organs

G Model

ARTICLE IN PRESS

COLSUB-6591; No. of Pages 14

Colloids and Surfaces B: Biointerfaces xxx (2014) xxx–xxx

Contents lists available at ScienceDirect

Colloids and Surfaces B: Biointerfaces journal homepage: www.elsevier.com/locate/colsurfb

Precise manipulation of cell behaviors on surfaces for construction of tissue/organs Wenfu Zheng, Xingyu Jiang ∗ Beijing Engineering Research Center for BioNanotechnology & CAS Key Laboratory for Biological Effects of Nanomaterials and Nanosafety, National Center for NanoScience and Technology, 11 BeiYiTiao, ZhongGuanCun, Beijing 100190, PR China

a r t i c l e

i n f o

Article history: Received 15 May 2014 Received in revised form 17 August 2014 Accepted 20 August 2014 Available online xxx Keywords: Microfluidics Cell Soft lithography Surface Chip

a b s t r a c t The use of micro/nanotechnology has become an indispensable strategy to manipulating cell microenvironments. By employing key elements of soft lithographical technologies including self-assembled monolayers (SAMs), microcontact printing (␮CP), and microfluidic pattering (␮FP) and a number of switchable surfaces such as electrochemical active, photosensitive, and thermosensitive surfaces, scientists can control the adhesion, proliferation, migration and differentiation of cells. By combining essential in vivo conditions, various physical or pathological processes such as cell–cell interaction in wound healing and tumor metastasis could be studied on well-defined surfaces and interfaces. By integrating key elements in live tissues, in vitro models mimicking basic structure and function of vital organs such as lung, heart, blood vessel, liver, kidney, and brain have been developed and greatly increased our knowledge of these important life processes. In this review, we will focus on the recent development of these interfacial methods and their application in fundamental biology research. © 2014 Elsevier B.V. All rights reserved.

1. Introduction Attaching to extracellular matrix (ECM) is a prerequisite for most mammalian cells to survive, proliferate, differentiate, and carry out their functions. The ECM, an insoluble scaffold largely comprised of proteins and other biomolecules, provides a wide range of physical and chemical cues to cells. Physical cues such as topography and stiffness in the ECM provide the mechanical support and transmit external stimulations to the cells. Chemical cues such as hormones, cytokines, ionic strength, and pH values are present in ECM, providing multiple chemical stimulations for the cells. In vivo, physical and chemical microenvironments synergistically regulate cell behavior. With the development of cell biology and regenerative medicine, the need for precise control of cell behaviors and construction of in vitro tissues has increased. Soft lithography, a set of techniques for patterning surfaces, provides convenient, effective, and low-cost means to meet the needs. The key elements of soft lithography include elastomaric stamps, masks, and prototyping. Soft lithographical techniques, such as self-assembled monolayers (SAMs), microcontact printing (␮CP), and microfluidic pattering (␮FP) have been widely used to pattern a variety of

∗ Corresponding author. Tel.: +86 1082545558. E-mail address: [email protected] (X. Jiang).

different substrates and have been extensively reviewed [1]. SAMs are 1–3 nm-thick nanostructures formed by regularly assembled organics by the adsorption of molecular constituents onto a surface. The molecules that form SAMs usually have a functional headgroup with a specific affinity for a substrate. The most extensively studied class of SAMs is derived from the adsorption of alkanethiols on gold, silver, copper, palladium, and platinum [2]. In particular, SAMs of alkanethiolates on gold are currently the most widely studied class of model organic surfaces that permit control over interfacial structure and properties. ␮CP is a technique that uses topographic patterns on the surface of an elastomeric PDMS stamp to form patterns on the surfaces of various substrates. The stamp inked with a solution containing the patterning component (thiols, activated silanes, and various ligands) can be dried and brought into conformal contact with the substrate surface to transfer the components to the substrate in the regions of contacts with high spatial definition. ␮FP is a technique that uses microfluidic channels to restrict the flow of fluids to desired regions to form patterns on various substrates. A variety of patterning components (ligands, proteins, or cells) and substrates (gold, glass, polystyrene) could be used in ␮FP. In recent years, a number of studies combining soft lithography technologies and basic biological issues emerged and have become useful for biological and medical research. This review will first discuss a variety of novel methods to control cell adhesion, migration, and differentiation on surfaces; these methods include electrochemical and photo-/thermo-responsive surfaces that result

http://dx.doi.org/10.1016/j.colsurfb.2014.08.026 0927-7765/© 2014 Elsevier B.V. All rights reserved.

Please cite this article in press as: W. Zheng, X. Jiang, Precise manipulation of cell behaviors on surfaces for construction of tissue/organs, Colloids Surf. B: Biointerfaces (2014), http://dx.doi.org/10.1016/j.colsurfb.2014.08.026

G Model COLSUB-6591; No. of Pages 14

ARTICLE IN PRESS W. Zheng, X. Jiang / Colloids and Surfaces B: Biointerfaces xxx (2014) xxx–xxx

2

in different chemistry, topography and stiffness. The following part will introduce the studies of cell–cell interactions in wound healing and tumor metastasis using microfluidic techniques. The last part will introduce the newly developed microfluidic models simulating organ-level functions of blood vessel, lung, heart, liver, kidney, and brain.

Engineered cardiac tissues were cultured on the micropatterns generated on gelatin hydrogels by ␮CP of fibronectin [14]. Based on inert/adsorptive surface, various multiple cell–cell interactions [15–18] were systematically studied. So, the inert/and adsorptive surfaces provide a convenient tool for cell biology research. 2.2. Manipulation of cell adhesion on surfaces

2. Control cell adhesion on surfaces 2.1. Patterning cells with inert and adsorptive surfaces Identification and design of new surfaces that can resist nonspecific adsorption of protein is a central goal in new biomaterial development and basic biological research. Inert surfaces provide the background necessary for spatially restricted protein adsorption or for preparing surfaces that only bind specific proteins, and are used in patterning proteins as well as cells. Poly (ethylene glycol) (PEG) is the most common molecule to result in inert surfaces, and a variety of strategies for tailoring surfaces with PEG have been developed [3]. SAMs terminated in oligo (ethylene glycol) (EGn, n > 2), resist the adsorption from solution of all known proteins and their mixtures [4]. SAMs terminated in methyl groups is hydrophobic and adsorptive for proteins. The combination of inert and adsorptive surfaces with soft lithographic techniques provides the basis for patterning of proteins and cells in most studies [2]. Using geometrical confined adsorptive patterns surrounding with inert surface based on thiol SAMs on gold surface, Chen et al. [5] controlled the endothelial cells to specific-shaped regions and studied the relationship between the cell live/death and cell shapes. The dynamic process of cell confining and releasing from patterns based on thiol-SAMs-on-gold-surface were realized by electrochemistry [6]. On the PEG based inert/adsorptive SAM system on gold surface, we studied the directional migration of cell on confined teardrop shaped pattern [7]. Kushiro et al. [8] further revealed that the gap size, teardrop asymmetry and the relative positioning of teardrop are all essential features influencing the directional cell migration. To achieve reversible control of cell adhesion, Yoon et al. fabricated gold electrode arrays on glass surface and functionalized gold electrodes with arginylglycylaspartic acid (RGD) terminated thiol to induce cell adhesion, whereas treated the glass substrate with PEG to prevent cell adhesion. By switching the electrodes to desorb the gold-thiol SAM, the attached cells on the electrodes could be detached [9]. Besides thiol SAMs, other molecules have been used to pattern cells. Fan et al. precoated pluronic copolymers on a dielectric surface with microelectrode arrays. By selectively switching on microelectrode, the perfusing of fibronectin solution could produce protein patterns on the surface which could reveal desired cell patterns on the surface [10]. By microcontact printing (␮CP) an amphiphilic, protein resisting comb polymer on glass following the incubation of the surface with fibronectin solution, Hyun et al. [11] fabricated mammalian cell patterns which could be maintained alive for more than 25 days [11]. To direct control over feature height of micropatterns, Khademhosseini et al. [12] synthesized a PEG based polymer poly ((3-trimethoxysilyl)-propyl methacrylate-ran-poly (ethylene glycol) methyl ether methacrylate (poly(TMSMA-␥-PEGMA)) and generated micropatterns on silicon oxide surface by using capillary force lithography. The patterns of cells were formed driven by patterned adhesive fibronectin and repulsive polymeric monolayers of poly(TMSMA-␥-PEGMA). Apart from gold, glass, and silicon surface, soft surfaces such as hydrogels have been utilized to pattern cells. Cao et al. [13] generated a series of microarrays of cell-adhesive RGD on a persistent non-fouling PEG hydrogel and researched the effects of spreading areas and aspect ratios of single cells on dedifferentiation of chondrocytes.

Cells are major constituents of living organisms. Most cells require adhesion to substrates in order to carry out biological activities. Understanding and exploiting cell adhesion are an active area in biological research [19]. Various model surfaces have been developed to control cell adhesion [20]. Compared with conventional “static” artificial surfaces, switchable surfaces will become possible to regulate the cellular environment in a spatiotemporally controlled manner and to address cellular dynamic activity, such as cell migration and signaling transduction, in response to matrix remodeling. Here, we will review switchable surfaces for cell biology based on electrochemical, photoresponsive, and thermal properties of surfaces. 2.2.1. Electrochemical active surfaces The stability of SAMs on gold surface varies at different range of electrochemical potentials. The potential at which the desorption of the SAMs occurs depends on the length of the alkyl chain, the degree of ordering and the number of intermolecular interactions within the SAMs, and the crystallinity of the gold substrate [21]. We have developed a method that can release patterned cells on cell adhering SAMs for free migration by applying a cathodic voltage on gold surface [6]. Furthermore, by electrochemical desorption of SAMs in localized areas defined by a microfluidic system [15,16], we selectively released patterned multiple types of cells on the substrate and simulated complex cell–cell interactions in vivo: (1) those between two types of cells that are both immobilized and confined to isolated areas; (2) those between one cell type that is immobile and another that moves freely; and (3) those between two or more types of cells that are both moving freely. This technique has the capability to pattern different types of cells with precisely controlled distances while allowing the free exchange of soluble molecules, making co-culturing of different types of cells more accessible to biologists (Fig. 1). The above methods provide the means to release attached cells, but, in some cases, precise immobilization of cells from solution onto the surface is needed. Wittstock et al. described a strategy to manipulate the cell-adhesive property of an OEG terminated SAM substrate using ultramicroelectrodes. The OEG SAMs can rapidly switch to cell adhesive by exposure to Br2 , which can be electrogenerated from Br− in aqueous solution. By using this method, cellular micropatterns could be fabricated in situ [22]. Nishizawa and colleagues reported a strategy to create a patterned surface within a microfluidic channel by locally generating hypobromous acid at a microelectrode, the heparin-coated channel surface rapidly switches from antibiofouling to protein-adhering, enabling sitespecific immobilization of proteins and cells under physiological conditions. In biological studies, repeated desorption and reattachment of cells on a substrate is required. Electrical potentials that stimulate a geometry change in the surface-bound molecules have been shown to facilitate rapid and reversible switching. Yeo et al. [23] developed a SAM that can convert O-silyl hydroquinone to benzoquinone and subsequently hydroquinone using oxidization and reduction reactions successively initiated by electrical potential. During this process, the RGD peptide present on the O-silyl hydroquinone can be selectively released to release adhered cells from the substrate. Benzoquinone group, which undergoes a selective immobilization reaction with a diene-tagged peptide, can be reduced to the hydroquinone, which prevents

Please cite this article in press as: W. Zheng, X. Jiang, Precise manipulation of cell behaviors on surfaces for construction of tissue/organs, Colloids Surf. B: Biointerfaces (2014), http://dx.doi.org/10.1016/j.colsurfb.2014.08.026

G Model COLSUB-6591; No. of Pages 14

ARTICLE IN PRESS W. Zheng, X. Jiang / Colloids and Surfaces B: Biointerfaces xxx (2014) xxx–xxx

3

Fig. 1. Left: The strategy for patterning multiple types of cells on the same substrate that allows three types of naturally occurring cell–cell interactions. Right: time-lapse phase fluorescence micrographs for the three types of cell–cell interactions between 3T6 and NIH3T3 cells. From the up to bottom: Type II cell–cell interactions: NIH 3T3 cells moved freely on the surface while 3T6 cells were confined in the stripe; Type I cell–cell interactions: all cells were well confined; Type III cell–cell interactions: both types of cells moved freely on the surface. Reproduced with permission [16].

immobilization of the diene-tagged ligand. The two dynamic properties of the substrate can sequentially release and attach cells. Ng et al. [24] reported a reversible switching method to control cell adhesion by developing a SAM that can switch from cell-repulsive to cell-adhesive and back again. The SAM formed on a silicon electrode and is composed of a protein-resistant hexa(ethylene glycol) species (EG6) containing a charged moiety, and a cell adhesive component terminating in GRGDS peptides. Under different potentials, the EG6 molecules project out or flip toward the surface to conceal or expose the RGD peptides to the cells, hence resisting or promote cell adhesion. These “electrochemical switchable surfaces” may find applications in different areas of cell biology. 2.2.2. Photoresponsive surfaces Regulating the attachment of adhered living cells from a surface is a key technological requirement to obtain specific cells in the field of cellular engineering. Photoresponsive surface is a powerful platform for achieving this goal. Photoresponsive surfaces are usually surfaces coated with photoresponsive materials whose physical and chemical properties, such as conformation, shape, wettability, potential, permeability, pH, solubility, sol–gel transition temperature, and phase separation temperature of polymer, can be changed by photoirradiation [25,26]. Many kinds of functional groups including azo groups, merocyanines, fulgides can render polymers photoresponsive. Azo groups are the most popular photoresponsive groups used in controlling cell adhesion. The azo group has been used as spacer between acrylamide anchor and RGD peptides to control cell adhesion. The distance and orientation of the RGD peptides at the surface can be changed and the poly(methyl methacrylate) surfaces allows some level of control of cell adhesion [27]. In our study, the azo unit was linked to a ligand containing RGD peptide and formed SAMs on a gold surface. The E isomer of azo can present the RGD peptide

for cell adhesion, while the Z isomer of azo can mask the RGD peptide in PEG terminated SAMs to prevent cell adhesion. The interconversion between E and Z can be achieved with two different wavelengths of lights. Because the E-to-Z isomerization is completely reversible, this method provides the means to control cell adhesion reversibly on a molecularly well-defined surface (Fig. 2A) [28]. Salierno et al. [29] used photo-activatable (“caged”) RGD adhesive peptides to enable precise light-mediated control over integrin–ligand interactions. The caged RGD peptide contains a photocleavable chromophore which inhibits integrin–ligand interactions and cell adhesion. The cage can be removed by light irradiation and the uncaged RGD is recognized by integrins and mediates cell adhesions. This photo-activatable surface is a convenient tool for studying dynamics of integrin-dependent processes (Fig. 2B). The photoreactivity of azo group could be combined with stiffness of the materials to regulate cell adhesion. Hanssens et al. [30] used polyelectrolyte multilayers (PEMs) to produce tunable substrates of different thickness and matrix stiffness. The surfaces were coated with a poly(acrylic acid)-poly(allylamine hydrochloride) polyelectrolyte bilayer functionalized with a small fraction (<1%) of an azobenzene-based photoswitchable sidegroup containing RGD peptide. The isomerization switch of the azobenzene-linked RGD and the softer substrate synergistically enhanced the cell adhesion and growth. Byambaa and co-workers [31,32] employed a photoreactive phospholipid polymer surface to control cell adhesion and detachment. The surface was prepared using the amphiphilic and water-insoluble substance poly(2-methacryloyloxyethyl phosphorylcholine-co-nbutyl methacrylate) (PMB) bearing 4-[4-(1-hydroxyethyl)-2methoxy-5-nitrophenoxy]butyric acid (PL) groups in its side chain (PMB-PL). The PL groups at the surface provide adhesion points for cells. Following UV-irradiation, the photocleavage of PL unit can efficiently convert on the PMB-PL surface to a neutral and

Please cite this article in press as: W. Zheng, X. Jiang, Precise manipulation of cell behaviors on surfaces for construction of tissue/organs, Colloids Surf. B: Biointerfaces (2014), http://dx.doi.org/10.1016/j.colsurfb.2014.08.026

G Model COLSUB-6591; No. of Pages 14 4

ARTICLE IN PRESS W. Zheng, X. Jiang / Colloids and Surfaces B: Biointerfaces xxx (2014) xxx–xxx

Fig. 2. (A) Left: the azobenzene moiety can be converted photochemically between the E and Z configurations to either present or mask the RGD ligand and hence modulate biospecific cell adhesion. Right: cells adhered onto SAMs with the azobenzene group in the E configuration. Few cells adhered to the same SAMs with azobenzene in the Z configuration. Cells adhered to the SAMs again when the conformation of azobenzene was changed from Z to E. Scale bars are 200 ␮m. Reproduced with permission [28]. (B) Left: chemical structure of the caged RGD molecule (cyclo[RGD(DMNPB)fK]) and the photolysis products (uncaged RGD and by-product). Right: migration of HUVECs on a cyclo[RGDfK] patterned SAMs was triggered after RGD activation on free spaces. Scale bar 200 ␮m. Reproduced with permission [29].

hydrophilic surface and detach the adhered cells while maintaining cell viability. 2.2.3. Thermoresponsive surfaces Temperature can be used as a very convenient tool for manipulating cell adhesion by controlling “on-off” switching of thermoresponsive surfaces prepared using thermoresponsive polymers. Thermoresponsive polymers generally exhibit a lower critical solution temperature (LCST), below which they are soluble, and above which they dehydrate and aggregate [33]. The surfaces made of these stimuli-responsive polymers switch from hydrophilic to hydrophobic states in response to temperature changes. Poly(Nisopropylacrylamide) (PIPAAm) and poly- or oligo (ethylene glycol) methacrylate (OEGMA) derivatives are popular thermosensitive polymers used for reversible cell adhesion control [34,35]. PIPAAmgrafted cell culture substrates exhibit excellent thermoresponse to control cell attachment/detachment [36]. Cells adhered on the substrate at 37 ◦ C were harvested when the temperature decreased from 37 to 20 ◦ C without the use of digestive enzymes or chelating agents. A confluent cultured cell monolayer, which is detached from the PIPAAm-grafted cell culture substrate by decreasing the temperature, is used in regenerative medicine [37]. The recovered cell sheets retaining the cell–cell junctions were easily transferred onto other materials to construct 3D tissues from cell sheets using a thermoresponsive cell culture dish [38]. Fukumori et al. [39] investigated the relationship between the concentration of PIPAAm and cell adhesion. They found that when the monomer concentration of PIPAAm was 5 wt.%, the grafted polymer was cell-adhesive, but when the monomer concentration increased to 35 wt.%, the polymer surface was cell repellent. Li et al. realized the gradient immobilization of RGD peptide on a constant thickness of poly(N -isopropylacrylamide) PNIPAAm. They fabricated a molecular weight gradient of PAA on the underlying uniform

PNIPAAm layer and covalently immobilized RGD peptide onto the PAA gradient by carbodiimide chemistry. The immobilization of the RGD peptide could accelerate HepG2 cell attachment, while the thermoresponsive layer beneath could effectively release the cells by simply lowering temperature [40]. Circulating tumor cells (CTCs) have been proven to be a marker to predict cancer progression and survival and have the potential to guide therapeutic management. However, how to keep CTCs undamaged during the processes of capture/release is still challenging. By coating thermoresponsive PNIPAAm on silicon nanopillar array, Liu et al. [41] achieved reversible capture and release of targeted cancer cells by combining hydrophobic interactions with topographic interactions. The study provides a new strategy to provide undamaged CTCs for subsequent analysis. The ability to reversibly control the interactions between the ECM and cell surface receptors such as integrins would allow one to investigate reciprocal signaling circuits between cells and their surrounding environment (Fig. 3A). Zhu et al. [42] reported a protocol to fabricate biofunctionalized micropatterned gold nanoarrays that support integrin-mediated cell adhesion and function as plasmonic nanostoves to physically block and orient the formation of focal adhesion sites and the subsequent cell polarization and migration. Being reversible and not restricted spatiotemporally, thermoplasmonic approaches will open new opportunities to further study the complex interactions between ECM and cells (Fig. 3B). 3. Manipulation of cell migration on surfaces Cell migration is essential for many fundamental biological processes including embryonic morphogenesis, tissue repair and regeneration, and disease progression in cancer, vascular disease, and chronic inflammations [43]. Cell migration is an integrated multistep process among which polarization is the first

Please cite this article in press as: W. Zheng, X. Jiang, Precise manipulation of cell behaviors on surfaces for construction of tissue/organs, Colloids Surf. B: Biointerfaces (2014), http://dx.doi.org/10.1016/j.colsurfb.2014.08.026

G Model COLSUB-6591; No. of Pages 14

ARTICLE IN PRESS W. Zheng, X. Jiang / Colloids and Surfaces B: Biointerfaces xxx (2014) xxx–xxx

5

Fig. 3. (A) Reversible capture and release of targeted cancer cells on as-prepared surfaces triggered by temperature. Using biotin-BSA as a hydrophobic anchor, targeted MCF-7 cells can be captured onto or released from the PNIPAAm-modified SiNP (PSiNP) reversibly, by changing the temperature between 37 ◦ C and 20 ◦ C. The insets indicate the quantification of the targeted cancer cells on the surface. Reproduced with permission [41]. (B) Left: High-resolution SEM images of micropatterned gold nanoarrays of various dimensions; Right: Cell adhesion on RGD-coated 10 ␮m wide lines. Phase contrast microscopy images of mouse fibroblasts migrating onto RGD-functionalized gold nanoparticles of 35 nm in diameter separated by 90 nm. Confocal fluorescence microscopy images of fixed cells expressing GFP-labeled paxillin and stained for filamentous F-actin (in red) displaying, respectively, adhesive sites and cytoskeleton of a cell migrating onto RGD-coated gold nanoparticles. Reproduced with permission [42]. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

initial response of a cell to a migration-promoting agent. Cells are inherently sensitive to local micro-/nano-scale topographic and molecular patterns in the ECM environment, a phenomenon called “contact guidance”. A typical outcome of contact guidance effect of the ECM is the polarization of cells and their directional migration. Surface structures can significantly influence cell orientation, migration direction, as well as migration speed in the directions parallel and perpendicular to grooves/ridge in a surface structure-dependent manner [44]. Besides surface topography, stimulations such as chemical gradient and fluid flow also play roles in cell migration. Han et al. [45] assembled poly(sodium 4styrenesulfonate) (PSS)/poly(diallyldimethylammonium) chloride (PDADMAC) multilayers on PDMS membrane with linear grooves and ridges and treated the surface with a gradient solution of 3–5 M NaCl. The synergetic effects of the surface topography and swelling gradient can effectively guide the unidirectional migration of single smooth muscle cells. A study on the combination of flow shear stress and groove guidance on endothelial cell migration revealed that when flow direction was oriented parallel to microgrooves, the cells migrated along the microgrooves. When microgrooves were oriented perpendicular to the flow, most cells migrated orthogonally to the grooves and downstream with the flow [46]. Compared with symmetrical topographies, asymmetric surface patterns can trigger the polarization of cells and bias the

direction of cell movement. For example, we studied the relationship between the direction of cell migration and its asymmetric shape. We [7] designed teardrop-like asymmetric geometries to first restrict and then release the cells. The cells tend to move toward blunt ends after desorption from the substrate (Fig. 4A). Combining stripe and teardrop geometric cues into a hybrid, spearshaped micropattern was reported to yield combinatorial benefits in cell speed, persistence, and directional bias. Cell migration speed and persistence were enhanced in a predictable, additive manner on the modular spear-shaped design. Meanwhile, the spear micropattern also improved the directional bias of cell movement compared to the standard teardrop geometry, revealing that combining geometric features can also lead to unexpected synergistic effects in certain aspects of cell motility (Fig. 4B) [47]. The mechanical microenvironment is known to influence single-cell migration; however, the extent to which mechanical cues affect collective migration of adherent cells is not well understood. Ng. et al. [48] measured the effects of varying substrate compliance on individual cell migratory properties. They found that increasing substrate stiffness increased collective cell migration speed, persistence, and directionality as well as the coordination of cell movements. Apart from physical and chemical cues, the cell–cell interactions also influence cell behavior. We employed Madin–Darby canine kidney (MDCK), a cell line with relatively strong intercellular interactions,

Please cite this article in press as: W. Zheng, X. Jiang, Precise manipulation of cell behaviors on surfaces for construction of tissue/organs, Colloids Surf. B: Biointerfaces (2014), http://dx.doi.org/10.1016/j.colsurfb.2014.08.026

G Model COLSUB-6591; No. of Pages 14 6

ARTICLE IN PRESS W. Zheng, X. Jiang / Colloids and Surfaces B: Biointerfaces xxx (2014) xxx–xxx

Fig. 4. (A) A cartoon illustration of the migration of a typical mammalian cell on a flat surface. This teardrop shape is found in many types of cells. Reproduced with permission [7]. (B) Design of a hybrid spear-shaped micropattern and its effect on the directional movement of cells. Left: design schematic of the spear-shaped pattern. Right: time-lapse images display the movement of cells on (upper) teardrop- and (bottom) spear-shaped patterns. The white arrows indicate the direction of cell movement. Scale bar is 20 ␮m. Reproduced with permission [47]. (C) Upper: three types of axon behaviors in the branch channel region. Scale bar is 100 ␮m. Bottom: an unbranched axon on the micropattern generated by ␮CP only, Scale bar is 100 ␮m; A type II axon branch on geometry with 50 ␮m strips, Scale bar is 50 ␮m; Axon branching on the triangle micropattern, Scale bar is 50 ␮m. Reproduced with permission [50]. (D) Integration of microfluidic channels with replica mold of aligned ES fibers for guiding neurites. The guidance of neurites on migrogrooves was compared with control experiments of neurons without microgrooves. The green fluorescence channel was the immunocytochemistry of with neuron-specific marker Tuj1. Reproduced with permission [52]. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

and NIH 3T3 fibroblast cells, a cell line with relatively weak intercellular interactions, to study the interplay between contact guidance and intercellular connections on substrates with microgrooves. Although MDCK cells migrate much more slowly than 3T3 cells on flat substrate, the velocity of migration of MDCK cells parallel to the grooves is higher than that of 3T3 cells perpendicular to the grooves. This may be due to synergistic interplay of the group behavior of the MDCK cells and the topography of the surface [49]. The migration and fasciculation of growth cones are essential for embryogenesis and nervous system development. Axons are guided by gradients of chemoattractant or chemorepellent cues and travel in bundles or fascicles to reach their final destination. By using a micropatterned coculture system, we produced a Slit gradient on a surface and induced directional axon migration and fasciculation. This approach establishes a simple and stable axon guidance model in vitro, and may be broadly applicable for investigating long-term events such as axon regeneration [50]. Axon branching enables neurons to contact with multiple targets and respond to their microenvironment. The chemical properties of

ECM and geometric constrains have been proposed to regulate axonal development. We demonstrated that both the sharp broadening of substrate geometry and the sharp change of laminin density stimulate axon branching by ␮CP and ␮FP techniques. We also found that the change of axon branching stimulated by laminin density depends on myosin II activity (Fig. 4C). These previously unknown mechanisms of change of laminin density-stimulated axon branching may explain how the ECMs regulate axon branching in vivo and facilitate the establishment of neuronal networks in vitro [51]. Neural circuits are composed of neurons interconnected through elongated neurites. The guidance of neurites to specific locations in order to connect with other neurons is a key step for achieving precise wiring in the nervous system. We [52] combined microfluidics with the replica-molded substrate from electrospun fibers to control the morphology of neurons, making possible the guidance of neurites to form interconnected circuits in vitro (Fig. 4D). Neurons are dynamically coupled with each other through neurite-mediated adhesion during development. Understanding the collective behavior of neurons in circuits is important

Please cite this article in press as: W. Zheng, X. Jiang, Precise manipulation of cell behaviors on surfaces for construction of tissue/organs, Colloids Surf. B: Biointerfaces (2014), http://dx.doi.org/10.1016/j.colsurfb.2014.08.026

G Model COLSUB-6591; No. of Pages 14

ARTICLE IN PRESS W. Zheng, X. Jiang / Colloids and Surfaces B: Biointerfaces xxx (2014) xxx–xxx

for understanding neural development. We [53] established a 2D model for studying collective neuronal migration of a circuit, with hippocampal neurons from embryonic rats on Matrigel-coated SAMs. When the neural circuit is subject to geometric constraints of a critical scale, the collective behavior of neuronal migration is spatiotemporally coordinated. Neuronal somata tend to aggregate at the geometric center of the circuit, forming mono-clusters. The discovery of geometry-dependent collective neuronal migration and the formation of somata clustering in vitro shed light on neural development in vivo.

7

the structures increased in height from 15 to 55 to 100 nm, showing the importance of feature height in cell differentiation induction. A major obstacle to the application of stem cells is their propensity to spontaneously differentiate, with the loss of multi-lineage potential when grown on standard tissue culture plastics in vitro. To address this problem, McMurray et al. [77] identify a nanostructured surface that retains stem-cell phenotype and maintains stem-cell growth over eight weeks, providing new opportunities for the maintenance of stem cell pluripotency. 4.2. Stiffness

4. Cell differentiation on surfaces Cell differentiation is manipulated by a wide variety of chemical cues, ranging from cell-surface proteins to soluble cytokines and molecules of the ECM. However, recent studies have shown that mechanical cues, including the micro/nanotopography of the adhesion surface, the stiffness of the substrate, and extracellular forces, are able to direct stem cell fate in vitro, even in the absence of biochemical factors [54–56]. Here, we focus on the influence of mechanical cues on the differentiation of various cell types. 4.1. Micro/nanotopography ECM can be organized into fibers, sheets, and other features that range in sizes from nanometers to many centimeters. Being comparable to these ECM sizes, micro/nanotopography has been reported to exert dramatic impact on cell differentiation. The studies on micro/nanotopography of various geometries and sizes have shown their effects on embryonic stem cells (ESC) [57,58], mesenchymal stem cells (MSCs) [59–62], and neuron stem cells [63,64]. For example, alignment patterns on the surface enhance cell orientation and improve the polarity, which is important in cell development and differentiation. Nanofibrous scaffolds showed neuron differentiation tendency with neural induction factors [65]. MSCs on aligned carbon nanotube networks exhibited enhanced proliferation and osteogenic differentiation compared to those on randomly oriented carbon nanotube networks [59]. Grooves and grids, inducing cells alignment and elongation, show a greater effect on osteogenic differentiation compared to flat surfaces in MSCs [60,61]. Moreover, aligned/nonaligned nanofibres [66] or low roughness silk-tropoelastin scaffolds [67] can lead to myogenic commitment of MSCs in the absence of differentiating medium. Nanoscale ridge/groove pattern arrays alone can effectively and rapidly induce the differentiation of ESC into a neuronal lineage without the use any differentiation-inducing agents, indicating the significant role of topography in determining cell fates [68]. Apart from aligned surface patterns, other topographies also have differentiation-inducing effects. For example, circular patterns can direct cardiac differentiation of uniform-sized mouse ESC aggregates [69] and adipogenic lineage differentiation of MSCs [44]. Nanogratings can induce osteogenic transition of MSCs [70] and an upregulation on neuronal markers of MSCs even in the absence of retinoic acid as a biochemical cue [71]. Random topographies such as disordered nanopits were able to induce MSCs to differentiate toward osteogenic lineage [72–74]. Other factors, such as size [75] and height of topographical features can also regulate cell differentiation [76]. Nanotopography alone can induce the differentiation of MSCs into neuronal lineage and induced a more significant upregulation of neuronal markers compared to microtopography, highlighting the importance of feature size in topography-induced differentiation [75]. Another work indicates that 15-nm high pillars can induce more apparent osteogenic differentiation of MSCs compared with that on the planar control. Moreover, the differentiation tendency decreased as

In tissues, cells and the ECM co-contribute to a relatively elastic microenvironment [78]. The stiffness of the ECM is known to impact cell activities such as gene transcription, cytoskeleton remodeling, and cell differentiation [79–82]. From neurons to osteoblasts, the stiffness of the ECM ranges from about 1 kPa in brain to 100 kPa in collageous bone [83]. Most of the cells not only sense but also respond to the mechanical properties of the ECM by adjusting their focal adhesion structure, cytoskeleton organization, and overall state [84–86]. Engler et al. [83,87] found that the differentiation of human mesenchymal stem cells (hMSCs) depend on elasticity of the collagen-coated polyacrylamide substrate on which they were cultured. hMSCs grown on moderately stiff substrates (E ∼ 10 kPa, similar to stiffness of muscle) exhibited upregulated myogenic markers, whereas hMSCs proceeded down an osteogenic pathway when cultured on a stiffer substrate (E ∼ 35 kPa, similar to stiffness of bone). Healy et al. [88] found that soft PEG-peptide-based materials (E ∼ 0.5 kPa, similar to stiffness of brain tissue) promote neural differentiation of neuronal stem cells in serum-free media, whereas stiff gels (E ∼ 1–10 kPa) promoted differentiation to glial cells. Stem cells that naturally reside in adult tissues exhibit robust regenerative capacity in vivo that is rapidly lost during in vitro culture. Gilbert et al. [89] reported that muscle stem cells cultured on soft hydrogel substrates that mimic the elasticity of muscle (12 kPa) self-renew in vitro and contribute extensively to muscle regeneration when subsequently transplanted into mice. Stiffness, the intrinsic property of tissues, will be useful in designing tissue constructs toward tissue regeneration and therapeutic objectives. 5. Cell–cell interactions on surfaces/interfaces 5.1. Wound healing Wound healing is a complex process that is critical for preserving the integrity of multicellular organisms and tissue homeostasis. Traditional in vitro models of wound healing involve scratching a confluent cell monolayer with a microneedle or micropipette tip. However, the poor controllability of the depth and width of the wound, the contraction of the cell wave front, the removal of underlying matrix of specific ligands all influence the outcome of experiments. Alternative strategies are developed to create wounds to overcome these shortcomings. Murrel et al. [90] utilized multiple laminar flows to selectively cleave cells enzymatically to deconstruct the classical wound healing scratch assay, providing direct and quantitative evidence of strong tension within the cell sheet. Lee et al. [91] introduced a method involving creation of defined patterns of damaged cell debris with PDMS stamping. This assay permitted the quantification of wound healing rates in the presence of cell debris in vitro. Being comparable but different to Lee’s method, Hettler et al. [92] used heatable thermostat-controlled aluminum stamps to investigate thermomechanical damage induced wound healing. They found that the unaffected cells surrounding the thermomechanical damage zone can migrate into the damaged area, resulting in a complete closure

Please cite this article in press as: W. Zheng, X. Jiang, Precise manipulation of cell behaviors on surfaces for construction of tissue/organs, Colloids Surf. B: Biointerfaces (2014), http://dx.doi.org/10.1016/j.colsurfb.2014.08.026

G Model COLSUB-6591; No. of Pages 14 8

ARTICLE IN PRESS W. Zheng, X. Jiang / Colloids and Surfaces B: Biointerfaces xxx (2014) xxx–xxx

Fig. 5. (A) A microchip model for co-culture and selective wounding. Magnified drawing of the device shows the position of all three microchannels: MDCK cells alone or co-cultured with NIH 3T3 were patterned on a surface. MDCK cells in the middle channel were selectively lysed by distilled water. Reproduced with permission [97]. (B) Top and side views show the device schematic with the endothelial monolayer, the tumor cells, and the location of the 3D ECM. Bottom: confocal images demonstrate intravasation of breast carcinoma cells (green) across the endothelium (MVEC, stained red for VE-cadherin). (Scale bar: 30 ␮m.) Reproduced with permission [104]. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

of the “wound” within 48 h. Poujade and coworkers [93] designed a model to obtain parallel cell stripes on a surface to mimic a “wound” with no injury. The cell stripes acquired by this model offered minimal damage to the boundary cells, and confirmed that the free space is sufficient to trigger epithelial collective migration. Other techniques have been proposed to generate wounds with specific shapes and sizes on confluent cell monolayer, such as electrical wounding [94], laser wounding [95], and protease trypsin wounding [96], to investigate the transmission of signals from a wound to adjacent cells. To study the influence of remnant damaged cells on epithelial collective migration, we [97] achieved co-culturing cells and making a “wound” without mechanical tension on adjacent normal cells. We confirmed that fibroblasts and lysed epithelial cells enhanced epithelial collective migration even without the influence of free physical space (Fig. 5A). To produce uniform-shaped wound model, Zhang et al. [98] presented a simple microfluidic strategy to minimize the injury of cells during the formation of wound. They create the cell free (wound) area and cell loading (residual) area by using a two-layer microfluidic device with micropillars, offering a uniform initial area for the dynamic wound healing assay. Understanding and accelerating the mechanisms of wound healing is of fundamental interest for biotechnology and clinical research in repairing invasive medical interventions. Franco et al. [99] reported the influence of substrate topography and flow on the efficiency of endothelial regeneration. They found that flow significantly accelerated the healing process

of the wound. The substrate topography modulated the mechanical connection between migrating cells and increased their coordination and integrity, thus accelerating wound healing. The development of materials and formulations for wound dressings is laborious and ethically challenging due to the use of animal experiments. A method for rapid and effective screening of wound dressings in vitro, therefore, is in great need. We [100] reported an in vitro cell-on-a-chip model for rapid wound dressing screening and found that the results acquired by microchip model corroborates well with animal experiments. The cell-on-achip wound model may change the way for wound dressing screen. 5.2. Tumor metastasis Tumor metastasis is a complicated and dynamic process involving angiogenesis, tumor cell proliferation and dissociation, stromal invasion, intravasation, circulation, extravasation, and metastatic tumor formation [20]. In the process, a number of interactions among tumor cells and between tumor cells and the host tissue take place. The recreation of cancer models on in vitro enables the study of spatiotemporal interaction of cells during the tumor metastasis. Entry of tumor cells into the blood stream is a critical step in cancer metastasis. Although significant progress has been made in visualizing tumor cell motility in vivo, the underlying mechanism of cancer cell intravasation remains largely unknown. Invadopodia or invasive feet, which are actin-rich membrane protrusions with

Please cite this article in press as: W. Zheng, X. Jiang, Precise manipulation of cell behaviors on surfaces for construction of tissue/organs, Colloids Surf. B: Biointerfaces (2014), http://dx.doi.org/10.1016/j.colsurfb.2014.08.026

G Model COLSUB-6591; No. of Pages 14

ARTICLE IN PRESS W. Zheng, X. Jiang / Colloids and Surfaces B: Biointerfaces xxx (2014) xxx–xxx

matrix degradation activity formed by invasive cancer cells, are a key determinant in the malignant invasive progression of tumors and represent an important target for cancer therapies. For tumor cells, the ability to form invadopodia is closely related to their invasive and metastatic properties [101]. During intravasation, invadopodia-like protrusions in tumor cells have been observed in vivo by intravital imaging [102]. Therefore, invadopodia are proposed to function in local ECM degradation during cancer invasion and metastasis. Wang et al. [103] presented a microfluidic 3D culture device mimicking tumor microenvironment in vivo. The device could be used to assay invadopodia formation and to study the mechanism of human lung cancer invasion. Their results demonstrated that epidermal growth factor receptor (EGFR) signaling played a significant role in invadopodia formation. The device provided an applicable platform for elucidating the mechanism of cancer invasion. To elucidate the tumor–endothelial interactions, Kamm’s research group [104] developed a microfluidic-based assay to recreate the tumor–vascular interface in 3D, allowing for high resolution, real-time imaging, and precise quantification of endothelial barrier function. They found that endothelial barrier impairment was associated with a higher number and faster dynamics of tumor–endothelial cell interactions. Endothelium poses a barrier to tumor cell intravasation which can be regulated by factors present in the tumor microenvironment (Fig. 5B). CTCs survive the vascular system can eventually extravasate across the endothelium to metastasize at a secondary site. The ability to properly model extravasation is essential in identifying key physical, cellular, and molecular determinants that can be targeted therapeutically to prevent metastatic disease. But the exact mechanism of extravasation is still unknown. With the development of microfluidics, many extravasation models have been developed [105,106]. Song et al. [106] present a microfluidic vasculature system to model interactions between circulating breast cancer cells with microvascular endothelium at potential sites of metastasis. The microfluidic vasculature produces spatially-restricted stimulation from the basal side of the endothelium that models both organ-specific localization and polarization of chemokines under variable flow conditions. Their studies suggest that targeting CXCL12-CXCR4 signaling in endothelium may limit metastases in breast. Kamm’s group [105] developed a microfluidic system to mimic tumor cell extravasation where cancer cells can transmigrate across an endothelial monolayer into a hydrogel that models the extracellular space. Their results showed that the presence of tumor cells increases endothelial permeability, and extravasation generally occurs within the first 24 h of tumor cell contact with the endothelium.

6. Organs on chips Although in vitro cell research systems have greatly promoted the development of biology and medicine, they often suffer from a lack of physiological relevance which has been a bottleneck in biological research and drug development [107]. The emergency and progress of micro/nanotechnology has enabled recapitulation and manipulation of physiological and pathological tissue microenvironment, which has led to the development of novel in vitro tissue or organ models termed “organ-on-a-chip” systems [108–111]. The micro-engineered in vitro models combining microfluidics, tissue organization, soluble gradients and mechanical stimulations are called “organs-on-chips”. These chips often consist of multiple channels, splitting and merging devices, pumps, valves, and integrated electrical and biochemical sensors. In these organ-on-chips models, some key aspects of organ-level microenvironment are mimicked and applied to cells or organ-like tissues by engineering and integrating the complex components. Various

9

“organ-on-a-chip” systems, such as blood vessel [112–122], lung [123–129], heart [130–137], kidney [138–142], liver [143–148], brain [149,150], and gut [151], have been reported to reproduce target organ structures and functions better than conventional in vitro model systems. In the following section, we will introduce latest progress in “organ-on-a-chip” in the context of precise manipulation of cells on interfaces. We highlight the recapitulation of blood vessel model on a chip. 6.1. Blood vessel 6.1.1. Vasculature and angiogenesis At the core of tissue engineering is the construction of 3D scaffolds out of biomaterials to provide mechanical support and guide cell growth into new tissues or organs. Blood vessel, composed of intima, media, and adventitia, is a typical, complex 3D structure. Recently, we [113] reported a new fabrication strategy that results in stable tubular tissue with a high structural similarity to blood vessel. First, we delivered and patterned different types of cells on a 2D stress-induced rolling membrane (SIRM) using microfluidic channels. Then, the SIRM was released to roll up into a 3D tube. The tubes had different types of cells at specific locations (Fig. 6A). To further recapitulate the spatial arrangement of cellular interactions in vitro, we first patterned a monolayer of SA-coated cells on the SIRM, then seeded biotinylated cells on the monolayer to form a bilayer which was transformed into 3D tubes. Within the tube, the two types of cells can directly interact and communicate with each other, mimicking the in vivo conditions of blood vessel [114]. Microvascular networks support metabolic activity and define microenvironmental conditions within tissues. Recapitulation of functional microvascular structures in vitro could provide a platform for the study of complex vascular phenomena, including angiogenesis and thrombosis. We [142] reported a microfluidic method that utilized fibrillogenesis of collagen and liquid mold to engineer 3D vascular networks in hydrogel. This technique enables the mimicry of passive diffusion in nephron and would be used for in vitro modeling of other vasculature-rich tissues and organs. Zheng et al. [121] used lithographic technique to form endothelialized microfluidic vessels within a native collagen matrix and elucidated the angiogenic activities of the endothelia. This microvascular network is a useful platform for the study of cardiovascular biology and pathophysiology. Angiogenesis is a complex morphogenetic process whereby endothelial cells from existing vessels invade as multicellular sprouts to form new vessels. Recently, Nguyen et al. [120] engineered a 3D ECMbased model of angiogenic sprouting and neovessel formation. This in vitro 3D biomimetic model reconstituted the morphogenetic steps of angiogenic sprouting. Miller [117] generated cylindrical networks that could be lined with endothelial cells and perfused with blood under high-pressure pulsatile flow. The perfused vascular channels sustained the metabolic function of primary rat hepatocytes in engineered tissue constructs. Ye et al. [116] reported a biodegradable microvessel scaffold employing poly(glycerol sebacate) (PGS) elastomer as the framework. In vivo studies demonstrated biodegradation of the membrane interface and host blood cell infiltration of the microvessels. This scaffold could serve as a basis for building tissue constructs of increasing scale and clinical relevance. 6.1.2. Blood vessel models In the cardiovascular system, endothelial cells (ECs) are constantly subjected to hemodynamic forces in the form of fluid shear stress (FSS) and cyclic stretch (CS) which are important for the maintenance of the physiological condition of the cardiovascular system and the development of cardiovascular diseases. We [152] developed a microfluidic flow-stretch chip that integrates FSS and

Please cite this article in press as: W. Zheng, X. Jiang, Precise manipulation of cell behaviors on surfaces for construction of tissue/organs, Colloids Surf. B: Biointerfaces (2014), http://dx.doi.org/10.1016/j.colsurfb.2014.08.026

G Model COLSUB-6591; No. of Pages 14 10

ARTICLE IN PRESS W. Zheng, X. Jiang / Colloids and Surfaces B: Biointerfaces xxx (2014) xxx–xxx

Fig. 6. (A) Left: schematic illustration of a SIRM and tubes with multiple types of cells. A thin PDMS membrane is stretched as the top layer of the SIRM and covers a semicured PDMS membrane. After curing the two layers to cause adhesion, a SIRM is obtained that rolled up when the ends were released. Different cells are delivered via microfluidic channels to the surface of the SIRM. Once the cells attach to the SIRM surface, the microchannels are removed. One end of the SIRM is released by cutting its edge. The SIRM rolls up into a tube and each type of cell is delivered to a designated position as the tubular wall. Right: three types of cells on a SIRM before and after rolling. Each color indicates a different type of cell. Red are endothelial cells (HUVECs); green: smooth muscle cells; blue: fibroblasts (NIH 3T3). Reproduced with permission [113]. B) Left: a schematic of the microfluidic flow-stretch chip. Right: the stress fibers tend to align in the direction parallel to that of the FSS, CS and the resultant force of the FSS and CS. Reproduced with permission [152]. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

CS for cultured cells. The chip can deliver FSS and CS simultaneously or independently to vascular cells to mimic the hemodynamic microenvironment of blood vessels in vivo. By imposing FSS-only, CS-only, and FSS + CS stimulation on rMSCs, we found the alignment of the cellular stress fibers varied with the types of stimulations (Fig. 6B). The flow-stretch chip is a reliable tool for simulating the hemodynamic microenvironment. The vascular structure consists of intima, media, and adventitia from inner to outer with distinct patterns. [153,154]. To mimic the vascular vessel structure, it is important to consider the alignment and orientation of the cells. Choi et al. [112] constructed a biomimetic smooth muscle cell layer which is aligned perpendicular to the axis of blood vessel. They generated orthogonally microwrinkle patterns inside a circular microchannel to guide the circumferential alignment of human aortic smooth muscle with high viability, providing a bioassay platform for in-depth study of vascular function. MSCs may migrate through the blood stream and initiate pathology of atherosclerosis. However, how the MSCs sense and respond to vascular FSS stimulation and lead to subsequent biological effects remains elusive.

By using an in situ time-lapse microfluidic cell culture and observation system [155], we found that MSCs presented a contraction and re-spread (CRS) process when they were initially subjected to a physiological FSS. Our study revealed the immediate response of the rMSCs to FSS. 6.2. Other organs The epithelial–endothelial interface, epithelium–air interface and endothelium–blood interface comprise the elementary unit of the lung tissue. How to reproduce this structure and its basic function is challenging. Recently, alveolar re-epithelialization has been studied on an alveolo capillary model system [125]; Pulmonary acinar flows were simulated on a microfluidic platform [123]; A microfluidic device comprising a branched microvascular network that provides controlled wall shear stress and uniform blood flow was developed to minimize blood damage, thrombosis and inflammatory responses seen in current oxygenators with high gas permeability for artificial lung applications[156]. Considering

Please cite this article in press as: W. Zheng, X. Jiang, Precise manipulation of cell behaviors on surfaces for construction of tissue/organs, Colloids Surf. B: Biointerfaces (2014), http://dx.doi.org/10.1016/j.colsurfb.2014.08.026

G Model COLSUB-6591; No. of Pages 14

ARTICLE IN PRESS W. Zheng, X. Jiang / Colloids and Surfaces B: Biointerfaces xxx (2014) xxx–xxx

11

Fig. 7. (A) Left: biologically inspired design of a human breathing lung-on-a-chip microdevice, the device recreates physiological breathing movements by applying vacuum to the side chambers and causing mechanical stretching of the PDMS membrane forming the alveolar–capillary barrier. Right: Long-term microfluidic coculture produces a tissue–tissue interface consisting of a single layer of the alveolar epithelium (stained with CellTracker Green) closely apposed to a monolayer of the microvascular endothelium (stained with CellTracker Red), both of which express intercellular junctional structures stained with antibodies to occludin or VE-cadherin, separated by a flexible ECM-coated PDMS membrane. Scale bar, 50 ␮m. Reproduced with permission [124]. (B) Failing myocardium on a chip design. Left: (i) Schematic of the heart. (ii) Ventricular myocardium consists of aligned, elongated cardiac myocytes. (iii) Schematic of failing myocardium on a chip system adapted for the muscular thin film assay. (iv) Photograph of stretchable muscular thin film experiment. Right: neonatal rat ventricular myocytes on micropatterned membrane under different conditions (Detail see Ref. [131]). Reproduced with permission [131]. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

the mechanical stimulations in the alveoli, a microfluidic “alveolion-a-chip” was reported [128]. This device incorporates cyclic propagation of a meniscus over a flexible PDMS membrane that recreated stresses in many surface tension diseases such as infant respiratory distress syndrome (IRDS). Another microdevice incorporating alveolar–capillary air–liquid interface was reported by Huh et al. [124]. The device allows controlled mechanical stretch of the endothelial–epithelial bilayer, mimicking the mechanical cues present in the lung during breathing (Fig. 7A). One significant challenge of treatment for cardiovascular diseases is the lack of an established in vitro model of human cardiac tissue. Recently, with the advent of microfluidic technologies, many cardiac models concerning mechanics emerged. Grosberg et al. [130] reported a “heart-on-a-chip” device with ventricular myocardium on an elastomeric thin film to replicate the hierarchical tissue architectures of laminar cardiac muscle. To recapitulate the architecture necessary for a physiologically accurate heart response, Chen et al. [134] developed a simple way to create large areas of aligned cardiomyocytes using shrink-wrap film. Drug assays on the chip indicate that the cells were more sensitive to the drugs than in other in vitro assays. Hemodynamics was considered and by Chen et al. [133]. They generated cardiac-like flow in a microfluidic device and observed the morphological of cells in the system. The heart is one of the least regeneratable organs in the body and any major insult can result in a significant loss of heart cells. The development of in vitro-based cardiac tissue is important for cardiology research [157]. Serena et al. [158] tested the effects of

electrical stimulation and H2 O2 on human cardiomyocytes (hCMs) cultured on a poly-acrylamide hydrogel with tunable mechanical properties and a micropatterning array. Ren et al. [89] described a micropillar array-aided tissue interface mimicking microfluidic device for the dynamic study of hypoxia-induced myocardial injury in a microenvironment-controllable manner. McCain et al. [131] designed a system to mimic mechanical overload in vitro by applying cyclic stretch to engineered laminar ventricular tissue on a stretchable chip. The overload led to pathological remodeling of the myocytes, suggesting that failing myocardium can be replicated by an in vitro microsystem (Fig. 7B). Aiming at arrhythmia, Ma et al. [135] developed an in vitro model of 3D human cardiac tissue by populating synthetic filamentous matrices with cardiomyocytes derived from pluripotent stem cells (iPS-CMs) to mimic the condensed and aligned human ventricular myocardium. The iPS-CMs exhibited different level of contractility abnormality and susceptibility to drug-induced cardiotoxicity. Kidney is one of the important organs in the body. There is an urgent need for recapitulating kidney function in vitro for fundamental research and tissue engineering. Jiang et al. [139] developed a simple multi-layer microfluidic device by integrating a microfluidic channel and a porous membrane substrate. FSS elicited enhanced renal tubular cell polarization. Exposure of the human kidney proximal tubule epithelial monolayer to an apical FSS resulted in enhanced albumin transport and glucose reabsorption in the cells [138]. Frohlich et al. [140] developed a microdevice which couples a topographically-patterned substrate with a microfluidic chamber to control both topographic and FSS

Please cite this article in press as: W. Zheng, X. Jiang, Precise manipulation of cell behaviors on surfaces for construction of tissue/organs, Colloids Surf. B: Biointerfaces (2014), http://dx.doi.org/10.1016/j.colsurfb.2014.08.026

G Model COLSUB-6591; No. of Pages 14 12

ARTICLE IN PRESS W. Zheng, X. Jiang / Colloids and Surfaces B: Biointerfaces xxx (2014) xxx–xxx

cues to human renal proximal tubule cells. The surface topography and FSS work in concert to influence cell functions. They also [141] constructed a model of renal reabsorptive barriers with topographically-patterned porous membranes. The topographic cues in the model resemble physiological cues found in vivo while the porous structure allows quantification of transport across the cell layer. Liver is one of the most important organs to study for its toxicological and pharmaceutical function. The orderly organization of hepatocytes is necessary to act as a functional unit and retain their phenotype. Wong et al. [143] developed a size-controllable spheroidal hepatosphere and heterosphere model. Heterospheres had higher albumin secretion and enzymatic activity as compared with hepatospheres. On the similar model, they [144] investigated the interaction of hepatocytes and hepatic stellate cells (HSCs). The continuous flow and HSCs assists the formation and long-term maintenance of spheroids. Current in vitro platforms for testing the liver metabolism suffers from the lack of resemblance to the human liver metabolism. To address this, Domansky et al. [145] developed a bioreactor that fosters maintenance of 3D tissue cultures under constant perfusion. Lee et al. [147] reported a microfluidic device with liver microsome encapsulated in 3-D poly(ethylene glycol) diacrylate (PEG-DA) hydrogel matrix, which can mimic the metabolism reaction and the transport phenomena in the liver. This platform can serve as a tool for screening the liver metabolism of various compounds. The blood brain barrier (BBB), a unique selective barrier for the central nervous system (CNS), hinders the passage of most compounds to the CNS, complicating drug development. Innovative in vitro models of the BBB can provide useful insights into its role in CNS disease progression and drug delivery. Booth et al. [149] developed a microfluidic blood brain barrier (mBBB) to mimic the in vivo BBB. The mBBB system is superior to Transwell and could be used to study BBB function and drug delivery. Prabhakarpandian and coworkers [150] developed a microfluidics based synthetic microvasculature model of the BBB (SyM-BBB). In the model, astrocyte conditioned media led to the upregulation of tight junction of endothelial cells.

7. Conclusion In this review, we introduced the manipulation of adhesion, migration, and differentiation of cells on surface, focusing on the dynamic control of cell adhesion using switchable surfaces. We also discussed recent advances in microfluidic cell–cell interactions models concerning wound healing and tumor metastasis. The integration of multiple cell types, ECM, and various physical and chemical cues on a chip allow us to address basic biological questions and explore novel approaches for manipulating cells, providing platforms for setting up tissue or organ level in vitro models for physiological and pathological research. The combination of surface functioning and soft lithographic techniques has already seen success and holds considerable promise for cell biology research. Despite the advances mentioned above, we still face many problems, such as the less robust performance of functionalized surfaces in complex microenvironments and still limited approach in precise imaging dynamic behavior of cells, all of which will be addressed in future studies. A further key goal is to construct highly defined, complex in vitro architectures that can comprehensively mimic organ-level structure and function of tissues in vivo. Achieving this goal will facilitate not only the progress in the field of fundamental cell biology research, but also the development of advanced materials for tissue engineering and replacement of animal tests in drug screening.

Acknowledgements The financial support was provided by the Ministry of Science and Technology of China (2009CB93001 and 2007CB714502), National Science Foundation of China (31170905, 50902025, 20890020 and 90813032), the Chinese Academy of Sciences, and the Beijing Natural Science Foundation. References [1] Y. Xia, G.M. Whitesides, Angew. Chem. Int. Ed. 37 (1998) 550. [2] G.M. Whitesides, E. Ostuni, S. Takayama, X. Jiang, D.E. Ingber, Annu. Rev. Biomed. Eng. 3 (2001) 335. [3] E. Ostuni, L. Yan, G.M. Whitesides, Colloids Surf. B 15 (1999) 3. [4] K.L. Prime, G.M. Whitesides, Science 252 (1991) 1164. [5] C.S. Chen, M. Mrksich, S. Huang, G.M. Whitesides, D.E. Ingber, Science 276 (1997) 1425. [6] X.Y. Jiang, R. Ferrigno, M. Mrksich, G.M. Whitesides, J. Am. Chem. Soc. 125 (2003) 2366. [7] X.Y. Jiang, D.A. Bruzewicz, A.P. Wong, M. Piel, G.M. Whitesides, Proc. Natl. Acad. Sci. U. S. A. 102 (2005) 975. [8] K. Kushiro, S. Chang, A.R. Asthagiri, Adv. Mater. 22 (2010) 4516. [9] S.H. Yoon, J.Y. Chang, L.W. Lin, M.R.K. Mofrad, Lab Chip 11 (2011) 3555. [10] C.Y. Fan, Y.C. Tung, S. Takayama, E. Meyhofer, K. Kurabayashi, Adv. Mater. 20 (2008) 1418. [11] J.H. Hyun, H.W. Ma, Z.P. Zhang, T.P. Beebe, A. Chilkoti, Adv. Mater. 15 (2003) 576. [12] A. Khademhosseini, S. Jon, K.Y. Suh, T.N.T. Tran, G. Eng, J. Yeh, et al., Adv. Mater. 15 (2003) 1995. [13] B. Cao, R. Peng, Z.H. Li, J.D. Ding, Biomaterials 35 (2014) 6871. [14] M.L. McCain, A. Agarwal, H.W. Nesmith, A.P. Nesmith, K.K. Parker, Biomaterials 35 (2014) 5462. [15] Y. Li, B. Yuan, H. Ji, D. Han, S.Q. Chen, F. Tian, et al., Angew. Chem. Int. Ed. 46 (2007) 1094. [16] Z.L. Chen, Y. Li, W.W. Liu, D.Z. Zhang, Y.Y. Zhao, B. Yuan, et al., Angew. Chem. Int. Ed. 48 (2009) 8303. [17] K. Sun, L.S. Song, Y.Y. Xie, D.B. Liu, D. Wang, Z. Wang, et al., Langmuir 27 (2011) 5709. [18] Z.L. Chen, Y. Dai, Z. Dong, M.H. Li, X. Mu, R. Zhang, et al., Integr. Biol. 4 (2012) 1090. [19] M. Mrksich, G.M. Whitesides, Annu. Rev. Bioph. Biom. 25 (1996) 55. [20] J. Robertus, W.R. Browne, B.L. Feringa, Chem. Soc. Rev. 39 (2010) 354. [21] H. Kuroki, I. Tokarev, S. Minko, Annu. Rev. Mater. Res. 42 (2012) 343. [22] C. Zhao, I. Witte, G. Wittstock, Angew. Chem. Int. Ed. 45 (2006) 5469. [23] W.S. Yeo, M.N. Yousaf, M. Mrksich, J. Am. Chem. Soc. 125 (2003) 14994. [24] C.C.A. Ng, A. Magenau, S.H. Ngalim, S. Ciampi, M. Chockalingham, J.B. Harper, et al., Angew. Chem. Int. Ed. 51 (2012) 7706. [25] M. Irie, Adv. Polym. Sci. 94 (1990) 27. [26] M. Irie, Pure Appl. Chem. 62 (1990) 1495. [27] J. Auernheimer, C. Dahmen, U. Hersel, A. Bausch, H. Kessler, J. Am. Chem. Soc. 127 (2005) 16107. [28] D.B. Liu, Y.Y. Xie, H.W. Shao, X.Y. Jiang, Angew. Chem. Int. Ed. 48 (2009) 4406. [29] M.J. Salierno, A.J. García, A.d. Campo, Adv. Funct. Mater. 23 (2013) 5974. [30] A. Goulet-Hanssens, K. Lai Wing Sun, T.E. Kennedy, C.J. Barrett, Biomacromolecules 13 (2012) 2958. [31] B. Byambaa, T. Konno, K. Ishihara, Colloids Surf. B 103 (2013) 489. [32] B. Byambaa, T. Konno, K. Ishihara, Colloids Surf. B 99 (2012) 1. [33] R.M.P. Da Silva, J.F. Mano, R.L. Reis, Trends Biotechnol. 25 (2007) 577. [34] M. Yamato, C. Konno, M. Utsumi, A. Kikuchi, T. Okano, Biomaterials 23 (2002) 561. [35] E. Wischerhoff, N. Badi, A. Laschewsky, J.F. Lutz, Adv. Polym. Sci. 240 (2011) 1. [36] A. Mizutani, A. Kikuchi, M. Yamato, H. Kanazawa, T. Okano, Biomaterials 29 (2008) 2073. [37] J. Yang, M. Yamato, T. Shimizu, H. Sekine, K. Ohashi, M. Kanzaki, et al., Biomaterials 28 (2007) 5033. [38] M. Hirose, O.H. Kwon, M. Yamato, A. Kikuchi, T. Okano, Biomacromolecules 1 (2000) 377. [39] K. Fukumori, Y. Akiyama, M. Yamato, J. Kobayashi, K. Sakai, T. Okano, Acta Biomater. 5 (2009) 470. [40] L.H. Li, J.D. Wu, C.Y. Gao, Colloids Surf. B 85 (2011) 12. [41] H. Liu, X. Liu, J. Meng, P. Zhang, G. Yang, B. Su, et al., Adv. Mater. 25 (2013) 922. [42] M. Zhu, G. Baffou, N. Meyerbroker, J. Polleux, ACS Nano 6 (2012) 7227. [43] P. Rorth, Annu. Rev. Cell Dev. Biol. 25 (2009) 407. [44] J.P. Kaiser, A. Reinmann, A. Bruinink, Biomaterials 27 (2006) 5230. [45] L.L. Han, Z.W. Mao, J.D. Wu, Y. Guo, T.C. Ren, C.Y. Gao, Colloids Surf. B 111 (2013) 1. [46] P. Uttayarat, M. Chen, M. Li, F.D. Allen, R.J. Composto, P.I. Lelkes, Am. J. Physiol. Heart Circ. Physiol. 294 (2008) H1027. [47] K. Kushiro, A.R. Asthagiri, Langmuir 28 (2012) 4357.

Please cite this article in press as: W. Zheng, X. Jiang, Precise manipulation of cell behaviors on surfaces for construction of tissue/organs, Colloids Surf. B: Biointerfaces (2014), http://dx.doi.org/10.1016/j.colsurfb.2014.08.026

G Model COLSUB-6591; No. of Pages 14

ARTICLE IN PRESS W. Zheng, X. Jiang / Colloids and Surfaces B: Biointerfaces xxx (2014) xxx–xxx

[48] M.R. Ng, A. Besser, G. Danuser, J.S. Brugge, J. Cell Biol. 199 (2012) 545. [49] B. Yuan, Y. Li, D. Wang, Y. Xie, Y. Liu, L. Cui, et al., Adv. Funct. Mater. 20 (2010) 3715. [50] W.W. Liu, W.F. Zheng, B. Yuan, X.Y. Jiang, Integr. Biol. 5 (2013) 617. [51] W.W. Liu, S.G. Xing, B. Yuan, W.F. Zheng, X.Y. Jiang, Integr. Biol. 5 (2013) 1244. [52] Y. Liu, Y. Sun, H. Yan, X. Liu, W. Zhang, Z. Wang, et al., Small 8 (2012) 676. [53] Y. Sun, Z. Huang, K. Yang, W. Liu, Y. Xie, B. Yuan, et al., PLOS ONE 6 (2011) e28156. [54] L. MacQueen, Y. Sun, C.A. Simmons, J. R. Soc. Interface 10 (2013) 20130179. [55] G.C. Reilly, A.J. Engler, J. Biomech. 43 (2010) 55. [56] D.A. Lee, M.M. Knight, J.J. Campbell, D.L. Bader, J. Cell. Biochem. 112 (2011) 1. [57] L.D. Markert, J. Lovmand, M. Foss, R.H. Lauridsen, M. Lovmand, E.M. Fuchtbauer, et al., Stem Cells Dev. 18 (2009) 1331. [58] D. Sasaki, T. Shimizu, S. Masuda, J. Kobayashi, K. Itoga, Y. Tsuda, et al., Biomaterials 30 (2009) 4384. [59] S. Namgung, K.Y. Baik, J. Park, S. Hong, ACS Nano 5 (2011) 7383. [60] E. Martinez, E. Engel, J.A. Planell, J. Samitier, Ann. Anat. 191 (2009) 126. [61] E. Engel, E. Martinez, C.A. Mills, M. Funes, J.A. Planell, J. Samitier, Ann. Anat. 191 (2009) 136. [62] K.A. Kilian, B. Bugarija, B.T. Lahn, M. Mrksich, Proc. Natl. Acad. Sci. U. S. A. 107 (2010) 4872. [63] G.T. Christopherson, H. Song, H.Q. Mao, Biomaterials 30 (2009) 556. [64] J.B. Recknor, D.S. Sakaguchi, S.K. Mallapragada, Biomaterials 27 (2006) 4098. [65] M.P. Prabhakaran, J.R. Venugopal, S. Ramakrishna, Biomaterials 30 (2009) 4996. [66] J.M. Dang, K.W. Leong, Adv. Mater. 19 (2007) 2775. [67] X. Hu, S.H. Park, E.S. Gil, X.X. Xia, A.S. Weiss, D.L. Kaplan, Biomaterials 32 (2011) 8979. [68] M.R. Lee, K.W. Kwon, H. Jung, H.N. Kim, K.Y. Suh, K. Kim, et al., Biomaterials 31 (2010) 4360. [69] L.A. Smith, X.H. Liu, J. Hu, P. Wang, P.X. Ma, Tiss. Eng. A 15 (2009) 1855. [70] R. McBeath, D.M. Pirone, C.M. Nelson, K. Bhadriraju, C.S. Chen, Dev. Cell 6 (2004) 483. [71] E.K.F. Yim, S.W. Pang, K.W. Leong, Exp. Cell Res. 313 (2007) 1820. [72] M.J. Dalby, N. Gadegaard, R. Tare, A. Andar, M.O. Riehle, P. Herzyk, et al., Nat. Mater. 6 (2007) 997. [73] F. Kantawong, K.E.V. Burgess, K. Jayawardena, A. Hart, R.J. Burchmore, N. Gadegaard, et al., Biomaterials 30 (2009) 4723. [74] M.J. Dalby, D. McCloy, M. Robertson, H. Agheli, D. Sutherland, S. Affrossman, et al., Biomaterials 27 (2006) 2980. [75] E.K. Yim, S.W. Pang, K.W. Leong, Exp. Cell Res. 313 (2007) 1820. [76] T. Sjostrom, M.J. Dalby, A. Hart, R. Tare, R.O.C. Oreffo, B. Su, Acta Biomater. 5 (2009) 1433. [77] R.J. McMurray, N. Gadegaard, P.M. Tsimbouri, K.V. Burgess, L.E. McNamara, R. Tare, et al., Nat. Mater. 10 (2011) 637. [78] A. Ehrlicher, J.H. Hartwig, Nat. Mater. 10 (2011) 12. [79] T. Lai, K.H. Chiam, Phys. Rev. E: Statist. Nonlin. Soft Matter Phys. 84 (2011) 061907. [80] C. Belizna, L. Loufrani, A. Ghali, A. Lahary, E. Primard, J.P. Louvel, et al., Stroke 43 (2012) 1129. [81] X. Peng, J. Huang, C. Xiong, J. Fang, J. Biomech. 45 (2012) 116. [82] O.V. Sazonova, K.L. Lee, B.C. Isenberg, C.B. Rich, M.A. Nugent, J.Y. Wong, Biophys. J. 101 (2011) 622. [83] A.J. Engler, S. Sen, H.L. Sweeney, D.E. Discher, Cell 126 (2006) 677. [84] D.E. Discher, P. Janmey, Y.L. Wang, Science 310 (2005) 1139. [85] P.Y. Wang, W.B. Tsai, N.H. Voelcker, Acta Biomater. 8 (2012) 519. [86] A.D. Bershadsky, N.Q. Balaban, B. Geiger, Annu. Rev. Cell Dev. Biol. 19 (2003) 677. [87] A.J. Engler, M.A. Griffin, S. Sen, C.G. Bonnemann, H.L. Sweeney, D.E. Discher, J. Cell Biol. 166 (2004) 877. [88] K. Saha, A.J. Keung, E.F. Irwin, Y. Li, L. Little, D.V. Schaffer, et al., Biophys. J. 95 (2008) 4426. [89] P.M. Gilbert, K.L. Havenstrite, K.E. Magnusson, A. Sacco, N.A. Leonardi, P. Kraft, et al., Science 329 (2010) 1078. [90] M. Murrell, R. Kamm, P. Matsudaira, PLOS ONE 6 (2011) e24283. [91] J. Lee, Y.L. Wang, F. Ren, T.P. Lele, Langmuir 26 (2010) 16672. [92] A. Hettler, S. Werner, S. Eick, S. Laufer, F. Weise, PLOS ONE 8 (2013) e82635. [93] M. Poujade, E. Grasland-Mongrain, A. Hertzog, J. Jouanneau, P. Chavrier, B. Ladoux, et al., Proc. Natl. Acad. Sci. U. S. A. 104 (2007) 15988. [94] C.R. Keese, J. Wegener, S.R. Walker, L. Giaever, Proc. Natl. Acad. Sci. U. S. A. 101 (2004) 1554. [95] A.G. Clark, A.L. Miller, E. Vaughan, H.Y.E. Yu, R. Penkert, W.M. Bement, Curr. Biol. 19 (2009) 1389. [96] A.D. van der Meer, K. Vermeul, A.A. Poot, J. Feijen, I. Vermes, Am. J. Physiol. Heart C 298 (2010) H719. [97] Y.Y. Xie, W. Zhang, L.M. Wang, K. Sun, Y. Sun, X.Y. Jiang, Lab Chip 11 (2011) 2819. [98] M. Zhang, H. Li, H. Ma, J. Qin, Wound Repair Regen. 21 (2013) 897. [99] D. Franco, F. Milde, M. Klingauf, F. Orsenigo, E. Dejana, D. Poulikakos, et al., Biomaterials 34 (2013) 1488. [100] Q. Zhao, S. Wang, Y. Xie, W. Zheng, Z. Wang, L. Xiao, et al., Adv Healthcare Mater. 1 (2012) 560. [101] H. Yamaguchi, Y. Takeo, S. Yoshida, Z. Kouchi, Y. Nakamura, K. Fukami, Cancer Res. 69 (2009) 8594.

13

[102] M. Sidani, J. Wyckoff, C. Xue, J.E. Segall, J.J. Mammary, Gland Biol. Neoplasia 11 (2006) 151. [103] S.S. Wang, E.C. Li, Y.H. Gao, Y. Wang, Z. Guo, J.R. He, et al., PLOS ONE 8 (2013) e56448. [104] I.K. Zervantonakis, S.K. Hughes-Alford, J.L. Charest, J.S. Condeelis, F.B. Gertler, R.D. Kamm, Proc. Natl. Acad. Sci. U. S. A. 109 (2012) 13515. [105] J.S. Jeon, I.K. Zervantonakis, S. Chung, R.D. Kamm, J.L. Charest, PLOS ONE 8 (2013) e56910. [106] J.W. Song, S.P. Cavnar, A.C. Walker, K.E. Luker, M. Gupta, Y.C. Tung, et al., PLOS ONE 4 (2009) e5756. [107] C.Y. Chan, P.H. Huang, F. Guo, X.Y. Ding, V. Kapur, J.D. Mai, et al., Lab Chip 13 (2013) 4697. [108] B. Jiang, W. Zheng, W. Zhang, X. Jiang, Sci. Chin. Chem. 56 (2013) 1. [109] J.H. Sung, M.B. Esch, J.M. Prot, C.J. Long, A. Smith, J.J. Hickman, et al., Lab Chip 13 (2013) 1201. [110] W.F. Zheng, W. Zhang, X.Y. Jiang, Adv. Healthcare Mater. 2 (2013) 95. [111] D. Huh, G.A. Hamilton, D.E. Ingber, Trends Cell Biol. 21 (2011) 745. [112] J.S. Choi, Y. Piao, T.S. Seo, Biomaterials 35 (2014) 63. [113] B. Yuan, Y. Jin, Y. Sun, D. Wang, J.S. Sun, Z. Wang, et al., Adv. Mater. 24 (2012) 890. [114] P.Y. Gong, W.F. Zheng, Z. Huang, W. Zhang, D. Xiao, X.Y. Jiang, Adv. Funct. Mater. 23 (2013) 42. [115] K.H.K. Wong, J.M. Chan, R.D. Kamm, J. Tien, Annu. Rev. Biomed. Eng. 14 (2012) 205. [116] X.F. Ye, L. Lu, M.E. Kolewe, H. Park, B.L. Larson, E.S. Kim, et al., Biomaterials 34 (2013) 10007. [117] J.S. Miller, K.R. Stevens, M.T. Yang, B.M. Baker, D.H.T. Nguyen, D.M. Cohen, et al., Nat. Mater. 11 (2012) 768. [118] C. Franco, H. Gerhardt, Nature 488 (2012) 465. [119] S.S. Verbridge, A. Chakrabarti, P. DelNero, B. Kwee, J.D. Varner, A.D. Stroock, et al., J. Biomed. Mater. Res. A 101 (2013) 2948. [120] D.H.T. Nguyen, S.C. Stapleton, M.T. Yang, S.S. Cha, C.K. Choi, P.A. Galie, et al., Proc. Natl. Acad. Sci. U. S. A. 110 (2013) 6712. [121] Y. Zheng, J.M. Chen, M. Craven, N.W. Choi, S. Totorica, A. Diaz-Santana, et al., Proc. Natl. Acad. Sci. U. S. A. 109 (2012) 9342. [122] L.L.Y. Chiu, M. Montgomery, Y. Liang, H.J. Liu, M. Radisic, Proc. Natl. Acad. Sci. U. S. A. 109 (2012) E3414. [123] R. Fishler, M.K. Mulligan, J. Sznitman, J. Biomech. 46 (2013) 2817. [124] D. Huh, B.D. Matthews, A. Mammoto, M. Montoya-Zavala, H.Y. Hsin, D.E. Ingber, Science 328 (2010) 1662. [125] C.H.M.P. Willems, L.J.I. Zimmermann, P.J.L.T. Sanders, M. Wagendorp, N. Kloosterboer, J.W.C. Tervaert, et al., Exp. Cell Res. 319 (2013) 64. [126] C. Long, C. Finch, M. Esch, W. Anderson, M. Shuler, J. Hickman, Ann. Biomed. Eng. 40 (2012) 1255. [127] T. Kniazeva, A.A. Epshteyn, J.C. Hsiao, E.S. Kim, V.B. Kolachalama, J.L. Charest, et al., Lab Chip 12 (2012) 1686. [128] N.J. Douville, P. Zamankhan, Y.C. Tung, R. Li, B.L. Vaughan, C.F. Tai, et al., Lab Chip 11 (2011) 609. [129] A. Steimer, E. Haltner, C.M. Lehr, J. Aerosol Med. 18 (2005) 137. [130] A. Grosberg, P.W. Alford, M.L. McCain, K.K. Parker, Lab Chip 11 (2011) 4165. [131] M.L. McCain, S.P. Sheehy, A. Grosberg, J.A. Goss, K.K. Parker, Proc. Natl. Acad. Sci. U. S. A. 110 (2013) 9770. [132] C.S. Simmons, B.C. Petzold, B.L. Pruitt, Lab Chip 12 (2012) 3235. [133] H. Chen, J. Cornwell, H. Zhang, T. Lim, R. Resurreccion, T. Port, et al., Lab Chip 13 (2013) 2999. [134] A. Chen, E. Lee, R. Tu, K. Santiago, A. Grosberg, C. Fowlkes, et al., Biomaterials 35 (2014) 675. [135] Z. Ma, S. Koo, M.A. Finnegan, P. Loskill, N. Huebsch, N.C. Marks, et al., Biomaterials 35 (2014) 1367. [136] L. Ren, W.M. Liu, Y.L. Wang, J.C. Wang, Q. Tu, J. Xu, et al., Anal. Chem. 85 (2013) 235. [137] V. Chan, R. Raman, C. Cvetkovic, R. Bashir, ACS Nano 7 (2013) 1830. [138] K.J. Jang, A.P. Mehr, G.A. Hamilton, L.A. McPartlin, S.Y. Chung, K.Y. Suh, et al., Integr. Biol. 5 (2013) 1119. [139] K.J. Jang, K.Y. Suh, Lab Chip 10 (2010) 36. [140] E.M. Frohlich, X. Zhang, J.L. Charest, Integr. Biol. 4 (2012) 75. [141] E.M. Frohlich, J.L. Alonso, J.T. Borenstein, X. Zhang, M.A. Arnaout, J.L. Charest, Lab Chip 13 (2013) 2311. [142] X. Mu, W.F. Zheng, L. Xiao, W. Zhang, X.Y. Jiang, Lab Chip 13 (2013) 1612. [143] S.F. Wong, D.Y. No, Y.Y. Choi, D.S. Kim, B.G. Chung, S.H. Lee, Biomaterials 32 (2011) 8087. [144] S.A. Lee, D.Y. No, E. Kang, J. Ju, D.S. Kim, S.H. Lee, Lab Chip 13 (2013) 3529. [145] K. Domansky, W. Inman, J. Serdy, A. Dash, M.H.M. Lim, L.G. Griffith, Lab Chip 10 (2010) 51. [146] I. Wagner, E.M. Materne, S. Brincker, U. Sussbier, C. Fradrich, M. Busek, et al., Lab Chip 13 (2013) 3538. [147] J. Lee, S.H. Kim, Y.C. Kim, I. Choi, J.H. Sung, Enzyme Microb. Technol. 53 (2013) 159. [148] J.M. Prot, A.S. Briffaut, F. Letourneur, P. Chafey, F. Merlier, Y. Grandvalet, et al., PLOS ONE 6 (2011) e21268. [149] R. Booth, H. Kim, Lab Chip 12 (2012) 1784. [150] B. Prabhakarpandian, M.C. Shen, J.B. Nichols, I.R. Mills, M. SidorykWegrzynowicz, M. Aschner, et al., Lab Chip 13 (2013) 1093. [151] H.J. Kim, D. Huh, G. Hamilton, D.E. Ingber, Lab Chip 12 (2012) 2165. [152] W. Zheng, B. Jiang, D. Wang, W. Zhang, Z. Wang, X. Jiang, Lab Chip 12 (2012) 3441.

Please cite this article in press as: W. Zheng, X. Jiang, Precise manipulation of cell behaviors on surfaces for construction of tissue/organs, Colloids Surf. B: Biointerfaces (2014), http://dx.doi.org/10.1016/j.colsurfb.2014.08.026

G Model COLSUB-6591; No. of Pages 14 14

ARTICLE IN PRESS W. Zheng, X. Jiang / Colloids and Surfaces B: Biointerfaces xxx (2014) xxx–xxx

[153] B.C. Isenberg, C. Williams, R.T. Tranquillo, Circ. Res. 98 (2006) 25. [154] A.A. Lee, D.A. Graham, S. Dela Cruz, A. Ratcliffe, W.J. Karlon, J. Biomech. 124 (2002) 37. [155] W. Zheng, Y. Xie, W. Zhang, D. Wang, W. Ma, Z. Wang, et al., Integr. Biol. 4 (2012) 1102.

[156] T. Kniazeva, J.C. Hsiao, J.L. Charest, J.T. Borenstein, Biomed. Microdev. 13 (2011) 315. [157] E. Ghafar-Zadeh, J.R. Waldeisen, L.P. Lee, Lab Chip 11 (2011) 3031. [158] E. Serena, E. Cimetta, S. Zatti, T. Zaglia, M. Zagallo, G. Keller, et al., PLOS ONE 7 (2012) e48483.

Please cite this article in press as: W. Zheng, X. Jiang, Precise manipulation of cell behaviors on surfaces for construction of tissue/organs, Colloids Surf. B: Biointerfaces (2014), http://dx.doi.org/10.1016/j.colsurfb.2014.08.026