Keywords

These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

Introduction

Key opportunities for the field of stem cell engineering involve identification of cues that regulate stem cell fate, constructing a systems level understanding of how cells sense and process information provided by the microenvironment, and designing environments to elicit the desired cell fate. Meeting these opportunities will be facilitated by collaborative, interdisciplinary interactions among engineers, scientists, and clinicians. Chapters “Physical and Engineering Principles in Stem Cell Research” and “Computational Modeling and Stem Cell Engineering” in this report address the principles by which physical cues can affect stem cells and how mathematical modeling can provide insight into mechanisms of stem cell regulation. For these efforts to be successful, spatial and dynamic control over the microenvironment is needed. This chapter will focus on how recent advances in cell culture platform design and manufacture permit systematic application of regulatory cues to stem cells, and the insight these systems have provided in stem cell biology and engineering.

Stem Cell Microenvironment

The stem cell microenvironment is defined as the set of cues provided to the cell in vitro or in vivo. The microenvironment is an expansion of the stem cell niche concept, the physiologic environment that regulates the quiescence and maintenance of stem cells under normal conditions and the mobilization of stem cells for tissue generation in response to damage or disease (Scadden 2006). Chemical and mechanical cues presented in the microenvironment include short and long-range soluble factors, extracellular matrix, cell-cell contact, and mechanical forces (Fig. 1). These cues are sensed by receptors on or in the stem cell, and initiate chemical signaling cascades that affect gene transcription and protein translation, thereby governing a stem cell’s status. Signals that guide stem cell fates are not only encoded in the composition of the microenvironment, but also the spatial organization and temporal presentation of the cues.

Fig. 1
figure 1

Schematic of the stem cell microenvironment. Soluble factors, extracellular matrix components, intercellular interactions, and biomechanical cues synergize to regulate cell fate (From Metallo et al. 2007)

Full size image

Stem cell microenvironment engineering often attempts to recreate the native stem cell niche in vitro to maintain stem cell potency and facilitate stem cell expansion and differentiation. Alternatively, design of in vitro microenvironments permits systematic analysis of how a particular cue or set of cues regulates stem cell fate. Libraries of microenvironments also can be used as tools to screen for conditions that produce desired fates.

High-Throughput Screening

High-throughput screening (HTS) is a method for discovery of active compounds that elicit a desired biological response. HTS is often used to generate lead compounds for drug development and can also be used to identify unknown biochemical interactions. Often, robotic liquid handling systems apply large chemical libraries to biological samples in microwell plates and automated analytic techniques collect data on the biological effects of the compounds. Hits identified in the primary screen are typically validated in a secondary screen and the mechanism of action of confirmed hits can be assessed by standard biochemical techniques.

High-throughput screens are a powerful approach to identify chemical compounds that regulate a cellular phenotype, and require no prior knowledge about mechanisms that regulate the phenotype. Appropriate design of the screening process is critical for achieving a successful outcome. The phenotypic readout should be rapid, inexpensive, and ideally quantitative. In addition to direct phenotypic assessment, reporter cell lines or fluorescent probes are commonly used to analyze molecular changes in cells subjected to HTS.

Application of HTS approaches to stem cells have identified small molecule compounds that enhance stem cell survival, facilitate stem cell self-renewal, direct stem cell differentiation, and improve the efficiency of stem cell reprogramming. Immobilized arrays of matrix proteins, peptides, and synthetic materials have been used to identify defined substrates for stem cell culture. Combinatorial screening platforms have uncovered synergies between different cues and provided insight into how pathways combine environmental signals to regulate stem cell fate. Examples of these advances will be discussed in this chapter.

Microfluidics

Microfluidics involves precise manipulation of small volumes of fluids. Initial applications of microfluidics were primarily in the fields of chemistry and physics, but recent integration of microfluidics with biological cells has produced new insight into microenvironmental regulation of cellular behavior. At the microscale surface tension and fluid energy dissipation are low. Reynolds numbers are often very small so that molecular transport at fluid interfaces is dominated by diffusion, leading to low mixing rates and establishment of predictable molecular gradients.

Microfluidic devices have facilitated analysis of the dynamic response of the stem cell to the chemical and mechanical microenvironment by enabling precise application of cues to stem cells. A recent review by Young and Beebe discusses concepts related to microfluidic control of the cellular microenvironment including state-of-the-art technologies and remaining challenges (Young and Beebe 2010). Stem cell research studies using microfluidic devices have provided insight into mechanisms of autocrine and paracrine signaling and mechanotransduction in different types of stem cells. The ability to culture and analyze individual cells in parallel in microfluidic devices has allowed researchers to assess stem cell population heterogeneity that exists over the stem cell donor source, culture conditions, passage number, and other environmental selections applied to the cells. Microfluidic devices have also been used to separate low abundance stem cells from larger cell populations.

Stem Cell Biosensors

A biosensor is a tool that integrates a biological sensor with an analytic method to detect and report upon components in the microenvironment. In contrast to molecular biosensors, cell-based biosensors can provide a direct readout of the effects of the environment on cell phenotype. Cells used in biosensors are often engineered to have desired properties, including a real-time readout such as expression of a fluorescent or enzymatic reporter, or sensitivity and specificity of detection. Stem cells are a promising source of cells for whole cell biosensors because of their ability to generate a large number of normal cells from a single clonal source. Induced pluripotent stem cells derived from patients can also produce cells that possess disease phenotypes for integration into biosensors. Many cell types important in sensing applications (e.g., neural, cardiac, hepatic) are difficult to obtain and expand from primary sources, and animal models and cell lines fail to adequately recapitulate human responses. Thus, stem cell-based biosensors have the potential to revolutionize toxicity testing, drug discovery and evaluation, environmental monitoring, and clinical sample analysis.

High-Throughput Screening of Factors RegulatingStem Cell Fates

One strategy to regulate stem cell fates involves construction of in vitro microenvironments inspired by the in vivo niche. However, this rational design strategy is limited by knowledge of developmental biology and physiology, a relatively poor understanding of how different cues synergize to control signaling and cell fate, and the possibility that developmental pathways other than those present in vivo may regulate cell fate in vitro. HTS platforms offer the potential to identify factors that regulate stem cell fates, to improve upon a lead compound, to optimize presentation of cues, and to identify combinatorial interactions of known regulatory factors. Recent advances in spatially patterning cells, development of compound and materials libraries, and screening methodologies have enhanced our understanding of factors that control stem cell fates, leading to improved stem cell culture and differentiation systems.

Screening Chemical Libraries

Standard HTS methodologies, involving the application of large natural products or combinatorial chemical libraries to stem cells cultured in microwell plates, have been used to identify compounds that regulate stem cell self-renewal and differentiation. In one example, Alves et al. at the University of Twente screened a 1,280 compound library of compounds with known pharmacological activity for the ability to increase osteogenesis in human mesenchymal stem cells (hMSCs) (Alves et al. 2011). High-throughput fluorimetric assays of alkaline phosphatase and acidic phosphatase were used to assess osteogenesis and cell proliferation, respectively (Fig. 2). This study identified novel lead compounds, which would have been difficult to predict based on their mechanism of action, that induced osteogenesis with greater activity than previously known chemical compounds. Another study used 384-well plate screening of a 2,880 compound library to identify novel compounds that promote short-term hESC self-renewal and direct differentiation toward specific lineages (Desbordes et al. 2008). Other high-throughput screens have identified small molecule compounds and siRNAs that promote the survival and differentiation of pluripotent stem cells and molecules involved in regulating reprogramming of somatic cells to a pluripotent state (Andrews et al. 2010; Barbaric et al. 2010; Outten et al. 2011; Xu et al. 2010).

Fig. 2
figure 2

Schematic of a high-throughput screen of osteogenic enhancers in human mesenchymal stem cells (hMSCs) (From Alves et al. 2011). hMSCs were plated into 96 well plates in osteogenic medium. After 4 days, alkaline phosphatase (ALP) and acid phosphatase (ACP) activity were measured in a fluorescent plate reader to assess osteogenesis and proliferation, respectively. Positive compounds were confirmed by flow cytometry and subjected to further evaluation

Full size image

Biomaterials Arrays

The standard chemical screening platforms developed and refined by the pharmaceutical industry work well for identifying soluble chemical factors that regulate stem cell fate, but are not effective at identifying other components of the microenvironment, including matrices and scaffolds, intercellular interactions, and mechanical forces. The development of biomaterial arrays has enabled the identification of how the chemical and mechanical properties of polymers and other materials affect stem cell fates. For example, Mei and colleagues at MIT screened a combinatorial polymer library for the ability of the compounds to provide a substrate that supports human embryonic stem cell (hESC) self-renewal (Mei et al. 2010). The investigators related the biological effects of the polymers to properties including elastic modulus, roughness, hydrophobicity, and the composition of surface functional groups. Another recent study investigated the effects of material topography on hMSC proliferation and osteogenesis (Unadkat et al. 2011). Poly(lactic acid) substrates with identical chemical compositions were printed in arrays with 2,176 random surface topographies. Machine learning algorithms related material topography to stem cell fate and identified features that instruct MSC proliferation and differentiation. Such efforts may prove valuable in designing materials that provide defined matrices for stem cell culture.

Patterned arrays of biomaterials have also been used to identify peptides that support hESC self-renewal when immobilized to a substrate. Laura Kiessling’s lab at the University of Wisconsin–Madison used phage display to identify a library of peptides that bind to embryonic carcinoma cells, then created photopatterned arrays of these cells using self-assembled monolayers of alkanethiols on a gold surface (Derda et al. 2010). This array was screened for the ability to maintain self-renewal in hESCs, and two peptides that support hESC culture in defined conditions were identified.

Combinatorial arrays coupled with high-throughput analysis can be used to identify synergistic and antagonistic interactions between factors that regulate signaling pathways controlling stem cell development. For example, an array that contained mixtures of extracellular matrix proteins and growth factors known to be involved in neural development was screened for ability induce differentiation in human neural progenitors (Soen et al. 2006; Fig. 3). Dose and time-dependent analysis of cell differentiation revealed novel insights into interactions between pathways regulating cell fate. In another study, Bhatia and colleagues constructed a microwell-based system to assess the combinatorial interactions of growth factors and extracellular matrix proteins on embryonic stem cell fates, including differentiation to cardiac lineages (Flaim et al. 2008). A microwell chip was developed by Rosenthal and others to provide defined microenvironments consisting of soluble factors, matrices, and cell-cell contact to single cells or small clusters of stem cells (Rosenthal et al. 2007). This culture system was used to determine that intercellular interactions repress mESC colony initiation.

Fig. 3
figure 3

Schematic of combinatorial screening of extracellular matrix (ECM) and growth factor (GF) effects on stem cells (From Soen et al. 2006). Arrays of pre-mixed combinations of proteins were printed using a noncontact piezoelectric arrayer. Human neural progenitor cells were seeded onto the arrays and cultured under differentiation-promoting conditions for 3 days. Proliferation and differentiation responses were analyzed by immunostaining for Tuj1, GFAP, and BrdU

Full size image

Array-based screening methods have the potential to elucidate how intercellular interactions regulate stem cell fate, including multicellular tissue organization. Patterns of identical composition, but different sizes and shapes, have been constructed to determine that mechanical interactions between cells affect multicellular organization of MSCs, which in turn regulates osteogenic and adipogenic fates (Ruiz and Chen 2008).

Arrays for Clonal Analysis of Stem Cells

Arrays of materials can be used to screen for how chemical and physical properties of the substrate regulates stem cell fate. However, heterogeneity in stem cell populations can lead to stem cell clones responding to the same environment in different manners. Arrays consisting of identical features can be used to probe the extent of heterogeneity that exists in a stem cell population. Ashton and colleagues created micropatterned substrates that enabled formation of microspheres from neural stem cell clones and monitored proliferation and differentiation of neurospheres derived from individual clones (Ashton et al. 2007). The spatial registry in this system permits isolation and characterization of individual neurospheres. This platform, which allows the user to monitor then select clonal-derived neurospheres, is also useful for screening genetic libraries.

3D microwell arrays have been developed as another platform to analyze responses of stem cell clones to the microenvironment. Matthias Lutolf’s lab at École Polytechnique Fédérale de Lausanne (EPFL) constructed 3D poly(ethylene glycol) (PEG) microwell arrays that contained gradients in concentrations of two different proteins (Gobaa et al. 2011). These arrays were used to identify effects of MSC seeding density on proliferation and differentiation to adipocytes, regulation of MSC differentiation to osteocytes by the mechanical properties of the PEG matrix, and combinatorial relationships between factors that control neural stem cell expansion (Gobaa et al. 2011; Roccio et al. 2012).

Microfluidic Devices for Stem Cell Culture and Characterization

Microfluidic devices provide precise, dynamic control of the fluid environment surrounding a stem cell. Because of the small sample size required in microfluidics, these devices can be incorporated into high-throughput screens, assessing effects of spatial gradients and patterns of soluble cues and via dynamic changes in fluid composition. Microfluidic devices are amenable to automation, which can provide reproducible and precisely controlled pumping, mixing, temperature regulation, and integrated sensing.

Microfluidic devices enable the analysis of stem cell function at the microscale, including autocrine and paracrine signaling. Flow can be used to provide or remove factors in the environment of precisely-arranged stem cells to identify mechanisms of stem cell regulation. In addition, microfluidic devices have been applied to stem cell processing by enabling cell separations.

Microfluidics for Understanding Mechanisms of Stem Cell Regulation

The greatest impact of microfluidics to date on the field of stem cell engineering has been as a tool to elucidate mechanisms of stem cell fate regulation. For example, Peter Zandstra’s lab at the University of Toronto used perfusion flow in a microfluidic channel to regulate accumulation of cell-secreted paracrine factors in the cell environment (Moledina et al. 2012). Using a modeling approach that accounted for fluid flow and cell position in the microchannel, the investigators identified that paracrine signaling is able to stimulate murine embryonic stem cell (mESC) self-renewal in the absence of leukemia inhibitory factor (LIF). Ellison and colleagues also used an integrated computational modeling and microfluidic culture approach to identify the presence of distinct cell-secreted factors that regulate mESC viability in serum-containing and defined culture media (Ellison et al. 2009).

Microfluidic devices are also useful in studying the application of mechanical forces, including shear stress, to stem cells and their derivatives. Toh and Voldman (2011) fabricated a microfluidic device to apply a known shear stress to ESCs cultured on a substrate and found that shear stress negatively impacts ESC colony growth. Using inhibitors and proteases in conjunction with shear stress, the investigators found that ESCs sense shear stress through the extracellular matrix, and that shear affects cell fate via Fgf5 signaling. An optimization of medium perfusion rates in a microchannel containing hESCs found that, at low flow rates, nutrient depletion and waste accumulation impair cell expansion while high flow rates can detach cells (Titmarsh et al. 2011).

Because of their small size, microfluidic devices are particularly valuable for analyzing the behavior of single cells in response to a dynamic microenvironment, including clonal behavior of stem cells. In one such study, Lecault and colleagues at the University of British Columbia constructed arrays of nanoliter chambers connected by microchannels (Lecault et al. 2011). Pumps and valves regulated flow between the chambers, allowing programmed dynamic application of soluble factors to hematopoietic stem cell clones seeded into the chambers. When coupled with live cell image analysis, this platform was used to identify the precise time point at which Steel factor stimulation stimulated the exit of adult hematopoietic stem cells from quiescence. This culture system combines the dynamic advantages of microfluidics with the spatial segregation of microwells to provide the opportunity to perform HTS at the clonal scale. Such studies have the potential to provide insight into regulation of stem cells and stem cell population heterogeneity.

In another example, Gomez-Sjorberg and colleagues developed a high-throughput microfluidic screening platform for proliferation, differentiation, and motility of hMSCs (Gomez-Sjoberg et al. 2007). This platform uses a multiplexing scheme to create 96 distinct experimental conditions in which cells can be cultured for several weeks.

Spatial Patterning in Microfluidic Devices

The small length scales in microfluidic chambers typically result in laminar flow profiles, even at relatively high fluid velocities. By introducing different reagents to a microfluidic channel by spatially-segregated entrance ports, concentration gradients across the channel can be constructed. The nature of these gradients depends on convective flow through the channel and diffusion across the channel. Such devices can be used to identify how a population of stem cells responds to signaling gradients, which are important regulators of tissue development in vivo. In addition, these devices can present gradients to stem cells that elicit distinct cell fates and create spatially-patterned cell populations. Zhang and colleagues at Columbia University established a gradient of doxycycline in a microfluidic device containing MSCs engineered to express doxycycline-inducible bone morphogenetic protein 2 (BMP2) (Zhang et al. 2011). This created a gradient in BMP2 signaling which produced spatially-patterned osteogenesis.

Philippe Renaud’s lab at EPFL used a microfluidic device to construct gradients of neural growth factor (NGF) and B27 in a 3D hydrogel (Kunze et al. 2011). Cortical neurons were seeded in the hydrogels, and synapse formation observed in response to the NGF/B27 gradients. The investigators observed synergies in NGF and B27 in inducing synapses and demonstrated the ability to control synapse density using gradients of these neurotrophic factors. A similar approach could be used to regulate stem cell expansion, differentiation, or organization in a 3D microenvironment. Lii and others constructed a microfluidic chip that uses pneumatically actuated valves to perfuse reagents into 3D extracellular matrix gels containing mESCs (Lii et al. 2008). Channels above the gel permitted rapid flow and gel-embedded cells were protected from shear forces.

Microfluidic Separation and Characterization of Stem Cells

The precise spatial control of fluid flow in a microfluidic device offers the potential to improve stem cell separations. Low throughput of microfluidic devices may limit applications in larger scale bioprocessing operations, but microfluidic devices have the ability to isolate low abundance cells for subsequent expansion or characterization. Mehmet Toner and colleagues constructed a platform that can separate low abundance circulating tumor cells (CTCs) from peripheral blood samples based on cellular interactions with anti-body coated microposts in microfluidic channels (Nagrath et al. 2007; Stott et al. 2010). A microfluidic strategy to capture circulating endothelial progenitor cells from peripheral blood based on adhesion to antibody-coated substrate was also reported (Plouffe et al. 2009). Similar strategies to isolate other rare stem and progenitor cell populations from tissues may be promising if specific surface markers are available.

Stem cell properties other than surface affinity may also be the basis for microfluidic separations. Investigators at Lund University used forces generated by an acoustic standing wave to separate platelets from peripheral blood progenitor cells in apheresis products (Dykes et al. 2011). Microfluidic channels at the outlet of the device allowed the researchers to collect purified cell populations. A spiral microfluidic system that separates cells based on cell diameter was used to isolate bone marrow hMSC populations in different stages of the cell cycle (Lee et al. 2011).

Microfluidic capture of CTCs has been integrated with downstream culture, including clonal expansion as 3D spheroids by injection of a hydrogel matrix to encapsulate captured CTCs (Bichsel et al. 2012). Also, automated imaging platforms have been integrated with microfluidic culture to enable single cell-based characterization and high-throughput screening of stem cells based on marker or reporter expression (Kamei et al. 2009, 2010). Microfluidic flow can be used to deliver compounds to specific regions of a culture. For example, localized delivery of an enzyme can remove specific colonies for collection and analysis, and different cell stains can be provided to distinct regions of the culture channel (Villa-Diaz et al. 2009).

Stem Cell Biosensors

Whole Cell Biosensors

Stem cell biosensors have been constructed to report on the state of the stem cell itself as well as the microenvironment. The most common type of stem cell biosensor employs a genetic promoter-reporter construct to produce a fluorescent protein or an enzyme when the stem cell is in a particular differentiation state. For example, human embryonic stem cells expressing GFP under control of the OCT4 promoter can be noninvasively monitored for the loss of the pluripotency marker while somatic cells containing this construct will gain GFP expression following reprogramming to a pluripotent state (Gerrard et al. 2005; Huangfu et al. 2008). Numerous cell lines expressing GFP or other reporters under lineage-specific promoters have been constructed to study differentiation or purify desired cell populations. For example, NKX2-5 eGFP cells were used to purify cardiac-committed hESCs and identify specific surface markers on these cells (Elliott et al. 2011). Similarly, a nestin-eGFP reporter hESC line has been used to identify neural progenitors during differentiation (Noisa et al. 2010). Stem cells have also been engineered to express reporters when certain developmental signaling pathways, such as canonical Wnt signaling or Notch, have been activated (Davidson et al. 2012; Fre et al. 2011).

Biosensors can be constructed to assess cell differentiation state based on expression of surface markers. Surface plasmon resonance (SPR) has been used for real-time analysis of osteogenic differentiation in mesenchymal stem cells based on upregulation of OB-cadherin expression in cells cultured on an SPR substrate (Kuo et al. 2011). Stem cell biosensors have also been developed to probe the composition of microenvironments in vitro and in vivo. Fluorescently-labeled aptamers that bind platelet-derived growth factor (PDGF) have been coupled to the surface of mesenchymal stem cells and used to spatially map PDGF concentrations near the stem cell niche in vivo (Zhao et al. 2011). Other biosensors report on stem cell phenotypes. For example, Ali Khademhosseini’s lab developed a cardiotoxicity biosensor that integrated mESC-derived cardiomyocytes with an automated imaging system to monitor changes in cell contraction rate in response to pharmacologic agents (Kim et al. 2011).

Organ-on-a-Chip

One application of stem cell biosensors is to model tissue and organ level functions in an in vitro microdevice using stem cell-derived cells organized to provide the appropriate biological complexity. Real-time analysis can be facilitated by engineered reporter cell lines and integrating imaging, mechanical, electrical, and chemical probes into the system. These “organs-on-a-chip” can bridge the gap between studies on individual cells in culture and in vivo studies, and may supplement or replace animal models for physiological studies or drug evaluation trials.

One challenge in constructing stem cell models of tissue and organ level function is manufacturing relevant cells. An appropriate stem cell source must be chosen and efficient differentiation and purification processes must be available. Engineering function or appropriate sensing capacity into the stem cells can improve function of the construct. A physiologic microenvironment that enables survival, function, spatial arrangement, and intercellular interactions between multiple cell types must be constructed. These microenvironments often use microfluidics and biomaterials approaches to provide 3D, spatially-controlled, dynamic environments. Cell culture must be integrated with analysis, and real-time control systems often enhance construct performance.

One promising application of stem cell-based organ-on-a-chip technology is production of cardiac constructs. Efficient methods for differentiating and purifying cardiomyocytes from human pluripotent stem cells have been reported (Mummery et al. 2012). In fact, hPSC-derived cardiomyocytes are commercially available from several companies, including Cellular Dynamics International (United States) and Pluriomics (Netherlands). Contractile forces and electrophysiologic function of the stem cell-derived cardiomyocytes can be monitored in real time (Hazeltine et al. 2012; Poon et al. 2011; Fig. 4). These individual hPSC-derived cardiomyocytes lack the structural organization of heart muscle, however. A microgroove culture platform has been used to align mESC-derived cardiomyocytes into fibers that exhibit organized sarcomeres (Luna et al. 2011). More complex engineered heart tissues that comprise muscle strips or organoid chambers have been constructed from primary animal ventricular cardiomyocytes (Lee et al. 2008). Tissues of hPSC-derived cardiomyocytes, endothelial cells, and stromal cells, illustrated roles of endothelial cells on cardiomyocyte proliferation, interactions between stromal and endothelial cells, vascular formation, and effects of cyclic stresses on cardiomyocyte hypertrophy and proliferation (Tulloch et al. 2011). Together, these studies demonstrate the near-term feasibility of 3D, functional stem cell-derived cardiac tissues integrated with functional readouts.

Fig. 4
figure 4

Mapping of contraction stress in cardiomyocytes (Courtesy of Laurie Hazeltine). (Left) Phase contrast image of a contracting cardiomyocyte differentiated from a human pluripotent stem cell cultured on a flexible polyacrylamide hydrogel containing embedded fluorescent beads. The cell outline is shown in white. Scale bar = 20 μm. (Right) A contraction stress map of the cell shown in the phase contrast image calculated based on bead displacement during the contraction cycle

Full size image

Future Directions

Stem Cell Screening

While proof-of-concept examples have demonstrated the power of high-throughput screening technologies to identify microenvironmental cues that regulate stem cell fates, the potential of these technologies is limited by structural features of HTS platforms and stem cells. The screens are typically constrained by the throughput of stem cell analysis. Thus, to screen larger libraries and increase the odds of obtaining a hit, more reliable, faster, and less expensive methods to assess cell response to the compounds in the library must be developed. Dynamic, nondestructive approaches for monitoring cell state, such as enzymatic activity or fluorescent reporter concentration, will facilitate identifying the temporal regulation of stem cells by microenvironmental cues. In addition, screening relies on observing the behavior of a single stem cell or small number of stem cells, and population heterogeneity can result in a high frequency of false positives or false negatives. An understanding of the heterogeneous responses of stem cells to microenvironmental cues and strategies to account for these differences are needed to realize the potential of HTS in stem cell applications.

An increase in screening throughput will enable higher order combinatorial screens to identify how cells process multiple cues in making fate decisions. Because of the costs and technical challenges associated with large-scale screening, it is anticipated that academic or nonprofit collaborations with industry will be productive in addressing the stem cell throughput issue. For example, researchers at I-STEM in France have collaborated with Roche to screen the effects of approximately 200,000 compounds on neural stem cell proliferation. When combined with modeling approaches and statistical analysis, these screens can provide fundamental insight into stem cell regulatory networks. Other screening outcomes may include more efficient and better defined microenvironments for expansion and controlled differentiation of stem cells.

An opportunity exists to expand HTS platforms beyond chemical libraries. Recent studies have identified important roles of microRNAs and long noncoding RNAs in developmental biology, including stem cell proliferation and differentiation as well as cell reprogramming (Pauli et al. 2011; Tiscornia and Izpisua Belmonte 2010; Yi and Fuchs 2011). Screening RNA libraries may identify new mechanisms of stem cell regulation and identify tools to control developmental programs in stem cells. Engineering platforms to screen cues such as intercellular contacts or mechanical forces in the context of a physiologically relevant chemical microenvironment would improve the potential of stem cell screening platforms. 3D screening platforms, such as microwells or biomaterials scaffolds would enable identification of factors that regulate stem cell assembly and tissue development or morphogenesis from stem cell sources.

Microfluidic Culture of Stem Cells

Microfluidic platforms offer the ability to understand the response of single stem cells or small populations of cells to defined, dynamic microenvironments. However, access to microfluidic culture systems is generally limited to labs with the fabrication capacity to produce these devices, and uniform standards for microfluidic culture of stem cells do not yet exist. To realize the potential of these devices in stem cell engineering, reliable and inexpensive sources of relatively simple microfluidic devices must be available. In addition, collaborative efforts between researchers with expertise in device manufacture and stem cell researchers posing questions that can be addressed by these devices will be important. Similarly, the use of microfluidic culture systems has been limited to the research laboratory. The opportunity exists to translate these devices to commercial or clinical applications, including production of pure populations of stem cells, stem cell-based HTS platforms, and patterned multicellular tissues for in vitro analysis.

The ability to construct defined microenvironments in microfluidic devices has outpaced the ability to characterize cells in these microenvironments in real time. Better integration of stem cell culture and separations platforms with cell characterization is needed. With the advent of single-cell gene expression analysis, the opportunity exists to deeply probe clonal differences in stem cell populations. Some microfluidic devices have been designed to probe gene expression and signaling activity in individual cells (Bennett and Hasty 2009; Cheong et al. 2009; Yin et al. 2010).

The opportunity exists to integrate precise microenvironmental control and cell analysis with signaling pathway modeling to obtain a systems level understanding of stem cell behavior. In fact, multiscale modeling will be needed to understand how spatial and temporal presentation of cues at the microscale can lead to longer term changes in cell and tissue level phenomena. This understanding will then in turn enable design of microenvironments that elicit the desired cell outcomes via application of chemical and physical cues to activate appropriate developmental signaling pathways.

Stem Cell Biosensors

Efforts to construct stem cell biosensors have only begun to realize the potential of stem cells in sensing applications. Next generation reporters will simultaneously monitor the activities of multiple promoters or cell processes in real time. This will require identification of sensitive and specific promoter sequences, methods to engineer stem cells, and integration of tools to monitor cell phenotypes in stem cell culture and differentiation platforms. Synthetic biology tools, which have been most widely applied to microbial cells, have the potential to redesign stem cell regulatory circuits for sensing applications.

Advances in microscale fabrication technologies will enable construction of more physiologically relevant, 3D structured cell and tissue biosensors (Gauvin and Khademhosseini 2011). For example, 3D vascular structures have been formed from microfluidic devices, direct inkjet printing, and assembly of microgels (Du et al. 2011; Wu et al. 2011; Choi et al. 2007; Fidkowski et al. 2005). These approaches will enable development of more advanced organ-on-a-chip models.

Global Assessment and Conclusions

The United States is currently a leader in development of microenvironments and platforms for using stem cells in HTS and biosensing applications. This is primarily the result of a strong and growing community of engineers trained in biomaterials and microsystem fabrication expanding their research to stem cells and collaborating with stem cell biologists. Other countries with pronounced strengths in this area include Canada, Japan, the Netherlands, and Switzerland, which have also made concerted efforts to engage engineers and stem cell biologists in research collaborations.

One area of this field that the United States does not necessarily lead is integration government and academic research with commercial efforts. Partnerships between I-STEM and pharmaceutical companies in France appear to be productive in leveraging the pharmaceutical industry’s expertise in HTS with the stem cell proficiency at I-STEM. Models of commercialization in the Netherlands Institute of Regenerative Medicine and the Berlin-Brandenburg Center for Regenerative Therapies offer advantages in designing and distributing new tools that have value to the stem cell field.

It is apparent that engineered microenvironments and high-throughput methods will become more sophisticated. These approaches will provide more detailed insight into basic regulation of stem cell function, generate predictive in vitro tissue models, and contribute to the regenerative potential of stem cells. Efforts to engage multidisciplinary teams of stem cell engineers, biologists, and clinicians in academia, industry, and medicine will accelerate progress in this field.