toward customized extracellular niche engineering ...download.xuebalib.com/5ex0ngdcmhco.pdf · such...

REVIEW

1703948 (1 of 20) © 2017 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

www.advmat.de

Toward Customized Extracellular Niche Engineering: Progress in Cell-Entrapment Technologies

Dilip Thomas, Timothy O’Brien, and Abhay Pandit*

DOI: 10.1002/adma.201703948

self-assembly of functional tissue-like structures.[2,3] Although such structures do not entirely match multicellular organi-zation, tissue engineering and regenera-tive medicine approaches often strives to replace damaged tissues by “nesting” cells in a tissue-specific ECM to trigger a reparative response and restore tissue homeostasis.[4]

In nature, hierarchical complexity in tissues is achieved by spontaneous self-assembly of macromolecules, whose unique composition dictates the blue-print of scaffolding for the ultimate com-plex functional structure. Remarkable strides have been made by bioengineers in fabricating materials, which “cosmeti-cally” resemble topographies, porosities, mechanics and surface chemistries of tissue precursors.[5,6] Regardless of the structural and mechanical mimicry, the

designed materials remain limited with respect to cell mediated proteolytic remodeling, exchange of small molecules across the ECM components, integrin-mediated adhesions and associated ligand dependent biological functions in 3D (Figure 1). For example, without the formation of cytoplasmic associations with the surrounding ECM, cells undergo “anoikis” or programmed cell death often triggered by caspase signaling pathway. On the other hand, nonadherent transformed cell types perceived as “cell-factories” have benefitted from advances in material sci-ence, enabling transplantation of cells as an immune-isolated ectopic organ.[7] In this case, design parameters and the effect of entrapped cells have been considered mutually exclusive to each other—although they are functionally intertwined.

The spurring interest in recapitulating extracellular matrices in simplified forms has generated a paradigm shift to engi-neer multifunctional cell-instructive niches that can regenerate living tissues without transient administration of cell factories. Immobilization of cells within a 3D tissue-like matrix allows nonpolar interactions due to the steric hindrance from the sur-rounding matrix.[8] Such an environment also allows freedom of self-assembly of ECMs, which are influenced by the pre-dominant macromolecules in the vicinity. In order to devise an optimal functional microenvironment, it is essential to define how simple is complex enough to trigger a reparative response. Our understanding of this extracellular niche has led to the development of biomimicking strategies to recapitulate a pas-sive or interactive extracellular environment where cells can be immobilized. In this review, we discuss the concept of cell immobilization, cellular effects on empirical biomaterial design

The primary aim in tissue engineering is to repair, replace, and regenerate dysfunctional tissues to restore homeostasis. Cell delivery for repair and regeneration is gaining impetus with our understanding of constructing tissue-like environments. However, the perpetual challenge is to identify inno-vative materials or re-engineer natural materials to model cell-specific tissue-like 3D modules, which can seamlessly integrate and restore functions of the target organ. To devise an optimal functional microenvironment, it is essen-tial to define how simple is complex enough to trigger tissue regeneration or restore cellular function. Here, the purposeful transition of cell immobiliza-tion from a cytoprotection point of view to that of a cell-instructive approach is examined, with advances in the understanding of cell–material interactions in a 3D context, and with a view to further application of the knowledge for the development of newer and complex hierarchical tissue assemblies for better examination of cell behavior and offering customized cell-based thera-pies for tissue engineering.

Tissue Engineering

1. Introduction: The Extracellular Niche

Cells in tissues are enmeshed in a network of functional extra-cellular matrix (ECM), which evolves due to time-dependent transitions or changes in biochemical or physiological pro-cesses. A functional ECM is a critical regulator of cellular processes which serve diverse functions such as sequestering signaling molecules, transmitting ligand specific cues via cell receptors and is amenable to synthesis, degradation and reas-sembly over time. Conventional 2D culture of stem cells over several passages influence cell phenotype and function.[1] 3D culture of adherent cell types on a biologically relevant matrix not only promises phenotypic maintenance, but also

Dr. D. Thomas, Prof. A. PanditCentre for Research in Medical Devices (CÚRAM)National University of Ireland GalwayGalway, IrelandE-mail: [email protected]. D. Thomas, Prof. T. O’BrienRegenerative Medicine Institute (REMEDI)National University of Ireland GalwayGalway, IrelandDr. D. ThomasCardiovascular InstituteStanford UniversityPalo Alto, CA 94305, USA

The ORCID identification number(s) for the author(s) of this article can be found under https://doi.org/10.1002/adma.201703948.

Adv. Mater. 2017, 1703948

© 2017 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim1703948 (2 of 20)

www.advmat.dewww.advancedsciencenews.com

parameters, current technologies for cell immobilization and biomaterials for immobilization are discussed. Furthermore, we distinguish biomaterial-based cell-entrapment devices on its role in imparting immunoprotective or immunomodula-tory properties. Finally, we summarize the important features and functions governed by cell–material interactions, high-lighting the need for fabrication of customized cellular micro-environments in the context of cell-delivery platforms that are more bioinstructive and bioresponsive in restoring tissue function.

2. Cell Immobilization: Concept

Cells entrapped within a functional extracellular matrix pro-mote cell survival, proliferation and material guided biochem-ical functions. Based on the spatial distribution of cells and localization within the material, there are three main immo-bilization strategies: embedding, encapsulation and surface seeding as schematically represented in Figure 2.

In an embedding approach, cells are mixed with a forming hydrogel solution, which allows uniform distribution of cells prior to the gelation process, thus enabling an injectable mode of delivery. The second approach, also referred to as an immu-noisolation approach, allows encapsulation of the cells within the core of the hydrogel matrix. A semipermeable membrane isolates the cells from its environment, which forms a barrier to immune response mediators (e.g., antibodies, cytotoxic T cells, macrophages), but allows exchange of gases and nutrients. The third approach involves seeding of cells post-fabrication of the hydrogel matrix. Such systems are generally synthesized under harsh chemical conditions or cytotoxic reagents, which excludes cell seeding during the fabrication process. However, this technique is widely used for in vitro expansion of cells in a 3D environment. Irrespective of the immobilization technique used, the material must provide a conducive environment for growth, proliferation, migration and some degree of hierar-chical self-assembly.[9] Hence a rational design of the hydrogel with respect to bulk transport of nutrients, porosity, stiffness, chemical composition and promotion of dynamic remodeling in a cell permissive environment is crucial.[10] 3D immobiliza-tion platforms recapitulate mechanical and biochemical stimuli as present in native tissue. For example, 3D systems allow isotropic interactions, resulting in polarization dictated by the surrounding matrix and not dictated by unnatural polarization along the 2D plane of contact with the substrate or plastic in most cases.[11] Moreover, cell migration in 3D systems occur via both localized proteolysis with membrane-bound proteases and integrins, and amoeboid migration by deformation of extracel-lular matrix and shape adaptation.[12]

2.1. Terminology

Cell immobilization is a technique to entrap cells within an artificially processed biomaterial in order to utilize its biochem-ical and biophysical effects to promote spatial self-assembly or directed-assembly into multidimensional tissue-like constructs. In the field of cell-immobilization technology, due to the rapid

Dilip Thomas is currently a postdoctoral researcher at the Cardiovascular Institute at Stanford University. Dilip obtained his B.Sc. degree in biotechnology from the University of Mumbai, M.Sc. degree in biochemical engineering from University College London, and Ph.D. degree from the National University of Ireland, Galway.

His Ph.D. at the Center of Research in Medical Devices (CÚRAM) focused on the development of a functionalized stem cell delivery platform for the treatment of Critical Limb Ischemia. His current research interests include 3D in vitro cell models, induced pluripotent stem cells (iPSCs), high-throughput drug screening, and modeling of cardiac disease.

Timothy O’Brien is the dean of College of Medicine, the director of the Centre for Cell Manufacturing Ireland (CCMI) and the Regenerative Medicine Institute (REMEDI) at NUI Galway. He received his honors MB BCh BAO degree from UCC in 1984. He completed a two-year residency in internal medi-cine at the Medical College

of Wisconsin in Milwaukee in 1990 followed by a subspe-cialty fellowship in endocrinology and metabolism at the Mayo Clinic in Rochester, MN, in 1992. He is a consultant endocrinologist whose research is focused on the use of advanced-therapy medicinal products, such as cells and genes, for the treatment of diabetic complications.

Abhay Pandit is an established professor in biomaterials and the director of the Science Foundation Ireland funded Centre for Research in Medical Devices (CÚRAM) at NUI Galway. His research interests focus on the development of functionalized biomaterial-based delivery systems designed to provide sustained and targeted biomolecule

(drugs, cells, nucleic acids, growth factors) delivery to address unmet clinical needs. Such delivery systems limit adverse side-effect profiles, reduce therapeutic degradation and loss, and increase bioavailability at the site of interest. He has studied the effect of the complex milieu of signals, consisting of biochemical and biophysical cues, on biomate-rial scaffolds in a complex in vivo microenvironment.

Adv. Mater. 2017, 1703948



development in new platforms, there is a need for systematic nomenclature to define cell-based systems. As highlighted pre-viously, the form factor of the pre-formed structures used for cell immobilization remains elusive under the commonly used term “encapsulation.”

Based on the size of the encapsulation device, cell-delivery platforms are categorized under “micro” or “macro.” Macro-encapsulation refers to entrapment of cells in larger structures with smaller surface-to-volume ratio, such as flat membranes, hollow fibers and cylindrical diffusion chambers. Macroencap-sulation-based devices developed as intravascular or extravas-cular implants have been used to treat metabolic and neurode-generative disorders. A recent, comprehensive review on mac-roencapsulation devices has been published elsewhere.[13]

The term “microencapsulation” generally encompasses structures that have a spherical geometry on the microscale. These structures are generally made of hydrogels, varying from tens of micrometers to sub-millimeter sizes (0.2–0.5 mm) up to ≈1.5 mm. Pre-formed cell laden constructs that are used for various therapeutic applications are summarized in Table S1 in the Supporting Information. The evolution of terminologies over the past five decades referring to the cell laden constructs namely, microcapsules,[14] microbeads,[15] microspheres,[16] microreactors,[17] microstructures,[18] micro-particles[19] and microgels[20] are represented on a time-scale in Figure 2. In most cases, the interchangeably used terms such as “microbeads” or “microparticles” do not provide any

information on the composition or shape of the construct. A spherical-bead-shaped structure with cells suspended in a flu-idic bed or within its void space, encased by a semi-permeable membrane is a shell-based capsule, also termed as microsphere-shell (mSS). Alternatively, the cells can be enmeshed within a crosslinked polymeric solid core, followed by a formation of the outer shell. These are termed as matrix core microsphere-shell (MC-mSS). Interpenetrating polymeric matrices that are used for homogenous cell embedding, such as agarose, collagen etc. do not offer total immunoisolation, and are termed as micro-sphere-gel (mSG). In some cases, an additional coating may be applied to mSG forming an external barrier and these are referred as coated/layered microsphere-gel. Pre-fabricated cell carriers used for surface seeding for scalable cell expansion are termed as microsphere-carriers.[21]

3. Recapitulating the ECM: Focus on Hydrogels

Cells readily shape and remodel their extracellular environ-ment by enzymatically degrading and resynthesizing the ECM. In tissues, ECM not only acts as an immobilization platform for higher order of self-assembly, but also facilitates the relay of biochemical and mechanical cues in a buffered and hydrated environment. The classical tissue engineering approach aims to engineer artificial 3D environments to direct stem cells to achieve a desired function, which can then be implanted into

Adv. Mater. 2017, 1703948

Figure 1. Cell interactions in a biological system. Cell–extracellular-matrix interactions that occur in a biological environment can be viewed as a func-tion of changing extracellular environments mediated by cell–cell and cell–matrix engagement. Cell–extracellular-matrix interactions are initiated by integrins, focal adhesions, and signaling receptors that sense soluble or tethered cues. Changes in physical properties of the matrix induced by tissue enzymes or fluid multiscale shear stresses affect cytoskeletal organization through gated ion channels or transmembrane receptors. A delicate temporal balance between cell–cell and cell–matrix relationships governs functional harmony and tissue homeostasis.



the host.[22] Although ECMs extracted from tissues can be reconstituted with resident cell population with easily, batch-to-batch variability and ambiguous qualification standards with regard to the composition and contaminants in ECM derived materials pose as pertinent issues. A bottom-up assembly of ECM components therefore has gained higher impetus with a view to translate the ECM language, letter-by-letter with the use of ECM mimetics, gaining better understanding of the tunability that can be achieved with precise patterning of bio-chemical cues, which would predictably model cell driven ECM organization. Thus, a plethora of natural and artificial ECM mimetics have been developed for immobilizing cells. From a tissue engineering perspective, delivery of stem cells in a cell permissive matrix provide a local stem cell niche that can give rise tissue-specific cells and participate in the regulation of tissue repair by the release of trophic factors. Based on the end-application, cell-immobilization technology has been exploited for two main strategies: one where xenogenic or genetically modified cells are entrapped to serve as “cell-factories” for ther-apeutic factor release. Second, where allogeneic or autologous cells are entrapped within a material for manipulation of cell behavior to promote a specific phenotype. Cell-immobilization technology was predominantly developed as an immunoisola-tion technique, to protect the beneficial cells from the body’s natural defense mechanisms. It was first demonstrated in 1933 that encapsulation of tumor cells within a nitrocellulose membrane maintained cell viability and the transplanted cells evaded the immune system in a guinea pig model.[23] Over four decades later in 1980, Lim and Sun adopted this technology, using endocrine porcine islets that were encapsulated within

alginate-poly-L-lysine-alginate (APA). Encapsulation platforms designed for a wide range of cell types have been developed for therapeutic applications in recent years.

Hydrophilic polymers or “hydrogels” have been a favored choice of materials for cell immobilization in tissue engineering applications due to its ability to mimic the architecture and mechanics of pliable cellular microenvironment.[24,25] Tissue-like fluidity, facile transport of soluble nutrients, ease of fabrica-tion and integration with biological interfaces are some of the main advantages of a hydrogel system. In recent years, different classes of hydrogels have been fabricated from natural and syn-thetic polymers or amalgamation of both, referred to as semi-synthetic polymers. Synthetic polymeric materials allow a higher degree of control over physical and mechanical properties over natural polymers. Natural polymers rely on spontaneous self-assembly, offering a seemingly familiar molecular recognition to the cells.[26] Semi-synthetic materials on the other hand, undergo synergistic arbitrary assembly due to steric interference of the synthetic component.[27] Natural polymers derived from ECM proteins are crucial for orientation and structural integrity of cellular microenvironment, and provide bioinstructive cues through ligand interactions between cell–cell and cell–matrix.[28]

3.1. Cellular Niche Design Considerations

3.1.1. Mass Transfer and Bulk Shape

Stability of a hydrogel is a function of the degree of crosslinking and its geometry, which influence transport of nutrients,

Adv. Mater. 2017, 1703948

Figure 2. 3D microscale cell-immobilization strategies. A) Embedding technology involves cells entrapped in a “bulk gel” forming without a cell-impeding boundary. Surface seeding technology allows cell expansion on the limited surface of a material with high nutrient availability. Encapsulation is an isolation approach adapted to protect cells from harsh physical or biochemical conditions. B) Evolution of terminologies used for cell-carrier systems over the last five decades.



metabolites and cytokines. The shape and size of the hydrogel determines the rate of molecular diffusion and the amount of fluid retained within the matrix.[29] In a 3D system, the shorter the average distances from the core to the outer boundary, the higher the diffusion of nutrients and analytes across the hydrogel. A porous 3D matrix can facilitate multidirectional dif-fusion compared to a unidirectional access to soluble mole cules on tissue culture plastic. Mass transfer limitations of nutrients can affect cellular osmolarity, growth and metabolite produc-tion. Hydrogels greater than ≈200 µm in thickness limit dif-fusion of nutrients and waste from the core,[30] that leads to a hypoxic gradient from the surface to the core. In such cases, it is necessary to simulate the permeability and diffusion in cell carriers where there is no mixing in static cell culture condi-tions, as this eventually leads to a necrotic core from poor nutrient diffusion. Most nutrients such as glucose, amino acids and oxygen easily diffuse as their molecular weights are below 200 Da. For adequate immunoisolation, it is essential that cell-carrier systems exclude high-molecular-weight compounds such as immunoglobulins and harmful immune cells with an ideal molecular-weight cutoff in the range of 50 000–100 000 Da.[31] Although there is a pertinent risk of diffusion of smaller com-plement proteins or antigens shed from the encapsulated cells leading to T-cell activation. For tissue regeneration strategies, higher porosities and pore sizes are beneficial to allow cell invasion, proliferation and vascularization in vivo. Larger pore size allows higher diffusion of oxygen and nutrients across the hydrogel. Porosities are manipulated either by crosslinking the hydrogel or increasing the concentration of the pre-polymer. Studies have highlighted the importance of varied porosities in hydrogel platforms. Intriguingly, porosity demands are cell type specific ranging from 5 to 350 µm.[32] One study demonstrated an optimal 10% wt porous Chitosan matrix with its stiffness matching the brain tissue led to higher differentiation of neural stem cells into neurons, oligodendrocytes and astrocytes under differentiation conditions.[33]

Diffusion within a spherical construct is a function of its radius and high surface area to volume ratio. Studies con-ducted on the rate of oxygen diffusion in encapsulated islet cells revealed higher mass transport in microencapsulated spheres over planar diffusion assemblies.[34] Studies have also indicated that maximal lengths of diffusion to avoid hypoxic core formation are 128, 182, and 224 µm for a slab, cylinder, and sphere, respectively.[35] Only a limited number of studies have been performed on simulating substrate mass transfer in microcapsules.

Limitations in efficient diffusion can also be observed by the distribution of cells. Cells tend to migrate and grow on the periphery or boundary of the system in order to meet their oxygen and nutrient demands. From the shape and mor-phology point-of-view, it has been demonstrated that size and surface characteristics of a cell-delivery platform can also influ-ence inflammatory responses.[36] Smaller microcapsules result in reduced immune reaction compared to macrocapsules, which may undergo distortion after implantation. Microcap-sules exhibit low interfacial tension with surrounding tissues, which minimizes protein adsorption and cell adhesion. Pro-tein adsorption and cell adhesion is prevented by nonspecific repulsion, resulting from entropic and mixing interactions

between polymeric chains and the surrounding tissue.[37] Addi-tionally softer, pliable features of spherical hydrogel reduce mechanical resistance or irritations due to friction at the site of implantation.[38]

3.1.2. Stiffness

Matrix stiffness and its mechanics are known to influence cell migration, spreading and cell fate based on the specific ligand-receptor interactions. Several mechanotransducers within a cell convert mechanical stimuli into biochemical signals that affect cell behavior. Tissue-specific adult stem cells such as notochordal cells[39] of the intervertebral disc or skeletal muscle satellite cells,[40] are challenging to expand on stiff tissue cul-ture plastic due to their sensitivity to stiff 2D environments. Mechanosensors such as G-protein receptors, integrins, and stretch ion channels convert mechanical cues into intracellular signaling cascades. ECM derived hydrogels possess viscoelastic property, which allow limited reversible conformational change upon application of deformational stress.[41] Cells are pulling and tugging at the ECM constantly, the degree of stiffness and viscoelasticity leads to considerable pull and recoil via adhesion ligands, resulting in the rearrangement of cytoskeletal compo-nents, just like pulling a spring soldered to a stationary surface or a spring embedded in clay, comparable to cells in a stiffer or softer nonelastic matrix.[42] Readily deformable low density of ECM clusters induces adipogenesis in mesenchymal stem cells (MSCs); whereas less resilient high density ECM induce osteo-genesis through RhoA-dependent cytoskeletal contractility.[43] Resistance in matrix pliability can induce formation of stress fibers leading to the activation of Rho-GTPases. However, acti-vation of the same pathway on stiffer ECM substrates lead to the inhibition of neurogenesis in neural stem cells.[44] On the other hand, elasticity rendered by tropoelastin promoted the expansion of undifferentiated hematopoietic progenitors solely through extensional elasticity sensing through myosin II and not integrin signaling.[45]

Synthetic materials are generally not affected by the rate or magnitude of deformation; hence they are defined by a single bulk modulus. Higher polymeric concentration is proportional to higher stiffness, which is also influenced by the arrangement of the macromolecules that can form physical or chemical crosslinking “bridges.” In photocrosslinkable synthetic mate-rials, stiffness is modulated by generation of free radicals that initiate crosslinking.[46] For example, modified methacrylated or acrylated hyaluronic acid have been used to encapsulate human MSCs, where spatial and temporal changes in the mechanical properties have been achieved by restricting light exposure.[47] Increase in material stiffness is usually accompa-nied by increase in focal adhesions and polarization of stress fibers and the traction forces experienced by the cells are lin-early associated with substrate stiffness.[48] Recently, it was shown that MSCs seeded on 2D polyacrylamide gels of varying stiffness induce differentiation via nuclear lamin A, where overexpression of lamin A on stiff matrix favor osteogenesis and abrogation of lamin A pushed the cells down adipogenic pathway.[49] Nuclear shape and size vary between 2D and 3D environments, which could enhance or dampen the influence

Adv. Mater. 2017, 1703948



on the nuclear mechanics and binding of lamins to transcrip-tional regulators.[50] Endothelial cells form robust cell networks on softer compliant hydrogels due to sufficient ligand availa-bility, weaker substrate adhesions and nonimpedance in migra-tion dependent cell–cell network formation.[51] Cells on stiffer matrix spread more isotropically compared to softer matrix,[52] and when subjected to stiffness gradients, cells migrate from softer to stiffer regions.[53] This phenomenon, called durotaxis, has also been observed in activated microglia.[54] Cell–cell com-munication and assembly are also influenced by matrix stiff-ness, where cells form associations through the strain caused by traction stresses and weak substrate adhesions.[55]

However, it must be noted that characterization of 3D matrix stiffness involves measurement of “bulk” stiffness rather than individual fiber stiffness. Individual fibers within a hydrogel can be of multiple orders of magnitude greater in stiffness comparison to a macroscopic hydrogel.[56] Studies carried out to identify cell responses to stiffness, stem from the need to mimic extracellular 3D environment of cells in different tissue environments, such as brain (≈1 kPa), adipose (≈5 kPa) liver (≈0.3–6 kPa), muscle (≈10–30 kPa), and precalcified bone (≈50 kPa).[57–59] In vivo crosslinking of proteins occurs in time-dependent manner through regulated and nonregulated mech-anisms. For example, enzymatic crosslinking of collagen by lysyl oxidase is critical in developmental processes and wound healing. Nonenzymatic crosslinking mediators through glyca-tion and transglutamination occurs over longer periods of time, which also result in several disease pathologies as a result of uncontrolled ECM homeostasis.[60] Hence, investigation of mechanical effects of ECM on cell behavior must be carried out in a more meaningful way, not only to stimulate desired cell function, but also to restore mechanical homeostasis in tissues.

3.1.3. Macromolecular Concentration

Some of the cell-fate determination effects occur due to the change in macromolecular concentration, which promote molecular clustering of cellular receptors and influence intra-cellular signaling. For example, a higher concentration of fibrin (50 mg mL−1) induced osteoblastic differentiation of MSCs,[61] likely due to the high stiffness of the fibrin matrix.[62] Unlike chemical crosslinking approaches used with other ECM mole-cules, stiffness in fibrin hydrogels can be tuned via supplemen-tation of thrombin, a serine protease that converts fibrinogen to fibrin under physiological conditions. However, the con-centration of thrombin is shown to significantly influence cell fate. Human umbilical MSCs embedded in fibrin microbeads crosslinked with lower thrombin concentration promoted myo-genic differentiation with the formation of multi-nucleated myotube structures.[63] Whereas, the use of five times higher thrombin concentration in a fibrin scaffold increased neuronal differentiation of embryoid bodies derived from murine pluri-potent stem cells without conventional retinoic acid induced differentiation.[64] Cell phenotype was shown to be maintained in MSCs embedded in a collagen type-I hydrogel.[65,66] Further studies into the collagen type, concentration and cell density have revealed that they can influence gene expression and the paracrine signaling, in some cases directing the cells into a

pro-angiogenic[67] or enhance chondrogenic phenotype[68,69] by promoting higher matrix deposition.

4. Moving from 3D Cell-Entrapment Devices to Cell-Modulation Platforms

4.1. Complex ECM-Based Platforms

ECM is composed of a complex array of macromolecules that are contributed by various resident and mobilized cell types. The macromolecules dictate biophysical signaling via mechan-ical forces generated by catalytic ECM remodeling, and bio-chemical signaling due to soluble factors tethered and stored within the matrix or secreted by other cell types. Because of this intricate exchange of signals between the cytoskeleton and ECM via cell surface receptors and consequent self-assembly of macromolecules dictate cell-limited function such as fate determination. In tissues, combinations of signals are needed by stem cell niches to control phenotypic maintenance and self-renewal. The complexity of ECM environments is seen as a huge challenge in the field of tissue engineering due to the difficulty in delineating overlapping effects which arise due to the dynamic composition. Tissue engineering has taken a sim-plified approach in recreating cellular microenvironments, by creating a relatively bioinert platform for cells to secrete their own extracellular matrix. A more fundamental developmental approach is utilizing a wide gamut of naturally derived macro-molecules such as collagen, fibronectin, elastin, laminin and glycosaminoglycans (GAGs) among others to emulate a native stem cell niche that allows natural anchoring via cell adhe-sion molecules. The central idea behind this approach is not to reconstitute the ECM complexity present in the tissues, but entrap cells in a cell-instructive (provisional) “ground material” that can help cells organize and orient cytoskeleton unrestricted within the cell-specific matrix, also influencing orientation of ECM “tracks” secreted outside. For example, in hyaline cartilage tissue, chondroblasts reside in a capsular space or “lacunae” in which deposition of collagen type-II and proteoglycans trigger chondrocytic differentiation. In other cartilaginous tissues like the intervertebral disc, in addition to secretion of cartilage spe-cific matrix, chondrocytes also receive phenotypic guidance by arranging themselves between the grooves of aligned collagen type-I bundles.

GAGs interspersed in the extracellular space aid in hydration of tissues by sequestering water molecules. Sulfated GAGs exist as a “signaling router,” rich in growth factor binding domains, released upon enzymatic action within the extracellular matrix. In most tissues, HA acts as a space-filler that can resist compressive forces applied within tissues. For cell-delivery applications, HA is chemically crosslinked by amidation of the carboxylic groups or modification with methacryl groups in order to form photo-induced crosslinks for higher stability.[70] Methacrylated HA hydrogels influences the differentiation of neural progenitors into mature neuronal phenotype via Rho and Rac pathways.[71,72] Recently, neuronal maturation and increased survival of hESC derived dopaminergic neurons was demonstrated on a RGD-modified HA gel with an optimal stiff-ness of ≈350 Pa.[73] Entrapment of myoblast in a combination

Adv. Mater. 2017, 1703948



of HA and gelatin casted into a cylindrical hydrogel induced myotube differentiation. In this study, it was demonstrated that 58% of cells differentiated into myotubes with 30/70 gelatin/HA combination with the gel stiffness of 5.5 kPa.[74] In terms of in vivo response HA hydrogel both methacrylated and thi-olated have shown to reduce inflammatory responses in tis-sues such as the heart[75] and spinal cord.[76] Gelatin on its own binds different soluble cytokines and growth factors based on a shift in its isoelectric point,[77] and allows significant matrix deposition.[78] Recently, the use of crosslinked gelatin micro-cryogels for embedding adipose derived MSCs demonstrated that higher matrix deposition led to induction of a pro-angi-ogenic cell response in vivo.[79] On the other hand, umbilical cord stem cells embedded within a gelatin-laminin composite matrix enhanced cell survival and promoted oligodendrocyte differentiation in an ouabain induced rat neurodegenerative model.[80]

Bioactive peptides self-assembled within a hydrogel also exhibit cell maturation through ECM-specific interactions. Incorporation of laminin derived pentapeptide IKVAV in col-lagen matrix improved proliferation of dorsal root ganglion neurons.[81] A combination of collagen, fibronectin, laminin has differential abilities in stimulation of embryonic stem cell differentiation. High collagen concentration inhibited apoptosis, fibronectin induced endothelial differentiation and vascularization, and laminin promoted differentiation into car-diomyocytes.[82] Similarly a mix of collagen, HA and fibrinogen combined with synovium derived MSCs improved healing of an osteochondral defect in vivo by inducing formation of neo-carti-lage matrix, rich in GAGs and type II collagen.[83] A composite hydrogel composed of HA, heparin and collagen crosslinked with poly(ethylene glycol) (PEG) diacrylate enhanced survival of neural progenitor cells in an undifferentiated state in a cor-tical stroke mouse model. The study also reported a decrease in Iba1-positive cells, which is indicative of active microglia/macrophage infiltration.[84] Such tailored ECM compositions act or influence every cell type in a unique manner. Although the exact mechanisms by which these individual components act is not thoroughly understood, it highlights the need for a high-throughput ECM microarray platform[85] to unveil the cell–matrix interactions that influence stem cell regulation in a 3D context.

Similar to tissue derived ECMs, Matrigel, a complex hetero-geneous mixture of extracellular macromolecules and growth factors derived from a sarcoma cell line has been used as the underlying material for generation of cell organoids. For example, pluripotent stem cells have been differentiated into intestinal organoids[86] and hepatocytes[87] on Matrigel. As a material that has been popularly used for 3D cell invasion assays and tumor outgrowth studies, its use for cell immobili-zation and its relevance to healthy stem cell ECM may be ques-tionable. In addition to the challenge in defining the variable composition, the lack of an equivalent experimental control makes it harder to interpret the functional outcomes. Hence engineering stem cell-specific ECM modules block-by-block is a viable tissue engineering approach instead of devising one complex matrix that fits all. Ultimately, the de novo tissue mor-phogenesis in encoded spatiotemporally within the extracel-lular niche.

4.2. Non-ECM-Based Cell-Delivery Platforms

4.2.1. Polysaccharide-Based Platforms

Polysaccharides are synthetic or naturally derived support matrices and require additional functionalization to allow cel-lular processes such as adhesion, proliferation and differentia-tion. Biochemically polysaccharides such as alginate, agarose, chitosan serve as excellent inert matrices, which allow entrap-ment of cells that readily form 3D clusters and do not undergo “anoikis” due to the lack of attachment motifs. Alginate, algal derived polysaccharide has been widely studied as a cell-immobilization matrix, due to its ease in gel formation in presence of divalent cations. Alginate in its natural form is non-degradable. However, modifications such as partial oxidation and methacrylation does make the hydrogels prone to hydro-lysis in aqueous conditions.[88,89]

Alginate microcapsules fall under the category of both mSG and mSS. In many applications an additional layer of cationic polyelectrolyte (poly-L-lysine) is applied to improve capsule durability. However, this additional coating causes reduction in diffusive properties of the capsule and also leads to fibrotic growth, due to affinity of activated macrophages to positively charged surfaces.[90,91] In order to achieve better encapsulation and lower surface area, alginate capsules have been gener-ated using techniques such as micromolding,[92] droplet gen-erator,[93] coaxial jetting[94] and microfluidics[95] with various cell types. Porcine neonatal pancreatic clusters encapsulated in alginate microcapsules were shown to differentiate into insulin producing beta-cells, thereby maintaining normal blood glu-cose in diabetic mice over 20 weeks.[96] Despite its applicability and ease of use there are several concerns of its cytotoxicity.

Several clinical studies are currently ongoing using algi-nate for the implantation of beta cells in diabetic patients (NCT01379729, NCT00940173, and NCT00790257). Human fetal liver stromal cells and rat hepatocytes have been utilized for liver regeneration by administering within APA capsules sup-plemented with basal fibroblast growth factor, in a partial hepa-tectomy mouse model.[97] Although treatment enhanced the survival rate and increase in liver mass post 48 hours, necrotic cluster of cells within the capsules and inflammatory cell infil-tration was not completely prevented. Hence, several efforts have been made to incorporate natural ECM molecules during the process of encapsulation to improve cell survival, or induce differentiation in a physiologically similar environment.[98]

Embryonic stem cell aggregates encapsulated within alginate-poly-L-lysine (PLL) capsules maintained embryonic phenotype and prevented cardiomyogenic differentiation.[99] Although alginate-PLL capsules have shown encouraging results, the PLL coating on the microcapsules evoke immunogenic and fibrotic responses.[100] Therefore, alternative modifications such as chemically modified triazole-thiomorpholine dioxide alginate has been recently identified to resist fibrosis in vivo up to 24 weeks.[101] Alginates have also been used in combination with HA[102] and fibronectin[103] with MSCs to drive osteoblast differentiation and bone formation in vivo. 3D systems com-posed of chitosan-collagen beads fabricated using an emulsifi-cation process, showed higher deposition of calcium and bone related matrix under the influence of osteogenic medium.[104]

Adv. Mater. 2017, 1703948



Composite systems also allow compartmentalization cells in order to improve cellular retention over a long period of time. In one such example of multi-modal platform, adipose derived stem cells have been seeded on chitosan microcarriers that was embedded in a collagen hydrogel. Cells that migrated from the microcarriers into the gel maintained stem cell phenotype.[105] Such multi-modal platforms can be also used to recapitulate tissue-like environments with the entrapment of different cell types spatially within a bi-phasic matrix sequestering sequential presentation of biochemical cues in a temporal fashion.

Agarose is another sea algae derived linear polysaccharide that is also used for cell immobilization. Agarose-based cell immobilization of pancreatic islet cells has been tested as allo-graft and xenograft carriers for reversal of diabetes in mice. Islets encapsulated within agarose capsules implanted in the peritoneum of diabetic mice maintained normal blood sugar levels to ≈100 mg dL−1.[106] It was demonstrated that increase in the percentage of agarose to 7% prevented rejection of hamster islets in a diabetic mouse model.[107] An additional coating of polyacrylamide or carboxymethyl cellulose (CMC) was shown to be effective in excluding foreign antibodies.[108,109] For stem cell applications agarose has been modified with cell adhesive peptides for better cell survival. Neural stem cells were shown to differentiate into all three primary cell types: oligodendro-cytes, astrocytes and neurons in a GRGDS-agarose hydrogel decorated with platelet derived growth factor-AA.[110] Agarose encapsulation using a water-in-oil emulsion technique was employed to form scalable cellular modules for embryonic stem cell growth to maintain its differentiation capacity into hemato-poetic progenitor cells.[111]

Chitosan is another polysaccharide derived from exoskeleton of crustaceans. Chitosan is structurally similar to GAGs and has an overall positive charge. Hence chitosan has been used mainly as a carrier for chondrocytes. Similar to alginates, chitosan forms a hydrogel in presence of multivalent anions, such as tri-poly-phosphate. However, the use of chitosan as a cell-immo-bilization matrix has been limited due to its poor solubility. Chitosan-based hydrogel laden with VEGF165-releasing lipid microtubules, ESCs derived ECs and CD31+ bone marrow cells and was shown to improve cell survival and induce angiogen-esis in vivo.[112] Bovine articular chondrocytes embedded within a chitosan-glycerol phosphate gel, delivered in a subcutaneous mouse model led to formation of cartilage specific matrix.[113] Degradation products of chitosan is known to play a role in the biosynthesis of cartilage ECM.[114] Chitosan has also been used in a multi-modal platform, where adipose derived stem cells seeded on chitosan microcarriers were embedded within a collagen hydrogel. Cells that migrated from the microcarriers into the gel maintained stem cell phenotype.[105] A combina-tion of collagen and chitosan seeded with porcine MSCs deliv-ered in a minipig arthrotomy model showed significant depo-sition of cartilage ECM with an increase in femur length.[115] Hydrophilic carbohydrate polymers such as dextran have also been studied in the context of cell-delivery applications. Dex-tran offers good resistance to protein adsorption. However, without the addition of cell adhesion ligands dextran does not support cell attachment and proliferation. High abundance hydroxyl groups on dextran allow addition of functional pep-tides through derivatization and crosslinking reactions.[116,117]

A recent study demonstrated that grafting of RGD peptide at a concentration as low as 0.1% shows comparable cell survival to dextran containing 10% RGD.[118] MSCs encapsulated in RGD functionalized dextran has also been used for sustained delivery of siRNA against noggin, a key molecular inhibitor of osteogen-esis, to drive osteogenesis in MSCs.[117] Mainly, polysaccharide matrices have been favorably used as an inert, semipermeable capsule for cytoprotection, which is not intended to integrate with the host and direct tissue development.

4.2.2. Synthetic Polymers and Derivatives

Synthetic polymers used as ECM (sECM) mimetics have been an attractive source of materials for cell entrapment due to three main reasons; tissue-like viscoelasticity, low batch-to-batch vari-ability and facile mechanical tunability. sECM fabrication takes a modular approach in mimicking natural ECM by recapitu-lating the tensegrity of a structural protein combined with spa-tial distribution of biochemical signals. sECMs were initially developed as an implantation device to contain therapeutic cells for longer periods of time in a living host. This approach does not aim to recapitulate cell-specific matrix instead aims to provide a cell friendly environment for prolonged survival. The technique often involves maintenance of an immunoisola-tion barrier, which prevents adverse immune recognition while maintaining sustained release of therapeutic factors from the encapsulated cells.

For example, PEG diacrylate encapsulating porcine islets grafted into the peritoneum of diabetic mice, remained viable and maintained insulin levels up to four months.[119] The lack of biochemical cues and bioinert nature of PEG do not sup-port cell adhesion and high cell survival. Although, one of the main advantages of using PEG is that it can be modified with functional groups that allows incorporation of enzymatic deg-radation motifs, growth factors and/or cell adhesion islands (e.g., RGD, YIGSR, MMPs), serving as a “backbone” or “tem-plate” into which biomimetic sequences can be engineered. In a recent example, a mixture of polyacrylic acid, PEG and agarose hydrogel containing RGD sequences was fabricated to deliver hMSCs in a spinal cord injury mouse model. The bio-mimetic nature of the hydrogel was shown to modulate the pro-inflammatory environment by increasing M2 macrophage population.[120] Fabrication of such sECMs with natural ECM analogs expand the scope of synthetic matrices beyond pas-sive cytoprotection, to decipher the cell behavior depending on the type of ligand, concentration and spatial distribution. In a recent study, PEG hydrogel decorated with ECM peptides including adhesion peptide RGDSP was demonstrated to be a better mimic of 3D microenvironment to maintain higher pluri-potency and reprogramming efficiency of iPSCs.[121] Although it was interesting to note that the PEG hydrogel without the enrichment peptides led to comparable reprogramming of cells as that of 3D Matrigel or on a 3D collagen scaffold.[122] User-defined control in synthesis of such designer sECMs rely on “click” chemistries that form orthogonal mesh-like polymeric architecture, thus the functional cues incorporated are also orthogonally arranged, unlike uniquely sequestered cues within the ECM macromolecule in tissues.

Adv. Mater. 2017, 1703948



Several other polymeric encapsulation devices made of poly(urethane), poly(ether sulfone), polypropylene have been tested with cell types such as pancreatic islets,[123] myo-blasts,[124–126] hepatocytes,[127] B16 melanoma cells[128] and fibroblasts.[129] The major disadvantage of these materials is the absence of cellular recognition sites and high hydrophobic nature that is associated with protein adsorption, making it sus-ceptible to inflammatory reactions.

Poly(vinyl alcohol) (PVA) a thermoplastic polymeric material has been traditionally used as hollow fiber-based cell-encapsula-tion device for trophic release of growth factors or neurotrophic factors in neurodegenerative diseases.[130] PVA itself is a non-supportive matrix for the cells; also in vivo it leads to protein fouling due to its low hydrophilicity, resulting in an inflamma-tory response. However, application of an additional coating of poly(ethylene oxide) has shown to improve reduction in protein adsorption.[131] In a recent example, PVA hydrogel with a stiff-ness gradient was demonstrated to promote differentiation of bone-marrow MSCs into neurogenesis on the softer regions (1 kPa) and osteogenesis on the stiffer (24 kPa) regions of the hydrogel.[132] Tunabilty achieved through such temporal presen-tation of biological cues and local manipulation of mechanical properties have increasingly gained attention to design sophis-ticated sECMs to understand and modulate cell regulation.

4.2.3. Stimuli Responsive Platforms

Stimuli responsiveness can be exploited in form of random or directed assembly of polymeric networks incorporating func-tional peptide sequences that influence cell function or via the release of cells upon induction of proteolytic degradation. Extrinsic conditions, such as pH, oxidative stress, temperature and light are exploited in the development of stimuli respon-sive cell carriers (Table S2, Supporting Information). Stimuli responsiveness can also be incorporated into self-assembling hydrogels, where degradation is triggered in presence of an enzyme or a reactive catalyst. Typically tethering a growth factor to MMP-liable synthetic matrices allow release of the growth factor followed by cell mediated proteolysis, mimicking the dynamic release of factors in vivo upon infiltration and degra-dation of ECM. Incorporation of MMP degradable crosslinks in low crosslinked gels promote natural remodeling of the gels by cellular proteases and help in deposition of ECM molecules.

Chondrocytes and MSCs co-encapsulated in a PEG-based hydrogel modified with an MMP-degradable linker and TGF-β1 allowed degradation of the gel by MSCs, playing a supporting role in the deposition of cartilage-specific ECM by the chon-drocytes.[133] In another example, MMP sensitive peptide and cell adhesive peptide engineered on a PEG hydrogel, with an overall stiffness ≈4.3 kPa was shown to induce quiescence in encapsulated porcine aortic valve interstitial cells, which oth-erwise spontaneously induce myofibroblastic phenotype.[134] Collagenase and elastase specific sensitivity has been achieved by grafting peptide sequences on PEG, enhancing proteolytic remodeling and deposition of extracellular matrix by smooth muscle cells.[135] A peptide linker C-VPLSLYSG-C sensitive to MMP-2 and MMP-9 incorporated within a PEG hydrogel underwent rapid degradation upon exposure to exogenous

MMPs. However in vivo, fibrous capsule formation and foreign body response by macrophages is unavoidable.[136]

In vivo, one of the early mediators of inflammation are reac-tive oxygen species. In order to tackle superoxide-mediated cell damage, a polymerizable manganese metalloporphyrin and superoxide dismutase mimetic was engineered on a PEG hydrogel. The hydrogel protected the encapsulated cells from oxidative damage by successfully neutralizing reactive oxygen species.[137] A semi-synthetic hydrogel was made sensitive to changes in pH by tuning the concentration of PVA and hep-arin. Heparin, a naturally occurring sulfated polysaccharide has high affinity to bind growth factors. A basic shift in the pH induced rapid degradation of the hydrogel, and acidic pH prolonged the degradation in vitro for up to six months.[138] In another example, temperature sensitive property of gelatin has been exploited for degradation mediated cell release in a matrix composed of gelatin, agarose and fibrinogen. Destabilization of gelatin at 37 °C triggered partial decomposition of the matrix, achieving controlled release of therapeutic cells at physiological temperature.[139]

Synthetic and natural materials have been widely used for delivery of growth factors via diffusion or proteolytic degrada-tion. Tunability in mechanical and degradation properties in synthetic hydrogels is achieved by photopolymerization tech-nique with incorporation of light sensitive linkers. Valvular interstitial cells embedded on a stiff PEG-based hydrogel, was demonstrated to reverse from a myofibroblastic phenotype after UV-based degradation of the photoliable linker, causing softening of the hydrogel.[140] In a similar technique, light sen-sitive HA was used to stiffen the gel at different times. Early stiffening led to osteoblastic differentiation of MSCs, whereas late stiffening promoted adipogenesis.[141] A recent study dem-onstrated, “on-demand” presentation of adhesion ligand RGD on a synthetic, light-sensitive hydrogel modulates inflamma-tion and vascularization in vivo. Hydrogels with sequestered adhesion sites were implanted subcutaneously in mice and exposed to UV light at various intervals. Hydrogels exposed to UV for presentation of adhesion sites at day 7 and 14 resulted in 50% thinner fibrous capsules than controls. Similarly, gels containing both RGD and VEGF when triggered by UV at day 0 and 7, resulted in functional vasculature by day 14.[142] Sys-tems that respond to ultrasound stimulus have also been devel-oped. Osteoblasts seeded on methacrylamide-modified gelatin cryogels when exposed to ultrasound stimulus enhanced depo-sition of bone specific matrix.[143] A stimulus can also trigger assembly or dis-assembly of cell-carrier platforms. Simulta-neous encapsulation of chondrocytes and magnetic alignment of HA enriched collagen, formed aligned collagen fibrils in a HA enriched collagen hydrogel.[144]

Magnetoceptive hydrogels have also been fabricated by incorporation of magnetic radicals[145] or nanoparticles.[146] This unique approach has led to a controlled self-assembly of microtissues in a 3D “Tetris”-like conformation under mag-netic actuation. However a critical dose must be evaluated, as higher concentrations of these magnetic additives can compro-mise cell survival. In a similar approach, acoustic frequencies have also been used for layer-by-layer assembly of cell-laden constructs.[147] Cell-delivery platforms can also be made “switch-able” in different states with an external on-demand stimulus.

Adv. Mater. 2017, 1703948



A composite hydrogel composed of collagen and alginate was made “switchable” to collagen only hydrogel, after the disso-lution of alginate mediated by the chelation of Ca2+ ions. The temporal switching of hydrogels promoted cardiomyocyte dif-ferentiation of pluripotent stem cells under transgenic or non-transgenic induction conditions.[148] Such unique platforms offer a great promise in tuning cellular processes by presenting cues “on-demand” as it would occur in physiological state.

5. Technologies for Fabrication of Microscale Cell-Delivery Platforms

Microscale technologies allow spatial self-assembly of cells and materials within a confined environment. One of the major challenges in engineering such environments is the diverse size scale of materials that can be combined without affecting the final construct size or design parameters. Such fabrication technologies are a step forward not only in developing cell-encapsulating structures, but also providing a bioinstructive template for early commitment toward cell-specific function. Figure 3 summarizes commonly used fabrication technologies for generating cell-laden constructs.

5.1. Microsphere Gel Technologies

Microdispensing technique is a top-down approach for cell entrapment. Hydrogel-forming precursor with cells is extruded

from a needle or a tip via positive air displacement. Upon reaching a critical mass, the drop is displaced into a polymer-izing solution or a hydrophobic surface[149] for the retention of a spherical shape. The cell containing droplets are generally of a fixed volume. Manual methods of dispensing usually lead to droplet sizes over 1 mm. However, for smaller working volumes and faster gelation chemistries, this technique allows for good reproducibility and distribution of cells due to lower standing times. Using this method, several ECM-derived molecules and other high viscosity materials have been employed to fabricate more physiologically relevant microenvironments.[69,103] The cells within these constructs are embedded uniformly and hence referred as microsphere gels. Materials that polymerize at certain equilibrium temperatures use an emulsion-based technique, where the hydrogel-cell mix is dispensed into an immiscible phase such as oil or liquid paraffin. Collagen-agarose beads containing MSCs have been generated with this technique in an immiscible poly(dimethyl siloxane) medium, maintained at 37 °C.[150] Due to less control of droplet-droplet interactions within the emulsion medium, the resulting bead sizes are highly polydisperse, ranging from several hundred micrometers to a few millimeters. Newer techniques such as micromolding or soft lithography can generate cell-laden con-structs with the use of a micromold of a fixed geometry. The hydrogel precursor with cells are pressed onto a mold, filling the shape of the mold, followed by a crosslinking step. Another common approach using this technique involves, crosslinking precursor gels with UV exposure through a photomask-induced gelation. This technique has been used with both synthetic and

Adv. Mater. 2017, 1703948

Figure 3. Microscale cell-immobilization technologies. Schematic illustration of the formation of shape-controlled cellular environments. Micro-dispensing or replica molding cell and prepolymer matrix that is physically or chemically crosslinked results in a nonenveloped cell-embedded con-struct. Matrices that are formed without cells serve as a template for cell expansion. High-throughput technologies such as microfluidics, coaxial droplet generators, and bioprinting achieve high spatial control over distribution of cells, which enables rapid generation of cell-encapsulated constructs.



semi-synthetic polymers in entrapping several cell types such as fibroblasts, dorsal root ganglia cells and endothelial cells.[151]

The micromolding technique allows fabrication of con-structs of different shapes and sizes. But to better allow suf-ficient nutrient and metabolite transfer, the distance from the periphery to the core is typically recommended to be between 200 and 300 µm. Efforts have been made to fabricate concave replica wells with a diameter of 300 µm to ensure nutrient diffusion, thereby preventing a necrotic core. Pancreatic islet spheroids were entrapped within a collagen–alginate com-posite in a concave microwell array, crosslinked via slow dif-fusion of calcium chloride, resulting in an encapsulated “cell sheet.” In vivo transplantation of the encapsulated spheroids in a diabetic mouse model led to the maintenance of glucose level below 200 mg dL−1 for four weeks.[152] A similar emerging technique using wettable super hydrophobic micropatterns has been recently developed to fabricate microdroplets of defined geometry and volume.[153] The sandwich method has not only shown to form freestanding microgels containing cells, but has also been used to construct responsive, shape-coded gels with distinct cell types.[154] A high-throughput microfluidic approach has also been proven successful in generating cell-laden micro-sphere gels. Microfluidics is a technique that allows bottom-up fabrication of cell-entrapment modules.[155] Depending on the final construct shape and size, cell-laden constructs can be produced using microfluidic flow-focusing or microfluidic spinning. The former generates spherical structures, whereas the later generates cell-containing microfibers. In the fluid-focusing technique, a cell suspension injected in a polymeric matrix is sheathed with an immiscible carrier as it flows through microchannels of different geometries. The stream of liquid passing through an intersection breaks into droplets containing cells, under controlled flow rates and pressure gra-dient. There are several advantages of using this technique for controlled generation of cell-entrapped constructs. Controlled construct size determine consistent cell numbers per con-struct, and enable rapid exchange of nutrients and metabolic products.

5.2. Microsphere Shell Technologies

Liquid jetting is one of the popular methods for generating microsphere shells. Extrusion of liquid through a nozzle results in formation of droplets, depending on the velocity of the jet and physical interactions near the orifice, such as surface ten-sion, friction and gravitational pull. An applied field at the orifice pulls the droplets from the nozzle into a hardening solu-tion. Jetting technique can produce capsules of 50 µm in diam-eter under reproducible conditions. A small pulsing electrical field[156] can be applied to overcome critical surface tension of the droplet that does not adversely affect cell viability.[157] The cell-laden constructs formed with this technique are called as microsphere shells, often used as a noninvasive injectable cell-delivery device. For example, cells re-suspended in alginate are dripped into a bath containing divalent ions for crosslinking. The resulting spheres can be then subsequently coated with layers of polymers with opposite charges in order to encapsu-late cells embedded in the core alginate matrix.

Controlling the size and shape is critical to allow shorter dif-fusion path lengths and pericapsular reactions, resulting from protein adsorption and fouling on irregular surfaces. To achieve higher rate of droplet formation, modifications in the nozzle design has led to significant advances in jetting technologies. A vibrating nozzle at the extrusion orifice breaks the laminar fluid containing cells rapidly,[93,158] or by creation of a positive drag force using a gas stream, breaking the liquid jets into smaller droplets.[159] Vibration nozzle jet break-up system has been employed to encapsulate several cell types, such as porcine islets, chicken hamster ovary cells in cellulose sulfate capsules (CellMAC),[160] for constitutive release of small molecules such as insulin and erythropoietin respectively.

One of the major advancement in cell-based therapy using this technology has been made in cancer treatment. Immortal-ized human kidney cells producing CYP2B1 were encapsulated to achieve enzymatic bioconversion of a pro-drug into lethal antitumor agent. The study demonstrated safety and efficacy in vivo in a porcine model,[161] followed by a successful outcome in human phase I/II trial on 51 patients with pancreatic car-cinoma.[162] Further modification of the encapsulation device has led to the introduction of a downstream rotating device with wires or disc,[163] that ensures cutting trailing liquid jets into monodispersed homogenous beads. Contrary to the suc-cess with encapsulating immortalized cell types, the absence of cell adhesive ligands and limited macromolecular diffusive properties (<50 kDa), these capsules impede growth of primary MSCs.[164] Although jet-droplet technologies are promising for consistent and rapid generation of capsules, high viscosity and shear forces can potentially damage cells and reduce overall encapsulation efficiency.

Gelatin microsphere shells were fabricated in a two-step process using a microfluidic approach, where cell-entrapped microspheres were encapsulated in an additional layer of HRP-catalyzed phenol modified gelatin. Incubation at 37 °C dissolved the inner gelatin matrix, resulting in a liquid core capsule.[165] Similar strategies have also been used with algi-nate as a cell-entrapment matrix, where embedded calcium car-bonate nanoparticles act as internal gelation initiators that are slowly released due to acidic oil carrier phase.[166]

A recent study demonstrated vinyl sulfone functionalized PEG and alginate can be used to “shrink-wrap” pancreatic islets with a “conformal coating” using microfluidics. Reducing the thickness of the enveloping membrane to few tens of microm-eters, aided in better graft survival for up to three months in a rodent model.[167] The use of synthetic polymers for cell entrap-ment with this technique often requires a gelation step via UV photo-irradiation or redox polymerization. However, due to free radical formation or longer polymerization times cellular via-bility is compromised.

5.3. Microsphere Carrier Technologies

Pre-formed spherical microstructures with high surface-to-area volume ratio and cell adhesive properties, used for cell seeding purposes are termed as microsphere carriers. Microsphere carriers are mainly fabricated using four main techniques, water-in-oil emulsion crosslinking, antiphase suspension,

Adv. Mater. 2017, 1703948



spray-drying and liquid extrusion. Water-in-oil emulsion tech-nique involves mechanical agitation of a mixture of polymer solution in a suitable organic solvent with a surfactant such as Span80. These carriers are stabilized by ionotropic gelation using sodium hydroxide or sodium chloride.

Chemical crosslinker such as gluteraldehyde has also been used, although its neutralization by reducing agents such as sodium borohydride is required to alleviate cellular toxicity. Microspheres produced by this technique have a mean size of 132–150 µm.[168] Large mammalian cells do not penetrate these constructs due to small pore size, instead colonize on the sur-face and eventually migrate after over proliferation. Gelatin microspheres seeded with MSCs implanted in a subcutaneous rodent model, regenerated skin tissue and restored sweat gland like structures.[169] Using a similar technique, amniotic stem cells loaded on poly(lactic-co-glycolic acid) spheres <300 µm, when delivered intramyocardially, showed significant improve-ment in heart systolic function after myocardial infarction in a rat model.[170]

Antiphase extraction method also involves the emulsion pro-cess, where microspheres formed are frozen in liquid nitrogen, and then crosslinked in a mixture of pre-cooled ethanol and sodium hydroxide.[171] This technique has been used to produce microsphere carriers at a large scale. The average internal pore size of the microspheres generated is between 10–20 µm.[172,173] Microcarriers with pore size higher than 50 µm support growth and enhance matrix deposition by chondrocytes, than those below 50 µm.[174] Spray-drying technique allows good control over the microcarriers and porosity. In this technique the poly-meric solution is electro-sprayed into a collection flask with liquid nitrogen. The frozen microcarriers are then sieved and lyophilized.[175] Porous chitosan microcarriers produced by this technique were noncytotoxic and successfully used for hepato-cyte culture.[176]

Extrusion is a conventional technique in generating micro-spheres, where the polymeric solution is extruded drop-wise into sodium hydroxide, sodium sulfate or sodium citrate for crosslinking.[177] The resulting porous microspheres allow adhesion and proliferation of the cells. Porosity of the micro-spheres can be tuned and loaded with growth factors during the fabrication process to allow slow release and enhance tissue ingrowth.[178] Microcarriers have also been fabricated from decel-lularized dermal and adipose tissue matrix. Human-fibroblast-seeded dermal source microcarriers stimulated a thick layer of tissue growth under subcutaneous muscle at three weeks.[179]

5.4. Multidimensional Tissue-Like Assemblies

Multiscale cell assembly approaches from microscale to macro-scale promise to generate complex tissues block-by-block. “Bio-printing” is a state-of-the-art prototyping technology that has the potential to recreate a cellular microenvironment with hierar-chical cell–matrix assemblies. For example, decellularized nat-ural extracellular matrices of a specific tissue can be processed as “bioinks” and generate cell-laden 3D micro- and macro-tissues. This technique can also be applied to ECMs derived from xenogenic sources. However, the only major concern with using xenogenic tissue is to ensure ≈99% reduction in foreign

DNA during the decellularization process. Once the cells and the ECM matrices are prepared, they are filled into syringe holders of a robotic multi-head tissue/organ building system. The cell–matrix suspensions are then dispensed in an ordered pattern or casted over a synthetic support framework, such as polycaprolactone. A recent study has successfully demonstrated that bioprinted scaffold from decellularized heart, cartilage and fat can increase commitment of stem cells toward mature myo-blasts, chondrocytes and adipocytes respectively.[180]

A new microfluidic spinning technology allows generation of microfibers that can be easily manipulated and handled in comparison to microbeads. The diameter or thickness of the fibers can be tuned according to the flow rate. The thickness of the fiber influences the pore sizes within the fibers, which mediate efficient nutrient and metabolite diffusion. Co-axial extrusion of fibers using this technology has been successfully used to entrap liver cells[181] endothelial cells and neuronal cells with functional signal transduction.[182] Pancreatic islets loaded microfibers made of collagen-alginate,[183] and a core-shell fibers generated using double coaxial extrusion of algi-nate-agarose hydrogel[182] demonstrated maintenance of blood glucose in vivo, with adequate immunoprotection of the trans-planted cells from the host immune system. Technologies that help create microenvironments by spatial printing of matrix and cells, eventually undergo spontaneous remodeling and maturation as result of cell–cell and cell–matrix interactions. To mediate self-organization in a “bottom-up” approach, unit cell structures can be brought together in defined geometries, which over time fuse into seamless tissue organoids. Following this principle, heterogeneous 3D tissues were fabricated using a modular approach. HepG2 and fibroblast cells co-cultured within collagen beads, matured within a 3D mold, giving rise to a complex tissue-like architecture with three-fold higher albumin secretion.[184] Similarly, individual collagen spheroids containing smooth muscle cells and endothelial cells under-went uniluminal fusion mimicking multilayer blood vessel organization. The inner endothelium expressed PECAM-1 surrounded by α-SMA expressing smooth muscle layer.[185] A comprehensive review on the principles of multiscale tissue assemblies for bottom-up tissue engineering has been pub-lished elsewhere.[186]

6. Immunology of Cell-Entrapment Devices

Immunoprotection and immunomodulation are two impor-tant aspects that are critical in designing materials for cell delivery in an immunocompetent living host. Tuning the physical properties of the immunoprotective barrier or mem-brane can eliminate infiltration of immune cells or high-molecular-weight immunoglobulins. Material properties can be exploited to modulate host cell responses or suppress immune activators that recruit immune cell types via antigen presentation. Several strategies that are employed to achieve optimal immune response are schematically depicted in Figure 4. Altogether the idea is to tailor cell-delivery materials to evade immune responses via direct or indirect signaling, thereby promoting a reparative response with restoration of function.

Adv. Mater. 2017, 1703948



6.1. Immunoprotective versus Immunomodulatory Property

Immunoprotection is a vital component of cell-based therapies involving allogenic or xenogenic cell sources. The consequence of allogenic or xenogenic cell transplantation in an immuno-competent host results in graft rejection and toxicity. Lim and Sun in 1980 tested the immunoisolation technique, where rat islets encapsulated in a liquid alginate core capsules survived up to three weeks in vivo. The principle of immunoisolation is to create a semi-permeable barrier to restrict reactive host immune cells or mediators but allow nutrient diffusion for encapsu-lated cells. Hydrogels are engineered to integrate with the host tissue or degrade over time. Degradation is also dependent on crosslinking and permeability of the hydrogel. Hydrogels can be made impermeable to biological molecules with higher con-centration and molecular weight of gel-forming precursors. Metabolic requirements are varied for different cell types; hence molecular-weight cutoff (MWCO) of the isolating membrane must be tailored to ensure diffusion of essential nutrients. It has been demonstrated that MWCO up to ≈150 kDa (size of the smallest antibody IgG) is correlated with reduced fibrotic reac-tion and graft rejection.[187] However, this can also drastically reduce diffusion of high-molecular-weight nutrients.

In a systematic study, it was demonstrated that PEG-based hydrogels can be selectively made permeable to proteins. 2000–8000 Mw PEG was found to be impermeable to proteins equal or larger than 22 kDa. Hydrogels made of 20 000 Mw PEG were impermeable to proteins equal or larger than 45 kDa.[188] Multi layer coating of hydrogel have been demonstrated to be effective in immunoisolation of islets without compromising the insulin release function to glucose response.[189]

Immunoreactivity is often initiated as a pericapsular response following adsorption of proteins, immunoglobulins, complement and growth factors. Pericapsular growth promoted by IL-1β and TNFα[190] leads to a fibrotic layer formation, which impedes nutrient diffusion to encapsulated cells. For example, Alginates containing higher glucuronic units were more immu-noreactive than, alginates with higher mannuronic content.[191] Immunomodulation on the other hand is dependent on the chemical and physical properties of the encapsulating material, bulk size of the construct and antigen shedding from encap-sulated cells. Immunomodulation in cell-based therapies can be achieved by virtue of the cell-entrapment material, coen-trapment of immunomodulatory additives or use of immuno-privileged cell types. The use of immunomodulatory strategies with materials are summarized in Table S3 in the Supporting Information.

Surface topographies of the cell-carrier platform also play an important role in fibrotic layer formation. Rough surfaces, sharp features or corners attract more acute inflammatory reactions than smooth surfaces. A recent study performed in rodents and primates concluded that spherical constructs of 1.5 mm and above suppress foreign body reaction for longer periods.[192]

7. Fate Determining Cell Interactions

7.1. Identifying Niche Functions to Create Cell Specific Matrices

Hydrogel composition for immobilization and cell survival have been developed from a generic cell-permissive point-of-view

Adv. Mater. 2017, 1703948

Figure 4. Cytoprotective immunotactics. Immunoisolation of cells with materials offers selective permeability to essential nutrients and efflux of trophic factors secreted by encapsulated cells. Defined molecular-weight cutoff of semipermeable membranes exclude entry of host immune reactive molecules and cells. Immunomodulation is achieved by biophysical interactions or bioactive release of immunomodulatory agents. The figure describes the role of surface topography/roughness, antiprotein fouling due to surface hydrophilicity, charge affinity, and antireactive coatings that inactivate or polarize naïve macrophages and desensitize T-cell-mediated responses. Immunomodulation with bioactive agents include release of degradation products, anti-inflammatory cytokines or peptides or enzymes, tethered antibodies to micro-/nanoparticles against reactive cytokines. Active immunomodula-tion can also be achieved by codelivery of immunosuppressive or educated immune cells that interact physically or via secretory paracrine signaling.



than that of a cell-selective one. Selected examples discussed here highlight the role of ECM and cell–cell interactions that dictate specific function. Epidermal stem cells responsible for skin regeneration, deposit ECM protein nephronectin to home MSCs, that subsequently differentiate into smooth muscle cells.[193] In diseases, where production of structural ECM com-ponents of the basement membrane is impaired, gene vector modified cells delivered on a suitable matrix aid in synthesis of the ECM protein and form natural networks or association in vivo. For example in junctional epidermolysis bullosa, a rare genetic disease, causes production of defective collagen VII and laminin V. Delivery of vector-modified fibroblasts[194] from the patients delivered on a type IV collagen or fibronectin rich matrix allows remodeling of the skin via natural deposition and formation of ECM networks.

Muscle stem cells, also known as satellite cells, under acti-vated state encode for ECM proteins such as fibronectin, ver-sican, glypicans, which induce expansion of these cells.[195] Conversely, quiescent satellite cells express collagen VI that is required for self-renewal.[196] Further demonstrating the utility of ECM macromolecules in stem cell niches, fibronectin helps in branching morphogenesis via cell–cell and cell–matrix inter-actions. A hydrogel containing fibronectin used to mimic the perivascular niche of adult neural stem cell (NSC), enhanced the neuronal differentiation potential of NSCs.[197] Another ECM molecule, perlecan has been exploited in order to impart neuroprotection in a rodent stroke model, promoting angio-genesis by VEGF secretion.[198] Exploratory studies using microarray-based ECM platforms in a combinatorial manner have been useful in identifying the important role of two ECM components laminin 1 and jagged 1 in NSC growth.[199] Tis-sues that experience mechanical load or electro-physiological stimulation require a combination of ECM cues and optimal mechanical stimulation, such as for osteogenic or cardiomyo-cyte differentiation.

Newer hydrogel-based materials have evolved to incorporate biological cues that cells receive, via adhesion and soluble sig-nals in the microenvironment. Synthetic materials combined with ECM mimetic peptides with desired mechanical proper-ties and biological cues have been shown to direct cell fate.[200] In an alternative approach, cell specific material design may be possible by creation of an “inducible microniche,” which prime encapsulated cells to deposit cell specific ECM matrix will enhance cell engraftment and therapeutic benefit. Gelatin[79] and collagen microgels have been used to prime human MSCs to promote a pro-angiogenic phenotype leading to neovasculari-zation in vivo.[67]

High-throughput technologies comprised of hydrogel micro-arrays[199] with different combinations of ECM proteins and soluble factors will be useful in determining physiologically rel-evant cell-specific environment. Studies need to include uncon-ventional protein-based matrix molecules. One such molecule sericin, derived from silk worms, has shown to promote axonal growth of primary neuronal cells, offering neuroprotection by modulating apoptosis signaling pathway.[201]

Although several reductive ECM approaches have been investigated, it is important to note that many different cell types contribute to the ECM in a physiological niche. For example, bone marrow microenvironment is majorly composed

of collagen I, IV and VI, tenascin-C and fibronectin. Functional studies on collagen VI have revealed its importance in main-tenance and proliferation of hematopoietic stem cells. How-ever it is important to note that, this niche is resident to osteo-clasts, MSCs, megakaryocytes, macrophages, schwann cells and endothelial cells. Hence, while engineering cell specific environments using materials it is critical to consider that ECM proteins are not only integral in maintenance of structural equi-librium, but also harbor bioinstructive signaling cues to other cell types.

7.2. Role of Integrins

Integrins are “outside-to-inside” cell sensing anchoring recep-tors that bind and transmit signals from the extracellular matrix to intracellular cytosolic space. Integrins are homodimers com-posed of α and β chains. Overexpression of specific integrins via presentation of its binding ligand in its natural form or on a semi-synthetic platform can influence the downstream cell signaling. For example, collagen type-I allows embryonic stem cell renewal and maintenance via α2β1 integrin and discoidin domain receptor 1, by increasing the expression of nuclear Bmi-1 protein. Abolishment of Bmi-1 impairs the prolifera-tion capacity and maintenance of an undifferentiated state.[202] Integrins also interact reciprocally with Rho family GTPases by integrin-cytoskeleton linkages and integrin triggered outside-in signaling. For example α5β1 integrin interact with fibronectin binding domain resulting in increased RhoA activity,[203] a GTPase protein responsible for stress fiber formation. Increase in RhoA activity enhances osteoinductive signaling in MSCs.[43] Downstream effects of Rho signaling occur through multiple mechanisms.

Src family of tyrosine kinases is one of the crucial media-tors of integrin signaling. Src-family kinases, Fyn and Src are required for rigidity sensing, and leads to activation of Src kinases driving RhoA mediated contractility.[204] RhoA and ROCK mediated contractility regulate differentiation of epithe-lial cells in collagen gels.[205] When fibroblasts are cultured in 3D type I collagen, integrin α2β1 activate pro-survival signal through phosphoinositide 3-kinase and Akt/protein kinase B. However, when the collagen contracts, Akt is inactivated which activates a pro-apoptotic signal.[206] This is an example of how integrin-dependent ECM remodeling results in turnover of cells at different stages of regeneration.

Extracellular availability of fibronectin and laminin leads to expression of fibronectin integrin α5β1 and laminin integrin α1β1. It has been observed that if the ECM contains ligands for both integrins, α1β1 is known to have a dominant inhibitory effect on α5β1, which leads to reduction in cell proliferation.[207] Differentiation of chondrocytes from MSCs are influenced by cellular expression of integrin αv and α5, where α5 inte-grin expression is significantly correlated with chondrogenic marker Sox9 under mechanical stimulation of MSCs.[208,209] In another example, N-Cadherin, a transmembrane protein modu-lates β-catenin signaling via cell–cell interaction that promotes chondrogenic differentiation. A recent study has reported on the influence of N-cadherin peptide mimic immobilized within MSCs embedded HA hydrogels. The study showed presence of

Adv. Mater. 2017, 1703948



the peptide mimic enhanced β-catenin expression by 80% in mesenchymal stem cells.[210] In another study, presence of both fibronectin and laminin help differentiate MSCs into insulin producing cells by activation of Akt and ERK signaling path-ways.[211] A list of integrins and ligands that have been iden-tified to determine cell-fate is summarized in Table S4 in the Supporting Information.

Cell attachment motifs that are engineered in synthetic materials, promote cell attachment via integrin αvβ3. This inte-grin is well-known in promoting cell proliferation and invasion in tumor metastasis.[212] Hence upregulation of ECM ligands for αvβ3 binding, promote angiogenesis in endothelial cells or pericytes. Given the importance of integrin engagement in driving biological functions, customized matrices can be engi-neered that serve as a cell signaling router to achieve desired phenotypic characteristics.

8. Time–Space-Dependent Interactions: Fourth Dimension

Cells are spatially and hierarchically arranged within tissues, interlaced with biochemical, biophysical and electrical stimuli. Spontaneous genesis and abrogation of soluble and insoluble signals secreted by resident and circulating cells together con-tribute a functional equilibrium called “niche.” Scientific devel-opments to date in the area of cell and matrix biology, have contributed to the basic understanding of how cells sense and translate cues. However two major points, which many studies do not take into account, are: cells perform survival and meta-bolic functions differently in 2D and 3D environments. Also, cells are partially polarized on a flat substrate that guides ani-sotropy. Whereas in a 3D environment cells attain polariza-tion independently, from an isotropic state. Second, there is a dynamic reciprocity of cues between cells and the surrounding environment, established and abolished temporally. Cells resi-dent within a tissue have this potential which enable creation of a functional extracellular microenvironment.

A unique combination of ECM information embedded within the matrix serves as a conspicuous roadmap. The type of extracellular environment, its recognition and tolerance to host-tissue directed self-assembly will determine its regenerative capacity. Figure 5 illustrates the differences between the type of materials and the “biosimilar” characteristics that play an important role in modulating cell behavior and functional res-toration. Tissue-derived natural extracellular-matrix molecules offer a superior self-assembly by the virtue of a biological back-bone, and provide “fingerprint” recognition to cells expanded on an artificial environment. In a clear example, myoblasts seeded on a decellularized muscle matrix under serum free condition not only rapidly secreted endogenous proteins, but also formed mature myotubes along the “tracks” within natural muscle matrix.[213] Over time cells maximize cell–cell contact, degrade the matrix and migrate by applying traction forces on the matrix. Magnitude of traction forces applied by the cells is governed by ligand availability and stiffness of the characteristic 3D ECM architecture. One hydrogel composition does not fit all cell-specific extracellular environments. The differences in cel-lular environments alter adhesion protein dynamics three-fold

faster than 2D environments.[214] Dimensionality can diversify cell–matrix adhesion patterns, due to high tension and distribu-tion of adhesion sites. For example, in a 3D fibronectin matrix, cells co-localize α5 integrin with paxillin, whereas α5 integrin expression is absent in 2D.[215] These effects can enhance the continuity of 3D adhesions over time. MMP-dependent peri-cellular proteolysis modulates a “second-order” process that widens the pores within a matrix, and facilitates migration through degradable substrates.[216] Creating artificial environ-ments puts pseudo-control in the hands of the experimenter. Meaning, cells respond to the limited cues presented in syn-thetic environments, but in a physiological state, soluble and matrix bound signaling molecules shuttle between single or multiple cell types in a dose-dependent manner. This time-dependent interaction in a “fluidic” tissue environment is referred to as the fourth dimension. Recent advances in “click-chemistry” have allowed the development of synthetic matrices that can selectively present or remove bioactive molecules with full spatial and temporal control,[217] in a quasi-4D like environ-ment. The challenge however is whether the spatial distribution of cues can be made accessible locally to the desired cell type, when hierarchical complexity is achieved. On the other hand, factors such as secretion of extracellular matrix, feedback mech-anisms, and enzymatic remodeling can be modulated on a cus-tomized platform made of bioactive extracellular components.

9. Biological Function and Outlook

It is inarguably clear from this review, that ECM informatics is essential in designing matrices to drive number of funda-mental and complex cellular processes. However, the field of cell injection therapies views “living cells” as defined chemical entities, oblivious of the fact that, normal cell function can only be achieved in a tissue-like environment. In the last few years, several clinical trials with cell transplantation strategies have been unsuccessful at different phases of development, due to limited cell function. One of the major reasons of such fail-ures is due to oversimplification of a complex biological milieu, combined with regulatory hurdles in defining a controlled cell-environment for optimal therapeutic effect.

Cell-immobilization approach introduced the paradigm of cytoprotection, which augmented the cellular effect by pro-viding spatial confinement and protection from harmful host reactions. The need for tissue replacement and regenera-tive strategies coincided with the advances in supramolecular chemistry, giving rise to a new class of “living” biomedical devices. ECM engineering has led to the development of a diverse portfolio of materials that modestly biomimic the tissue environment. Although there has been a significant develop-ment in material sciences, a shift in focus, toward creating a controlled model environment has uncoupled the rationale for choosing a material from designing a cell specific tissue micro-environment. Well-defined synthetic ECM modules developed with tunable compositions, mechanics and sensitivity have immensely contributed to our understanding of cell behavior and function, by manipulating one-factor at a time.

The ECM found within tissues is not just scaffolding, but an instruction manual, which is characteristic of a specific tissue

Adv. Mater. 2017, 1703948



appliance. Number of different physical, mechanical and chem-ical activities is dynamically orchestrated to maintain cell func-tions. Conventional cell immunoisolation approach using xeno-genic or genetically modified cells only addresses short-term restoration of tissue function by implantation of exogenous cell-factories. However unconditional immunoisolation cannot be achieved against toxic effector molecules without compro-mising nutrient availability to encapsulated cells. Therefore, a combination of immunomodulatory elements embedded within tailored, tissue specific ECM matrix will allow to create an “inducible” niche, that will alter inflammatory responses and kick-start innate regenerative mechanisms.

High-throughput ECM arrays can be systematically assessed to understand functional and developmental mechanisms in a matrix induced tissue morphogenesis. In addition to cus-tomized matrices, it is also imperative to consider co-culture systems to allow nascent self-assembly prior to in vivo implan-tation studies. The key challenge is to define and characterize these systems for clinical use. Some of these factors include the composition, degradation kinetics, diffusion and release of therapeutic factors and time-scale for cell/tissue integration. With reliable recombinant protein expression technologies, a smorgasbord of ECM proteins can be produced in a reproduc-ible and cost-effective manner. The idea of synthesizing cus-tomized matrices for transplantation of therapeutic cells offers a promising avenue for biomaterial-based cell therapies focused toward engineering commercially viable regenerative solutions.

Supporting InformationSupporting Information is available from the Wiley Online Library or from the author.

AcknowledgementsThis publication has emanated from research supported in part by a research grant from Science Foundation Ireland (SFI) and was cofunded under the European Regional Development Fund under Grant Numbers [13/RC/2073] and [09/SRC/B1794]. The authors would also like to thank Maciek Doczyk for the illustrations in the manuscript.

Conflict of InterestThe authors declare no conflict of interest.

Keywords3D cell-immobilization, extracellular matrix, fabrication technologies, fourth-dimension, immunomodulation, immunoprotection, synthetic ECM

Received: July 15, 2017Revised: September 12, 2017

Published online:

Adv. Mater. 2017, 1703948

Figure 5. The expanding matrix. Current technologies allow us to create complex artificial 3D environments with biochemical and biophysical presets. These biological presets determine a smooth or interrupted response in a living system. Most materials, irrespective of their composition, interface with the host elements in a dynamic interplay. Resistance from a material against the rate of tissue re-modeling distorts the regenerative signaling. Pristine synthetic materials can be tuned by the user to mimic bulk elasticity of a target tissue that is composed of an array of molecules with micromodulus. Semisynthetic materials with borrowed biological mimetics can present spatial cues to modulate cell behavior. However, use of a tissue-specific matrix derived from natural extracellular-matrix components offers a superior self-assembly by the virtue of a biological backbone and provides a “fingerprint” recognition that can prime a cell-guided reparative stimulus for seamless integration with tissues or organs.



[1] G. J. Hickman, D. J. Boocock, A. G. Pockley, C. C. Perry, ACS Bio-mater. Sci. Eng. 2016, 2, 152.

[2] X. Yin, B. E. Mead, H. Safaee, R. Langer, J. M. Karp, O. Levy, Cell Stem Cell 2016, 18, 25.

[3] N. J. Walters, E. Gentleman, Acta Biomater. 2015, 11, 3.[4] S. Chen, M. Lewallen, T. Xie, Development 2013, 140, 255.[5] J. Thiele, Y. Ma, S. M. C. Bruekers, S. Ma, W. T. S. Huck, Adv.

Mater. 2013, 26, 125.[6] N. Annabi, A. Tamayol, J. A. Uquillas, M. Akbari, L. E. Bertassoni,

C. Cha, G. Camci-Unal, M. R. Dokmeci, N. A. Peppas, A. Khademhosseini, Adv. Mater. 2014, 26, 85.

[7] A. D. Celiz, J. G. W. Smith, R. Langer, D. G. Anderson, D. A. Winkler, D. A. Barrett, M. C. Davies, L. E. Young, C. Denning, M. R. Alexander, Nat. Mater. 2014, 13, 570.

[8] M. H. Zaman, L. M. Trapani, A. L. Sieminski, A. Siemeski, D. Mackellar, H. Gong, R. D. Kamm, A. Wells, D. A. Lauffenburger, P. Matsudaira, Proc. Natl. Acad. Sci. USA 2006, 103, 10889.

[9] D. Thomas, D. Gaspar, A. Sorushanova, G. Milcovich, K. Spanoudes, A. M. Mullen, T. O’Brien, A. Pandit, D. I. Zeugolis, Biotechnol. Bioeng. 2016, 113, 1155.

[10] A. Sala, P. Hänseler, A. Ranga, M. P. Lutolf, J. Vörös, M. Ehrbar, F. E. Weber, Integr. Biol. 2011, 3, 1102.

[11] R. S. Gieni, M. J. Hendzel, J. Cell. Biochem. 2008, 104, 1964.[12] K. Wolf, I. Mazo, H. Leung, K. Engelke, U. H. von Andrian,

E. I. Deryugina, A. Y. Strongin, E.-B. Bröcker, P. Friedl, J. Cell Biol. 2003, 160, 267.

[13] J. Schweicher, C. Nyitray, T. A. Desai, Front. Biosci., Landmark Ed. 2014, 19, 49.

[14] T. M. Chang, Science 1964, 146, 524.[15] H. Gin, B. Dupuy, C. Baquey, D. Ducassou, J. Aubertin,

J. Microencapsulation 1987, 4, 239.[16] R. P. Lanza, W. M. Kühtreiber, D. Ecker, J. E. Staruk, W. L. Chick,

Transplantation 1995, 59, 1377.[17] R. P. Lanza, J. L. Hayes, W. L. Chick, Nat. Biotechnol. 1996, 14, 1107.[18] W.-G. Koh, A. Revzin, M. V. Pishko, Langmuir 2002, 18, 2459.[19] J.-M. Rabanel, P. Hildgen, J. Microencapsulation 2004, 21, 413.[20] J. Yeh, Y. Ling, J. M. Karp, J. Gantz, A. Chandawarkar, G. Eng,

J. Blumling, R. Langer, A. Khademhosseini, Biomaterials 2006, 27, 5391.

[21] C. Nogués, D. Thomas, A. Pandit, T. O’Brien, in Biomaterials for Stem Cell Therapy (Eds: L. DeBartolo, A. Bader), CRC Press, Boca Raton, FL 2013, pp. 386–424.

[22] D. Seliktar, Science 2012, 336, 1124.[23] V. Bisceglie, Z. Krebsforsch. 1934, 40, 122.[24] P. M. Kharkar, K. L. Kiick, A. M. Kloxin, Chem. Soc. Rev. 2013, 42,

7335.[25] F. Yanagawa, H. Kaji, Y.-H. Jang, H. Bae, D. Yanan, J. Fukuda,

H. Qi, A. Khademhosseini, J. Biomed. Mater. Res. 2011, 97, 93.[26] L. Gasperini, J. F. Mano, R. L. Reis, J. R. Soc., Interface 2014, 11,

20140817.[27] A. Top, C. J. Roberts, K. L. Kiick, Biomacromolecules 2011, 12, 2184.[28] J. Lee, A. A. Abdeen, D. Zhang, K. A. Kilian, Biomaterials 2013, 34,

8140.[29] C.-C. Lin, A. T. Metters, Adv. Drug Delivery Rev. 2006, 58, 1379.[30] M. Nomi, A. Atala, P. D. Coppi, S. Soker, Mol. Aspects Med. 2002,

23, 463.[31] L. Christenson, K. E. Dionne, M. J. Lysaght, in Fundamentals of

Animal Cell Encapsulation and Immobilization (Ed: M. F. A. Goosen), CRC Press, Boca Raton, FL, USA 1993, pp. 7–41.

[32] N. Annabi, J. W. Nichol, X. Zhong, C. Ji, S. Koshy, A. Khademhosseini, F. Dehghani, Tissue Eng., Part B 2010, 16, 371.

[33] H. Li, A. Wijekoon, N. D. Leipzig, PLoS One 2012, 7, e48824.[34] R. M. Buitelaar, C. Bucke, J. Tramper, R. H. Wijffels, Immobi-

lized Cells: Basics and Applications, Elsevier, Amsterdam, The Netherlands 1996, pp. 239–486.

[35] E. S. Avgoustiniatos, C. K. Colton, Design Considerations in Immu-noisolation, Principles of Tissue Engineering, R. G. Landes, Austin, TX, USA 1997, pp. 333–346.

[36] P. de Vos, R. van Schilfgaarde, Cell Encapsulation Technology and Therapeutics, Birkhäuser Boston, Boston, MA 1999, pp. 63–75.

[37] T. F. Tadros, in Polymer Adsorption and Dispersion Stability, (Ed: E. D. Goddard), ACS Press, Washington DC, USA 1995, pp. 411–428.

[38] S. Nagaoka, R. Akashi, Biomaterials 1990, 11, 419.[39] X. Tang, L. Jing, W. J. Richardson, R. E. Isaacs, R. D. Fitch,

C. R. Brown, M. M. Erickson, L. A. Setton, J. Chen, J. Orthop. Res. 2016, 34, 1316.

[40] D. Montarras, J. Morgan, C. Collins, F. Relaix, S. Zaffran, A. Cumano, T. Partridge, M. Buckingham, Science 2005, 309, 2064.

[41] P. M. Gilbert, H. M. Blau, Stem Cell Res. Ther. 2011, 2, 3.[42] F. Guilak, D. M. Cohen, B. T. Estes, J. M. Gimble, W. Liedtke,

C. S. Chen, Cell Stem Cell 2009, 5, 17.[43] R. McBeath, D. M. Pirone, C. M. Nelson, K. Bhadriraju, C. S. Chen,

Dev. Cell 2004, 6, 483.[44] A. J. Keung, E. M. de Juan-Pardo, D. V. Schaffer, S. Kumar, Stem

Cells 2011, 29, 1886.[45] J. Holst, S. Watson, M. S. Lord, S. S. Eamegdool, D. V. Bax,

L. B. Nivison-Smith, A. Kondyurin, L. Ma, A. F. Oberhauser, A. S. Weiss, J. E. J. Rasko, Nat. Biotechnol. 2010, 28, 1123.

[46] R. Sunyer, A. J. Jin, R. Nossal, D. L. Sackett, PLoS One 2012, 7, e46107.

[47] A. D. Rape, M. Zibinsky, N. Murthy, S. Kumar, Nat. Commun. 2015, 6, 8129.

[48] M. Ghibaudo, A. Saez, L. Trichet, A. Xayaphoummine, J. Browaeys, P. Silberzan, A. Buguin, B. Ladoux, Soft Matter 2008, 4, 1836.

[49] J. Swift, I. L. Ivanovska, A. Buxboim, T. Harada, P. C. D. P. Dingal, J. Pinter, J. D. Pajerowski, K. R. Spinler, J.-W. Shin, M. Tewari, F. Rehfeldt, D. W. Speicher, D. E. Discher, Science 2013, 341, 1240104.

[50] P. Friedl, K. Wolf, J. Lammerding, Curr. Opin. Cell Biol. 2011, 23, 55.[51] R. L. Saunders, D. A. Hammer, Cell. Mol. Bioeng. 2010, 3, 60.[52] J. P. Califano, C. A. Reinhart-King, Cell. Mol. Bioeng. 2010, 3, 68.[53] C. M. Lo, H. B. Wang, M. Dembo, Y. L. Wang, Biophys. J. 2000, 79,

144.[54] L. Bollmann, D. E. Koser, R. Shahapure, H. O. B. Gautier,

G. A. Holzapfel, G. Scarcelli, M. C. Gather, E. Ulbricht, K. Franze, Front. Cell. Neurosci. 2015, 9, 363.

[55] C. A. Reinhart-King, M. Dembo, D. A. Hammer, Biophys. J. 2008, 95, 6044.

[56] M. Guthold, W. Liu, E. A. Sparks, L. M. Jawerth, L. Peng, M. Falvo, R. Superfine, R. R. Hantgan, S. T. Lord, Cell Biochem. Biophys. 2007, 49, 165.

[57] H. M. Blau, B. D. Cosgrove, A. T. V. Ho, Nat. Med. 2015, 21, 854.[58] D. E. Discher, D. J. Mooney, P. W. Zandstra, Science 2009, 324,

1673.[59] A. M. Cozzolino, V. Noce, C. Battistelli, A. Marchetti, G. Grassi,

C. Cicchini, M. Tripodi, L. Amicone, Stem Cells Int. 2016, 2016, 5481493.

[60] T. R. Cox, J. T. Erler, Dis. Model & Mech. 2011, 4, 165.[61] I. Catelas, N. Sese, B. M. Wu, J. C. Y. Dunn, S. Helgerson, B. Tawil,

Tissue Eng. 2006, 12, 2385.[62] H. Duong, B. Wu, B. Tawil, Tissue Eng., Part A 2009, 15, 1865.[63] J. Liu, H. H. K. Xu, H. Zhou, M. D. Weir, Q. Chen, C. A. Trotman,

Acta Biomater. 2013, 9, 4688.[64] A. Montgomery, A. Wong, N. Gabers, S. M. Willerth, Biomater. Sci.

2015, 3, 401.[65] C. M. R. Pérez, A. Panitch, J. Chmielewski, Macromol. Biosci. 2011,

11, 1426.[66] B. P. Chan, T. Y. Hui, C. W. Yeung, J. Li, I. Mo, G. C. F. Chan, Bio-

materials 2007, 28, 4652.

Adv. Mater. 2017, 1703948



[67] D. Thomas, G. Fontana, X. Chen, C. Sanz-Nogués, D. I. Zeugolis, P. Dockery, T. O’Brien, A. Pandit, Biomaterials 2014, 35, 8757.

[68] Y. Y. Li, H. W. Cheng, K. M. C. Cheung, D. Chan, B. P. Chan, Acta Biomater. 2014, 10, 1919.

[69] G. Fontana, D. Thomas, E. Collin, A. Pandit, Adv. Healthcare Mater. 2014, 3, 2012.

[70] Y. Du, M. Ghodousi, E. Lo, M. K. Vidula, O. Emiroglu, A. Khademhosseini, Biotechnol. Bioeng. 2010, 105, 655.

[71] W. Lu, J. Wang, T. Wen, Mol. Cell. Biochem. 2008, 311, 233.[72] S. K. Seidlits, Z. Z. Khaing, R. R. Petersen, J. D. Nickels,

J. E. Vanscoy, J. B. Shear, C. E. Schmidt, Biomaterials 2010, 31, 3930.

[73] M. M. Adil, T. Vazin, B. Ananthanarayanan, G. M. C. Rodrigues, A. T. Rao, R. U. Kulkarni, E. W. Miller, S. Kumar, D. V. Schaffer, Bio-materials 2017, 136, 1.

[74] S. Poveda-Reyes, V. Moulisova, E. Sanmartín-Masiá, L. Quintanilla-Sierra, M. Salmerón-Sánchez, G. G. Ferrer, Macro mol. Biosci. 2016, 16, 1311.

[75] J. L. Young, J. Tuler, R. Braden, P. Schüp-Magoffin, J. Schaefer, K. Kretchmer, K. L. Christman, A. J. Engler, Acta Biomater. 2013, 9, 7151.

[76] H. J. Lim, T. H. Perera, T. S. Wilems, S. Ghosh, Y. Y. Zheng, A. Azhdarinia, Q. Cao, L. A. Smith Callahan, J. Mater. Chem. B 2016, 4, 6865.

[77] L. Dreesmann, M. Ahlers, B. Schlosshauer, Biomaterials 2007, 28, 5536.

[78] L. C. Abraham, J. F. Dice, K. Lee, D. L. Kaplan, Exp. Cell Res. 2007, 313, 1045.

[79] Y. Li, W. Liu, F. Liu, Y. Zeng, S. Zuo, S. Feng, C. Qi, B. Wang, X. Yan, A. Khademhosseini, J. Bai, Y. Du, Proc. Natl. Acad. Sci. USA 2014, 111, 13511.

[80] A. Sarnowska, A. Jablonska, M. Jurga, M. Dainiak, L. Strojek, K. Drela, K. Wright, A. Tripathi, A. Kumar, H. Jungvid, B. Lukomska, N. Forraz, C. McGuckin, K. Domanska-Janik, Cell Transplant. 2013, 22, S67.

[81] H. Hosseinkhani, Y. Hiraoka, C.-H. Li, Y.-R. Chen, D.-S. Yu, P.-D. Hong, K.-L. Ou, ACS Chem. Neurosci. 2013, 4, 1229.

[82] S. Battista, D. Guarnieri, C. Borselli, S. Zeppetelli, A. Borzacchiello, L. Mayol, D. Gerbasio, D. R. Keene, L. Ambrosio, P. A. Netti, Bio-materials 2005, 26, 6194.

[83] J.-C. Lee, S. Y. Lee, H. J. Min, S. A. Han, J. Jang, S. Lee, S. C. Seong, M. C. Lee, Tissue Eng., Part A 2012, 18, 2173.

[84] J. Zhong, A. Chan, L. Morad, H. I. Kornblum, G. Fan, S. T. Carmichael, Neurorehabil. Neural Repair 2010, 24, 636.

[85] A. Ranga, S. Gobaa, Y. Okawa, K. Mosiewicz, A. Negro, M. P. Lutolf, Nat. Commun. 2014, 5, 4324.

[86] J. R. Spence, C. N. Mayhew, S. A. Rankin, M. F. Kuhar, J. E. Vallance, K. Tolle, E. E. Hoskins, V. V. Kalinichenko, S. I. Wells, A. M. Zorn, N. F. Shroyer, J. M. Wells, Nature 2011, 470, 105.

[87] M. Ruhnke, H. Ungefroren, G. Zehle, M. Bader, B. Kremer, F. Fändrich, Stem Cells 2003, 21, 428.

[88] K. H. Bouhadir, K. Y. Lee, E. Alsberg, K. L. Damm, K. W. Anderson, D. J. Mooney, Biotechnol. Prog. 2001, 17, 945.

[89] O. Jeon, D. S. Alt, S. M. Ahmed, E. Alsberg, Biomaterials 2012, 33, 3503.

[90] P. de Vos, M. M. Faas, B. Strand, R. Calafiore, Biomaterials 2006, 27, 5603.

[91] S. Mazzitelli, L. Capretto, F. Quinci, R. Piva, C. Nastruzzi, Adv. Drug Delivery Rev. 2013, 65, 1533.

[92] S. S. Ho, K. C. Murphy, B. Y. K. Binder, C. B. Vissers, J. K. Leach, Stem Cells Transl. Med. 2016, 5, 773.

[93] S. Jitraruch, A. Dhawan, R. D. Hughes, C. Filippi, D. Soong, C. Philippeos, S. C. Lehec, N. D. Heaton, M. S. Longhi, R. R. Mitry, PLoS One 2014, 9, e113609.

[94] M. Ma, A. Chiu, G. Sahay, J. C. Doloff, N. Dholakia, R. Thakrar, J. Cohen, A. Vegas, D. Chen, K. M. Bratlie, T. Dang, R. L. York, J. Hollister-Lock, G. C. Weir, D. G. Anderson, Adv. Healthcare Mater. 2013, 2, 667.

[95] A. Moshaverinia, C. Chen, X. Xu, K. Akiyama, S. Ansari, H. H. Zadeh, S. Shi, Tissue Eng., Part A 2014, 20, 611.

[96] A. Omer, V. F. Duvivier-Kali, N. Trivedi, K. Wilmot, S. Bonner-Weir, G. C. Weir, Diabetes 2003, 52, 69.

[97] Y. Teng, Y. Wang, S. Li, W. Wang, R. Gu, X. Guo, X. Nan, X. Ma, X. Pei, Tissue Eng., Part C 2010, 16, 1125.

[98] E. E. Coates, C. N. Riggin, J. P. Fisher, J. Biomed. Mater. Res. 2013, 101, 1962.

[99] J. L. Wilson, M. A. Najia, R. Saeed, T. C. McDevitt, Biotechnol. Bioeng. 2014, 111, 618.

[100] V. F. Duvivier-Kali, A. Omer, R. J. Parent, J. J. O’Neil, G. C. Weir, Diabetes 2001, 50, 1698.

[101] A. J. Vegas, O. Veiseh, M. Gürtler, J. R. Millman, F. W. Pagliuca, A. R. Bader, J. C. Doloff, J. Li, M. Chen, K. Olejnik, H. H. Tam, S. Jhunjhunwala, E. Langan, S. Aresta-Dasilva, S. Gandham, J. J. McGarrigle, M. A. Bochenek, J. Hollister-Lock, J. Oberholzer, D. L. Greiner, G. C. Weir, D. A. Melton, R. Langer, D. G. Anderson, Nat. Med. 2016, 22, 306.

[102] L. Bian, D. Y. Zhai, E. Tous, R. Rai, R. L. Mauck, J. A. Burdick, Bio-materials 2011, 32, 6425.

[103] A. C. Lima, P. Batista, T. A. M. Valente, A. S. Silva, I. J. Correia, J. F. Mano, Tissue Eng., Part A 2013, 19, 1175.

[104] L. Wang, R. R. Rao, J. P. Stegemann, Cells Tissues Organs 2013, 197, 333.

[105] S. Natesan, D. G. Baer, T. J. Walters, M. Babu, R. J. Christy, Tissue Eng., Part A 2010, 16, 1369.

[106] T. Kobayashi, Y. Aomatsu, H. Iwata, T. Kin, H. Kanehiro, M. Hisanaga, S. Ko, M. Nagao, Y. Nakajima, Transplantation 2003, 75, 619.

[107] H. Iwata, K. Kobayashi, T. Takagi, T. Oka, H. Yang, H. Amemiya, T. Tsuji, F. Ito, J. Biomed. Mater. Res. 1994, 28, 1003.

[108] H. Gin, B. Dupuy, A. Baquey, C. H. Baquey, D. Ducassou, J. Micro-encapsulation 1990, 7, 341.

[109] T. Tun, K. Inoue, H. Hayashi, T. Aung, Y. J. Gu, R. Doi, H. Kaji, Y. Echigo, W. J. Wang, H. Setoyama, M. Imamura, S. Maetani, N. Morikawa, H. Iwata, Y. Ikada, Cell Transplant. 1996, 5, S59.

[110] Y. Aizawa, N. Leipzig, T. Zahir, M. Shoichet, Biomaterials 2008, 29, 4676.

[111] S. M. Dang, S. Gerecht-Nir, J. Chen, J. Itskovitz-Eldor, P. W. Zandstra, Stem Cells 2004, 22, 275.

[112] S. Lee, C. M. Valmikinathan, J. Byun, S. Kim, G. Lee, N. Mokarram, S. B. Pai, E. Um, R. V. Bellamkonda, Y.-S. Yoon, Biomaterials 2015, 63, 158.

[113] A. Chenite, C. Chaput, D. Wang, C. Combes, M. D. Buschmann, C. D. Hoemann, J. C. Leroux, B. L. Atkinson, F. Binette, A. Selmani, Biomaterials 2000, 21, 2155.

[114] W. Zhao, X. Jin, Y. Cong, Y. Liu, J. Fu, J. Chem. Technol. Biotechnol. 2013, 88, 327.

[115] L. Plánka, A. Necas, R. Srnec, P. Rauser, D. Starý, J. Jancár, E. Amler, E. Filová, J. Hlucilová, L. Kren, P. Gál, Physiol. Res. 2009, 58, 885.

[116] S. Noel, C. Fortier, F. Murschel, A. Belzil, G. Gaudet, M. Jolicoeur, G. De Crescenzo, Acta Biomater. 2016, 37, 69.

[117] M. K. Nguyen, A. McMillan, C. T. Huynh, D. S. Schapira, E. Alsberg, J. Mater. Chem. B 2017, 5, 485.

[118] N. Riahi, B. Liberelle, O. Henry, G. De Crescenzo, Carbohydr. Polym. 2017, 161, 219.

[119] G. M. Cruise, O. D. Hegre, F. V. Lamberti, S. R. Hager, R. Hill, D. S. Scharp, J. A. Hubbell, Cell Transplant. 1999, 8, 293.

Adv. Mater. 2017, 1703948



Adv. Mater. 2017, 1703948

[120] I. Caron, F. Rossi, S. Papa, R. Aloe, M. Sculco, E. Mauri, A. Sacchetti, E. Erba, N. Panini, V. Parazzi, M. Barilani, G. Forloni, G. Perale, L. Lazzari, P. Veglianese, Biomaterials 2016, 75, 135.

[121] M. Caiazzo, Y. Okawa, A. Ranga, A. Piersigilli, Y. Tabata, M. P. Lutolf, Nat. Mater. 2016, 15, 344.

[122] W. Deng, X. Cao, Q. Wang, Y. Wang, J. Chen, Q. Yu, Z. Zhang, J. Zhou, W. Xu, P. Du, J. Chen, X. Gao, J. Yu, X. Xu, ACS Appl. Mater. Interfaces 2016, 8, 19916.

[123] R. S. Ward, K. A. White, C. A. Wolcott, A. Y. Wang, R. W. Kuhn, J. E. Taylor, J. K. John, ASAIO J. 1993, 39, M261.

[124] N. Déglon, B. Heyd, S. A. Tan, J. M. Joseph, A. D. Zurn, P. Aebischer, Hum. Gene Ther. 1996, 7, 2135.

[125] C. Rinsch, P. Dupraz, B. L. Schneider, N. Déglon, P. H. Maxwell, P. J. Ratcliffe, P. Aebischer, Kidney Int. 2002, 62, 1395.

[126] A. Lathuilière, S. Cosson, M. P. Lutolf, B. L. Schneider, P. Aebischer, Biomaterials 2014, 35, 779.

[127] K. Takebe, T. Shimura, B. Munkhbat, M. Hagihara, H. Nakanishi, K. Tsuji, Cell Transplant. 1996, 5, S31.

[128] F. Schwenter, S. Zarei, P. Luy, V. Padrun, N. Bouche, J. S. Lee, R. C. Mulligan, P. Morel, N. Mach, Cancer Gene Ther. 2011, 18, 553.

[129] F. Schwenter, B. L. Schneider, W. F. Pralong, N. Déglon, P. Aebischer, Hum. Gene Ther. 2004, 15, 669.

[130] R. H. Li, M. White, S. Williams, T. Hazlett, J. Biomater. Sci., Polym. Ed. 1998, 9, 239.

[131] M. S. Shoichet, S. R. Winn, F. T. Gentile, S. Athavale, J. Milton Harris, Biotechnol. Bioeng. 1994, 43, 563.

[132] T. H. Kim, D. B. An, S. H. Oh, M. K. Kang, H. H. Song, J. H. Lee, Biomaterials 2015, 40, 51.

[133] B. V. Sridhar, J. L. Brock, J. S. Silver, J. L. Leight, M. A. Randolph, K. S. Anseth, Adv. Healthcare Mater. 2015, 4, 702.

[134] Y. Wu, D. S. Puperi, K. J. Grande-Allen, J. L. West, J. Tissue Eng. Regener. Med. 2015, n.

[135] B. K. Mann, A. S. Gobin, A. T. Tsai, R. H. Schmedlen, J. L. West, Biomaterials 2001, 22, 3045.

[136] L. D. Amer, S. J. Bryant, Ann. Biomed. Eng. 2016, 44, 1959.[137] P. S. Hume, K. S. Anseth, J. Biomed. Mater. Res. 2011, 99, 29.[138] J. J. Roberts, P. Naudiyal, L. Jugé, L. E. Bilston, A. M. Granville,

P. J. Martens, ACS Biomater. Sci. Eng. 2015, 1, 1267.[139] W. C. Mak, K. Olesen, P. Sivlér, C. J. Lee, I. Moreno-Jimenez,

J. Edin, D. Courtman, M. Skog, M. Griffith, J. Funct. Biomater. 2015, 6, 439.

[140] A. M. Kloxin, J. A. Benton, K. S. Anseth, Biomaterials 2010, 31, 1.[141] S. Khetan, M. Guvendiren, W. R. Legant, D. M. Cohen, C. S. Chen,

J. A. Burdick, Nat. Mater. 2013, 12, 458.[142] T. T. Lee, J. R. García, J. I. Paez, A. Singh, E. A. Phelps, S. Weis,

Z. Shafiq, A. Shekaran, A. Del Campo, A. J. García, Nat. Mater. 2015, 14, 352.

[143] L. Fassina, Conf. Proc. IEEE Eng. Med. Biol. Soc. 2013, 2013, 846.[144] T. Novak, S. L. Voytik-Harbin, C. P. Neu, Acta Biomater. 2015, 11,

274.[145] S. Tasoglu, C. H. Yu, H. I. Gungordu, S. Guven, T. Vural,

U. Demirci, Nat. Commun. 2014, 5, 4702.[146] E. Popa, V. Santo, M. Rodrigues, M. Gomes, Polymers 2016, 8, 28.[147] F. Xu, T. D. Finley, M. Turkaydin, Y. Sung, U. A. Gurkan, A. S. Yavuz,

R. O. Guldiken, U. Demirci, Biomaterials 2011, 32, 7847.[148] J. E. Dixon, D. A. Shah, C. Rogers, S. Hall, N. Weston,

C. D. J. Parmenter, D. McNally, C. Denning, K. M. Shakesheff, Proc. Natl. Acad. Sci. USA 2014, 111, 5580.

[149] W. Song, A. C. Lima, J. F. Mano, Soft Matter 2010, 6, 5868.[150] A. Batorsky, J. Liao, A. W. Lund, G. E. Plopper, J. P. Stegemann,

Biotechnol. Bioeng. 2005, 92, 492.[151] A. Khademhosseini, G. Eng, J. Yeh, J. Fukuda, J. Blumling,

R. Langer, J. A. Burdick, J. Biomed. Mater. Res. 2006, 79, 522.

[152] B. R. Lee, J. W. Hwang, Y. Y. Choi, S. F. Wong, Y. H. Hwang, D. Y. Lee, S.-H. Lee, Biomaterials 2012, 33, 837.

[153] Y. Li, P. Chen, Y. Wang, S. Yan, X. Feng, W. Du, S. A. Koehler, U. Demirci, B.-F. Liu, Adv. Mater. 2016, 28, 3543.

[154] A. I. Neto, K. Demir, A. A. Popova, M. B. Oliveira, J. F. Mano, P. A. Levkin, Adv. Mater. 2016, 28, 7613.

[155] D. Velasco, E. Tumarkin, E. Kumacheva, Small 2012, 8, 1633.[156] D. Poncelet, AIChE J. 1999, 45, 2018.[157] V. Manojlovic, J. Djonlagic, B. Obradovic, V. Nedovic, B. Bugarski,

Int. J. Nanomed. 2006, 1, 163.[158] D. Serp, E. Cantana, C. Heinzen, U. Von Stockar, I. W. Marison,

Biotechnol. Bioeng. 2000, 70, 41.[159] G. H. Wolters, W. M. Fritschy, D. Gerrits, R. van Schilfgaarde,

J. Appl. Biomater. 1991, 3, 281.[160] W. Weber, M. Rinderknecht, M. Daoud-El Baba, F.-N. de Glutz,

D. Aubel, M. Fussenegger, J. Biotechnol. 2004, 114, 315.[161] J.-C. Kröger, S. Benz, A. Hoffmeyer, Z. Bago, H. Bergmeister,

W. H. Günzburg, P. Karle, G. Klöppel, U. Losert, P. Müller, H. Nizze, R. Obermaier, A. Probst, M. Renner, R. Saller, B. Salmons, I. Schwendenwein, K. von Rombs, R. Wiessner, T. Wagner, K. Hauenstein, M. Löhr, Pancreatology 2003, 3, 55.

[162] M. Löhr, A. Hoffmeyer, J. Kröger, M. Freund, J. Hain, A. Holle, P. Karle, W. T. Knöfel, S. Liebe, P. Müller, H. Nizze, M. Renner, R. M. Saller, T. Wagner, K. Hauenstein, W. H. Günzburg, B. Salmons, Lancet 2001, 357, 1591.

[163] U. Prüße, U. Jahnz, P. Wittlich, J. Breford, Landbauforsch. Völken-rode, Sonderh. 2002, 241, 1.

[164] C. Sanz-Nogués, J. Horan, K. Thompson, L. Howard, G. Ryan, M. Kassem, T. O’Brien, J. Biomed. Mater. Res. 2015, 103, 3676.

[165] S. Sakai, S. Ito, H. Inagaki, K. Hirose, T. Matsuyama, M. Taya, K. Kawakami, Biomicrofluidics 2011, 5, 13402.

[166] W. H. Tan, S. Takeuchi, Adv. Mater. 2007, 19, 2696.[167] A. A. Tomei, V. Manzoli, C. A. Fraker, J. Giraldo, D. Velluto,

M. Najjar, A. Pileggi, R. D. Molano, C. Ricordi, C. L. Stabler, J. A. Hubbell, Proc. Natl. Acad. Sci. USA 2014, 111, 10514.

[168] X. Zhou, X. J. Cheng, W. F. Liu, J. Li, L. H. Ren, Q. F. Dang, C. Feng, X. G. Chen, J. Appl. Polym. Sci. 2013, 127, 4433.

[169] S. Huang, G. Lu, Y. Wu, E. Jirigala, Y. Xu, K. Ma, X. Fu, J. Dermatol. Sci. 2012, 66, 29.

[170] C.-C. Huang, H.-J. Wei, Y.-C. Yeh, J.-J. Wang, W.-W. Lin, T.-Y. Lee, S.-M. Hwang, S.-W. Choi, Y. Xia, Y. Chang, H.-W. Sung, Biomate-rials 2012, 33, 4069.

[171] M.-H. Ho, P.-Y. Kuo, H.-J. Hsieh, T.-Y. Hsien, L.-T. Hou, J.-Y. Lai, D.-M. Wang, Biomaterials 2004, 25, 129.

[172] F. R. M. Huss, J. P. E. Junker, H. Johnson, G. Kratz, J. Plast. Reconstr. Aesthet. Surg. 2007, 60, 543.

[173] C.-J. Gustafson, A. Birgisson, J. Junker, F. Huss, L. Salemark, H. Johnson, G. Kratz, Burns 2007, 33, 726.

[174] D. J. Griffon, M. R. Sedighi, D. V. Schaeffer, J. A. Eurell, A. L. Johnson, Acta Biomater. 2006, 2, 313.

[175] Y.-J. Maeng, S.-W. Choi, H. O. Kim, J.-H. Kim, J. Biomed. Mater. Res. 2010, 92, 869.

[176] X.-B. Wu, C.-H. Peng, F. Huang, J. Kuang, S.-L. Yu, Y.-D. Dong, B.-S. Han, Hepatobiliary Pancreatic Dis. Int. 2011, 10, 509.

[177] S. L. Levengood, M. Zhang, J. Mater. Chem. B 2014, 2, 3161.[178] Y. Wang, X. Shi, L. Ren, Y. Yao, F. Zhang, D.-A. Wang, J. Biomed.

Mater. Res., Part B 2010, 93, 84.[179] X. Zhang, Z. Deng, H. Wang, Z. Yang, W. Guo, Y. Li, D. Ma, C. Yu,

Y. Zhang, Y. Jin, Biomaterials 2009, 30, 2666.[180] F. Pati, J. Jang, D.-H. Ha, S. Won Kim, J.-W. Rhie, J.-H. Shim,

D.-H. Kim, D.-W. Cho, Nat. Commun. 2014, 5, 3935.[181] B. R. Lee, K. H. Lee, E. Kang, D.-S. Kim, S.-H. Lee, Biomicrofluidics

2011, 5, 22208.[182] H. Onoe, T. Okitsu, A. Itou, M. Kato-Negishi, R. Gojo, D. Kiriya,

K. Sato, S. Miura, S. Iwanaga, K. Kuribayashi-Shigetomi,



Adv. Mater. 2017, 1703948

Y. T. Matsunaga, Y. Shimoyama, S. Takeuchi, Nat. Mater. 2013, 12, 584.

[183] Y. Jun, M. J. Kim, Y. H. Hwang, E. A. Jeon, A. R. Kang, S.-H. Lee, D. Y. Lee, Biomaterials 2013, 34, 8122.

[184] Y. T. Matsunaga, Y. Morimoto, S. Takeuchi, Adv. Mater. 2011, 23, H90.

[185] P. A. Fleming, W. S. Argraves, C. Gentile, A. Neagu, G. Forgacs, C. J. Drake, Dev. Dyn. 2010, 239, 398.

[186] S. Güven, P. Chen, F. Inci, S. Tasoglu, B. Erkmen, U. Demirci, Trends Biotechnol. 2015, 33, 269.

[187] R. Li, Adv. Drug Delivery Rev. 1998, 33, 87.[188] G. M. Cruise, D. S. Scharp, J. A. Hubbell, Biomaterials 1998, 19,

1287.[189] S. Krol, S. del Guerra, M. Grupillo, A. Diaspro, A. Gliozzi,

P. Marchetti, Nano Lett. 2006, 6, 1933.[190] M. de Groot, T. A. Schuurs, R. van Schilfgaarde, J. Surg. Res. 2004,

121, 141.[191] B. Kulseng, B. Thu, T. Espevik, G. Skjåk-Braek, Cell Transplant.

1997, 6, 387.[192] O. Veiseh, J. C. Doloff, M. Ma, A. J. Vegas, H. H. Tam, A. R. Bader,

J. Li, E. Langan, J. Wyckoff, W. S. Loo, S. Jhunjhunwala, A. Chiu, S. Siebert, K. Tang, J. Hollister-Lock, S. Aresta-Dasilva, M. Bochenek, J. Mendoza-Elias, Y. Wang, M. Qi, D. M. Lavin, M. Chen, N. Dholakia, R. Thakrar, I. Lacík, G. C. Weir, J. Oberholzer, D. L. Greiner, R. Langer, D. G. Anderson, Nat. Mater. 2015, 14, 643.

[193] H. Fujiwara, M. Ferreira, G. Donati, D. K. Marciano, J. M. Linton, Y. Sato, A. Hartner, K. Sekiguchi, L. F. Reichardt, F. M. Watt, Cell 2011, 144, 577.

[194] F. Mavilio, G. Pellegrini, S. Ferrari, F. Di Nunzio, E. Di Iorio, A. Recchia, G. Maruggi, G. Ferrari, E. Provasi, C. Bonini, S. Capurro, A. Conti, C. Magnoni, A. Giannetti, M. De Luca, Nat. Med. 2006, 12, 1397.

[195] G. Pallafacchina, S. François, B. Regnault, B. Czarny, V. Dive, A. Cumano, D. Montarras, M. Buckingham, Stem Cell Res. 2010, 4, 77.

[196] A. Urciuolo, M. Quarta, V. Morbidoni, F. Gattazzo, S. Molon, P. Grumati, F. Montemurro, F. S. Tedesco, B. Blaauw, G. Cossu, G. Vozzi, T. A. Rando, P. Bonaldo, Nat. Commun. 2013, 4, 1964.

[197] K. Saha, A. J. Keung, E. F. Irwin, Y. Li, L. Little, D. V. Schaffer, K. E. Healy, Biophys. J. 2008, 95, 4426.

[198] B. Lee, D. Clarke, A. Al Ahmad, M. Kahle, C. Parham, L. Auckland, C. Shaw, M. Fidanboylu, A. W. Orr, O. Ogunshola, A. Fertala, S. A. Thomas, G. J. Bix, J. Clin. Invest. 2011, 121, 3005.

[199] S. Gobaa, S. Hoehnel, M. Roccio, A. Negro, S. Kobel, M. P. Lutolf, Nat. Methods 2011, 8, 949.

[200] C. A. DeForest, K. S. Anseth, Annu. Rev. Chem. Biomol. Eng. 2012, 3, 421.

[201] Z. Wang, J. Wang, Y. Jin, Z. Luo, W. Yang, H. Xie, K. Huang, L. Wang, ACS Appl. Mater. Interfaces 2015, 7, 24629.

[202] H. N. Suh, H. J. Han, J. Cell. Physiol. 2011, 226, 3422.[203] S. Huveneers, H. Truong, R. Fässler, A. Sonnenberg,

E. H. J. Danen, J. Cell Sci. 2008, 121, 2452.[204] A. Kostic, M. P. Sheetz, Mol. Biol. Cell 2006, 17, 2684.[205] M. A. Wozniak, R. Desai, P. A. Solski, C. J. Der, P. J. Keely, J. Cell

Biol. 2003, 163, 583.[206] H. Xia, R. S. Nho, J. Kahm, J. Kleidon, C. A. Henke, J. Biol. Chem.

2004, 279, 33024.[207] P. Moreno-Layseca, C. H. Streuli, Matrix Biol. 2014, 34, 144.[208] G. Friedl, H. Schmidt, I. Rehak, G. Kostner, K. Schauenstein,

R. Windhager, Osteoarthritis Cartilage 2007, 15, 1293.[209] Y. Y. Li, T. H. Choy, F. C. Ho, P. B. Chan, Biomaterials 2015, 52, 208.[210] S. L. Vega, M. Kwon, R. L. Mauck, J. A. Burdick, Ann. Biomed. Eng.

2016, 44, 1921.[211] H.-Y. Lin, C.-C. Tsai, L.-L. Chen, S.-H. Chiou, Y. J. Wang,

S.-C. Hung, J. Biomed. Sci. 2010, 17, 56.[212] J. Schittenhelm, E. I. Schwab, J. Sperveslage, M. Tatagiba,

R. Meyermann, F. Fend, S. L. Goodman, B. Sipos, J. Neuropathol. Exp. Neurol. 2013, 72, 194.

[213] V. Chaturvedi, D. E. Dye, B. F. Kinnear, T. H. van Kuppevelt, M. D. Grounds, D. R. Coombe, PLoS One 2015, 10, e0127675.

[214] A. D. Doyle, N. Carvajal, A. Jin, K. Matsumoto, K. M. Yamada, Nat. Commun. 2015, 6, 8720.

[215] J. C. Friedland, M. H. Lee, D. Boettiger, Science 2009, 323, 642.[216] K. Wolf, M. Te Lindert, M. Krause, S. Alexander, J. Te Riet,

A. L. Willis, R. M. Hoffman, C. G. Figdor, S. J. Weiss, P. Friedl, J. Cell Biol. 2013, 201, 1069.

[217] R. G. Wylie, S. Ahsan, Y. Aizawa, K. L. Maxwell, C. M. Morshead, M. S. Shoichet, Nat. Mater. 2011, 10, 799.

本文献由“学霸图书馆-文献云下载”收集自网络，仅供学习交流使用。

学霸图书馆（www.xuebalib.com）是一个“整合众多图书馆数据库资源，

提供一站式文献检索和下载服务”的24 小时在线不限IP

图书馆。

图书馆致力于便利、促进学习与科研，提供最强文献下载服务。

图书馆导航：

图书馆首页文献云下载图书馆入口外文数据库大全疑难文献辅助工具

http://www.xuebalib.com/cloud/

http://www.xuebalib.com/

http://www.xuebalib.com/cloud/


http://www.xuebalib.com/vip.html

http://www.xuebalib.com/db.php

http://www.xuebalib.com/zixun/2014-08-15/44.html


toward customized extracellular niche engineering ...download.xuebalib.com/5ex0ngdcmhco.pdf · such...

Documents