ARTICLE Microfluidic Perfusion Culture of Human Induced Pluripotent Stem Cells Under Fully Defined Culture Conditions Ryosuke Yoshimitsu,1 Koji Hattori,2 Shinji Sugiura,2 Yuki Kondo,1 Rotaro Yamada,1 Saoko Tachikawa,1 Taku Satoh,2 Akira Kurisaki,2 Kiyoshi Ohnuma,1,3 Makoto Asashima,2 Toshiyuki Kanamori2 1

Department of Bioengineering, Nagaoka University of Technology, 1603-1 Kamitomioka, Nagaoka, Niigata 940-2188, Japan; telephone: þ81-258-47-9454; fax: þ81-258-47-9454; e-mail: [email protected] 2 Research Center for Stem Cell Engineering, National Institute of Advanced Industrial Science and Technology (AIST), Tsukuba, Ibaraki 305-8562, Japan 3 Top Runner Incubation Center for Academia-Industry Fusion, Nagaoka University of Technology, 1603-1 Kamitomioka, Nagaoka, Niigata 940-2188, Japan

ABSTRACT: Human induced pluripotent stem cells (hiPSCs) are a promising cell source for drug screening. For this application, self-renewal or differentiation of the cells is required, and undefined factors in the culture conditions are not desirable. Microfluidic perfusion culture allows the production of small volume cultures with precisely controlled microenvironments, and is applicable to high-throughput cellular environment screening. Here, we developed a microfluidic perfusion culture system for hiPSCs that uses a microchamber array chip under defined extracellular matrix (ECM) and culture medium conditions. By screening various ECMs we determined that fibronectin and laminin are appropriate for microfluidic devices made out of the most popular material, polydimethylsiloxane (PDMS). We found that the growth rate of hiPSCs under pressure-driven perfusion culture conditions was higher than under static culture conditions in the microchamber array. We applied our new system to self-renewal and differentiation cultures of hiPSCs, and immunocytochemical analysis showed that the state of the hiPSCs was successfully controlled. The effects of three antitumor drugs on hiPSCs were comparable between microchamber array and 96-well plates. We believe that our system will be a platform technology for future large-scale screening of fully defined conditions for differentiation cultures on integrated microfluidic devices. Biotechnol. Bioeng. 2014;111: 937–947. ß 2013 Wiley Periodicals, Inc.

The authors have no conflict of interest to declare. Ryosuke Yoshimitsu and Koji Hattori contributed equally to this work. Correspondence to: K. Ohnuma Contract grant sponsor: The Japan Science and Technology Agency Received 20 July 2013; Revision received 9 October 2013; Accepted 8 November 2013 Accepted manuscript online 13 November 2013; Article first published online 30 November 2013 in Wiley Online Library (http://onlinelibrary.wiley.com/doi/10.1002/bit.25150/abstract). DOI 10.1002/bit.25150

ß 2013 Wiley Periodicals, Inc.

KEYWORDS: human induced pluripotent stem cells; defined culture conditions; monolayer culture; microchamber array; perfusion culture; cell-based assay

Introduction Human pluripotent stem cells (hPSCs), including both human embryonic stem cells (hESCs) and human induced pluripotent stem cells (hiPSCs), are capable of self-renewal and pluripotency (Takahashi et al., 2007; Thomson et al., 1998; Yu et al., 2007). Because hiPSCs contain the donor’s genetic information, hiPSC-derived cells with diseasespecific phenotypes provide a new methodology for drug screening (Egawa et al., 2012; Itzhaki et al., 2011; Liu et al., 2011). For efficient and accurate drug screening applications, use of a culture system that accurately controls the proliferation and differentiation of hiPSCs is important. However, application of conventional drug screening systems to hiPSCs has two problems: the effects of undefined components in the culture medium or extracellular matrix (ECM), and the limitations of the types of culture vessels used, as outlined below. Conventional culture methods for hiPSCs use many undefined supplements including liquid additives and ECM, such as Knockout Serum-Replacement (KSR; Life technologies, Grand Island, NY,) (Price and Goldsborough, 1998; Takahashi et al., 2007; Thomson et al., 1998; Yu et al., 2007), and Matrigel (BD Biosciences, Mississauga, Canada; International Stem Cell Initiative Consortium, 2010; Vukicevic et al., 1992). Because these undefined factors contain a variety of growth factors, hormones, and integrin receptors,

Biotechnology and Bioengineering, Vol. 111, No. 5, May, 2014

937

which maintain hiPSCs in an undifferentiated state, they are not suitable for differentiation assays. Therefore, fully defined hiPSCs culture systems, which have defined supplements such as specific hormones and growth factors in the culture medium and a defined coating matrix such as fibronectin or laminin, have been extensively studied (Braam et al., 2008; Furue et al., 2008; Hayashi et al., 2010). By using such systems, the direct responses of added factors can be assessed without masking and distortion by undefined factors, enabling efficient differentiation and accurate drug screening. The use of conventional culture vessels for large-scale screening of culture conditions (Minami et al., 2012; Yang et al., 2013) could be cost prohibitive because of the expensive culture medium and reagents. Furthermore, screening of differentiation conditions in such vessels is intrinsically limited by poor spatiotemporal control of the microenvironment, because it is difficult to determine whether exogenous factors act directly or through paracrine-dependent mechanisms (Gupta et al., 2010; Toh et al., 2010; van Noort et al., 2009). To overcome these limitations, we and other research groups have applied a microfluidic device to high-throughput cellular microenvironment screening (El-Ali et al., 2006; Hattori et al., 2011; Titmarsh and Cooper-White, 2009). More importantly, control of the culture environment at the micro-scale allows the study of many physiologically relevant phenomena that cannot be investigated using conventional culture vessels (Gupta et al., 2010; Toh et al., 2010; van Noort et al., 2009). For example, microfluidic perfusion culture can uncover autocrine and paracrine processes, which are generally hidden in conventional static culture systems, and it allows investigation of the minimal amount of soluble factors required to drive cells toward a desired fate (Blagovic et al., 2011; Ellison et al., 2009; Moledina et al., 2012). Thus, the microfluidic system has many advantages as a drug screening system. Despite these advantages, the use of microfluidics in hiPSC studies is still limited (Kamei et al., 2009; Khoury et al., 2010; Kim et al., 2006; Titmarsh et al., 2011, 2012; Villa-Diaz et al., 2009). Furthermore, as far as we know, there has been no report of the use of microfluidics for hiPSC experiments under fully defined culture conditions. These limited culture conditions prompt unique challenges not encountered with other cell types. Here, we investigated the type of purified ECM suitable for growing hiPSCs in defined culture medium on polydimethylsiloxane (PDMS), which is the most popular material used for the fabrication of microfluidic devices (Deng et al., 2000; Duffy et al., 1998). We also investigated the effect of medium perfusion rate on the growth of hiPSCs in self-renewal culture. Finally, we demonstrated that fully defined culture conditions controlled by medium perfusion in a microchamber array chip enabled the culture of hiPSCs in self-renewal and differentiation conditions.

938

Biotechnology and Bioengineering, Vol. 111, No. 5, May, 2014

Materials and Methods Culture of hiPSCs The hiPSC 201B7 line (Takahashi et al., 2007) was obtained from RIKEN BRC Cell Bank (Tsukuba, Ibaraki, Japan) through the National Bio-Resource Project for the Ministry of Education, Culture, Sports, Science and Technology, Japan. The hiPSC Tic line (JCRB1331), which was derived from fetal lung fibroblasts (MRC-5), was obtained from the Japan Collection of Research Bioresources Cell Bank (Osaka, Japan; Nagata et al., 2009; Nishino et al., 2011). hiPSCs were maintained in a KSR-based medium on mouse embryonic fibroblast (MEF) feeder cells, and subcultured as described in Supplementary Methods (Hayashi et al., 2010). For all experiments, hiPSCs in KSR-based medium on MEFs were transferred to fully defined culture conditions based on hESF9a medium (Table I) on 2 mg/mL fibronectin-coated dishes, and passaged at least once before use (Hayashi et al., 2010). For the differentiation experiments, the culture medium was replaced with hESF-6 medium (Table I) supplemented with bone morphogenetic protein 4 (BMP4; R&D systems, Minneapolis, MN). Mitomycin C, mithramycin, and actinomycin D with D-mannitol were obtained from Wako Pure Chemical Industries, Ltd. (Osaka, Japan). Screening of ECM-coating Conditions on PDMS Surfaces and Drug Assays Twenty-four-well–plates made from PDMS (Sylgard 184, Dow Corning, Midland, MI) and PDMS-coated 6-well plates were coated with various ECM solutions (fibronectin [Sigma], laminin [Sigma], collagen [Nitta Gelatin Inc., Osaka, Japan], or gelatin [Sigma]) in phosphate buffered saline (PBS) for 1 h at 37 C. hiPSCs were plated on the ECMTable I. Composition of the defined culture mediums. hESF-9aa

hESF-6b

Basal medium hESF-Growc hESF-Growc Supplements 10 mg/mL 10 mg/mL Bovine pancreas insulin (I-5500d) Human apotransferrin (T-1147d) 5 mg/mL 5 mg/mL 10 mM 10 mM 2-Mercaptoethanol (M-7522d) Ethanolamine (E-0135d) 10 mM 10 mM 20 nM 20 nM Sodium selenite (S-9133d) 100 ng/mL L-ascorbic acid-2-phosphate (013-196411e) 100 ng/mL Oleic acidf (O-3008d) 4.7 mg/mL Heparin sodium salt from 100 ng/mL porcine intestinal mucosa (H-3149d) 10 ng/mL Basic FGFe 2 ng/mL Human recombinant activin A (338-ACg) a

Furue et al. (2008); Hayashi et al. (2010). Kinehara et al. (2013); Nakanishi et al. (2009). c Obtained from Cell Science & Technology Institute, Miyagi, Japan. d dSigma–Aldrich, St. Louis, MO, USA. e Wako Pure Chemical Inc., Osaka Japan. f Conjugated with 0.5 mg/mL of fraction V fatty acid-free bovine serum albumin. g R&D Systems, Minneapolis, MN, USA. b

coated plates at a concentration of 5  103 cells/cm2 and cultured in hESF-9a medium with 5 mM Rho-associated coil kinase (ROCK) inhibitor (Y-27632; Wako). For antitumor drug assays, hiPSCs cultured in the microchamber were stained with calcein-AM (Wako) and green fluorescent microphotographs were taken; and cells cultured in a 96-well plate were stained with crystal violet and the absorbance was read at 595 nm with a microplate reader (Hayashi et al., 2007). Design of the Microchamber Array Chip We designed a perfusion culture microchamber array chip that was suitable for ECM coating, cell loading, and hiPSC perfusion culture based on our previous reports (Hattori et al., 2011; Sugiura et al., 2008), and fabricated it from PDMS as described in Supplementary Methods. In this system, ECM-coating solution and cell suspensions are loaded into all 64 microchambers (4 lanes  16 microchambers) from a cellinlet/medium-outlet port (Fig. 1a, right) via five cell-inlet/ medium-outlet main channels (Fig. 1b and c). The culture

media are supplied from a medium-inlet port (Fig. 1a, left) via four medium-inlet main channels (Fig. 1b and c). By applying pressure to these four channels, four different culture conditions can be generated in four lanes of microchambers on a single chip (Fig. 1b). Figure 1c shows the detailed structure of a cell culture microchamber (depth [mean  SD], 223.0  5.2 mm, n ¼ 6; diameter, 1,530 mm) and connecting microchannels. The connecting channels are a medium-inlet branch channel (depth, 12.6  0.6 mm, n ¼ 6; width, 50 mm) and cell-inlet/medium-outlet branch channel (depth, 53.2  2.9 mm, n ¼ 6; width, 100 mm). Because the medium-inlet branch channel is much shallower than the other microchannels, the flow rate in each microchamber is determined by the maximum fluidic resistance of the medium-inlet branch channel. Thus, the same flow rate between each microchamber is achieved during coating, cell loading, and perfusion culture. The microchamber used in this study has a slightly deeper medium-inlet branch channel than the 5-mm-deep channel used in our previous studies (Hattori et al., 2011; Sugiura et al., 2008). This modification enabled the introduction of coating solution before cell

Figure 1. Structure of the perfusion culture microchamber array chip. (a) Overview of the perfusion culture microchamber array chip. (b) Enlarged view of the array. To visualize the microfluidic networks on the perfusion culture microchamber array chip, dye solutions (red, new coccine [Kanto Chemical Co., Inc., Tokyo, Japan]; blue, gardenia blue [Wako]; yellow, gardenia yellow [Wako]; green, fast green [Wako]) were introduced into the array from the medium-inlet port (left). The numbers above each microchamber represent position number. (c) Enlarged view of the microchamber. Note that the color density designates three different depths of microstructure.

Yoshimitsu et al.: Defined Microfluidic Perfusion Culture of hiPSCs Biotechnology and Bioengineering

939

loading; in our previous studies, cells were loaded into a precleaned and dried microchamber, which was non-coated with ECM. Perfusion Culture in the Microchamber Array Chip The assembled microchamber array chip was filled with PBS, coated with 1 mg/cm2 fibronectin solution in PBS for more than 12 h at 37 C, and used without any washing process. hiPSCs were harvested by incubation in 0.02% (w/w) EDTA in PBS for 10 min and suspended in hESF-9a medium containing 5 mM ROCK inhibitor. The cell suspension (4.2  105 cells/mL) was added to the cell-inlet/mediumoutlet port by using a micropipette, and the cells were loaded into the microchambers by applying 5 kPa of pressure to the same port. The cell-loaded array chips were first incubated under static culture conditions in a CO2 incubator to induce cell adherence on the microchamber surface. After 24 h, intermittent perfusion of hESF-9a medium was started by applying sequenced pressure (Table II) with a sequenced pressure control system (Engineering System Co. Ltd., Matsumoto, Japan) equipped with an S100 air pump (Atem Corp., Tokyo, Japan) and PR-4102 pressure regulators (GL Science, Tokyo, Japan). Controlling the range of applied pressure at 5–50 kPa enabled intermittent perfusion culture with the desired medium flow-rate range of 10–50 mL/h. During cell cultivation, the top surface of the microchamber array chip was covered with a piece of wet gauze to avoid drying of the array through the gas-permeable PDMS. Flow Cytometry and Immunocytochemistry Conventional flow cytometric analysis and immunocytochemistry were performed as described in Supplementary Methods. A cell sorter (JSAN, Bay Bioscience Co., Ltd., Hogo,

Japan) was used for data acquisition. Antibody information is listed in Supplementary Table I. For immunocytochemistry in the microchamber array, solutions were perfused by applying 30 kPa pressure. In each lane, the hiPSCs were rinsed with 400 mL of PBS containing 0.5 mM CaCl2 and 0.5 mM MgCl2 (PBSþ/þ), and fixed with 150 mL of 4% formaldehyde containing 0.5 mM MgCl2 and 0.5 mM CaCl2 for 20 min. The cells were then rinsed with 400 mL of PBSþ/þ, permeabilized and blocked with 150 mL of PBSþ/þ containing 0.2% Triton X-100 and 10 mg/mL BSA for 90 min, and reacted with 150 mL of primary antibodies in PBSþ/þ containing 0.2% Triton X-100 and 10 mg/mL BSA. The primary antibody binding was visualized with 150 mL of secondary antibodies in the same solution for 3 h at room temperature. Nuclei were stained with 0.4 mM DAPI (Wako). Micrographs were produced by using a BZ-8100 microscope (Keyence, Osaka, Japan) and analyzed with Image J software (NIH, Bethesda, ML).

Results ECM-Coating Conditions for Culturing hiPSCs on PDMS in a Defined Culture Medium We assessed which ECM component, fibronectin, laminin, collagen, or gelatin, was suitable for culturing 201B7 hiPSCs on PDMS in a defined culture medium, hESF-9a (Fig. 2a). First we examined the differences between ECMs in terms of attachment efficiencies (i.e., the number of cells at day 1 normalized by the number of plated cells). The attachment efficiency of fibronectin was significantly larger than that of collagen or gelatin, and the attachment efficiency of laminin was significantly larger than that of gelatin (Fig. 2b). Next, we assessed two self-renewal properties of hiPSCs: proliferation and the expression of a self-renewal marker, TRA-1-60. The number of cells in each well at day 4 was normalized by that at

Table II. Applied pressure sequence and flow conditions for perfusion culture. Medium change (MC) conditions Static culture Parameter

Unit

Pressure Period of pressure Air vent time Cycle time Flow rate during pressuring for the 16 microchambers Average flow rate for the 16 microchambers Maximum shear stress in the microchamberb

kPa min min min mL/min mL/h Pa

1 MC/day a

4–5 >3.0a — 1,440 — — 10–3c

Intermittent perfusion

3 MCs/day a

4–5 >3.0a — 480 — — 10–3c

Low 4.1 0.1 3.3 3.4 7.1 12.5

10–3d

High 4.1 0.4 3.0 3.4 7.1 50

10–3d

a In static cultures, pressure was applied to the medium-inlet port during the change of medium until 20–50 mL of medium appeared in the cell-inlet/ medium-outlet port. b Shear stress generated by a fluid flowing in the microchamber wasestimated by t ¼ 6mQ/wd2, where t is the shear stress, Q is the flow rate during the medium change or perfusion, m is the viscosity, and w and d are the width and depth of the microchamber, respectively. Note that the maximum shear stress in the microchamber had the same order of magnitude in the different perfusion conditions, because the flow rate during the medium change and that during the pressuring were almost the same. c Shear stress during the change of medium was estimated. d Shear stress during applied pressure was estimated.

940

Biotechnology and Bioengineering, Vol. 111, No. 5, May, 2014

b

PDMS -bottomed well

PDMS

Fb

Lm

Col

Gel

ECM coat Day 0 Plate hiPSCs

Day 4 cell count & FCM

Growth rate

0.6 0.4 0.2 0

14 12 10 8 6 4 2 0

Lm ** **

Fb TRA-1-60 +(%)

e

**

0.8

TRA-1-60+

101 10 2 10 3 10 4 Fluorescent intensity (TRA-1-60)(a.u.)

** **

Fb

c

Cell counts (a.u.)

d

1

Attachment efficiency

Growth rate

Day 1 cell count

Attachment efficiency

a

100

Col

Gel

Col

Gel

** **

Lm **

**

80 60 40 20 0 Fb

Lm

Col

Gel

Figure 2. Screening for ECMs suitable for coating PDMS surfaces for the culture of hiPSCs in defined culture medium. (a) Schematic of the experimental procedure. 201B7 hiPSCs were plated at a density of 1  104 cells/cm2 in defined culture medium, hESF-9a, on PDMS coated with 2 mg/cm2 fibronectin (Fb, red), 2 mg/cm2 collagen (Col, green), 2 mg/ cm2 laminin (Lm, blue), or 0.1% gelatin (Gel, purple). Cells were counted at days 1 and 4 and flow cytometric analysis was performed at day 4. (b) Cell attachment efficiencies, defined as the number of cells at day 1 normalized by the number of plated cells, under each ECM condition. Data are presented as mean  SE, n ¼ 4. (c) Growth rates of cells, defined as the number of cells at day 4 normalized by that at day 1, under each ECM condition. Data are presented as mean  SE, n ¼ 4. (d and e) Flow cytometric analysis of the self-renewal marker detected by anti-TRA-1-60 antibody. Histograms showing the number of cells at various fluorescent intensities (flow cytometry profile) (d) and the percentages of anti-TRA-160 positive cells (fluorescent intensity, >102 a.u.) under each ECM condition (e). Data are presented as mean  SE, n ¼ 3.   P < 0.01, Tukey multiple comparison.

day 1 to estimate proliferation. hiPSCs proliferated at a significantly higher rate on fibronectin and laminin coating than on collagen or gelatin coating (Fig. 2c). The results were similar even when a 10-times concentration of ECM was used (Supplementary Fig. 1). Flow cytometric analysis of the selfrenewal marker, TRA-1-60, showed that the number of TRA1-60 positive hiPSCs was significantly greater when the cells were cultured on fibronectin or laminin than when they were cultured on gelatin (Fig. 2d and e). Taken together, these results suggest that both fibronectin and laminin are suitable ECMs for culturing hiPSCs on PDMS. Because we routinely use fibronectin for culturing hiPSCs on polystyrene culture dishes (Hayashi et al., 2010), we decided to use fibronectin as

the ECM in the following experiments using the microchamber array chip made from PDMS. Effect of Perfusion Culture on Self-Renewal of hiPSCs After coating the microchamber array with fibronectin, we loaded 201B7 hiPSCs into the array from the cell inlet/ medium outlet port with applied pressure at 5 kPa (Fig. 1a). After a 24-h period of static culture, we observed the adhesion and extension of hiPSCs on the microchamber surface (Fig. 3a). We then started perfusion of hESF-9a medium with sequenced applied pressure at a total flow rate of 12.5 mL/h for all microchambers (Table II). After 2 days of perfusion

Yoshimitsu et al.: Defined Microfluidic Perfusion Culture of hiPSCs Biotechnology and Bioengineering

941

Figure 3.

Self-renewal of hiPSCs under static or perfusion culture conditions in the microchamber array. 201B7 hiPSCs in hESF-9a medium were loaded into the fibronectincoated microchamber at day 0. (a and b) Phase contrast images of hiPSCs in the microchamber at day 1 (a) and day 3 (i.e., 2 days after start of medium perfusion [12.5 mL/h]) (b). (c) Histogram of the numbers of cells at day 4 after they were loaded in the microchambers at various positions (only data for the non-edge positions, 2–7, is shown [see Fig. 1b]). There was no significant difference between the number of cells at each position examined (n ¼ 4, P > 0.19, One-way ANOVA). (d) Schematic of experimental protocols for cells in static culture (upper two panels) and perfusion culture (lower two panels) (See Tables II and III). After loading, cells were cultured overnight without perfusion to allow them to adhere on the microchamber surface. From day 1, the medium was changed once per day (1 MC/day) or three times per day (3 MCs/day) for static culture or cells were perfused at an average flow rate of 12.5 or 50 mL/h. Immunocytochemical staining was performed at day 4. Longer term culture (1 week) was performed at an average flow rate of 12.5 mL/h. (e) Immunocytochemistry of the self-renewal marker, Oct 3/4 (red). Nuclei were stained with DAPI (blue). (f) Growth rate of hiPSCs in the microchamber array. Growth rate was defined as the numbers of cells at day 4 normalized by that at day 1. Control cells were cultured on a 24-well plate coated with 1 mg/cm2 fibronectin, and the 500 mL volume of culture medium was changed once a day. Data are presented as mean  SE, n ¼ 3,   P < 0.01, Tukey multiple comparison. (g) Immunocytochemistry of self-renewal markers, Oct 3/4 (red), TRA 1-60 (green), SSEA4 (green), and SSEA3 (green), and the early differentiation marker, SSEA1 (green). hiPSCs were cultured for 1 week. Nuclei were stained with DAPI (blue).

culture, the cells formed normal hiPSC colonies (Fig. 3b), which were tightly packed, and flat colonies consisting of cells with large nuclei and scant cytoplasm (Takahashi et al., 2007; Thomson et al., 1998), suggesting that we had successfully

942

Biotechnology and Bioengineering, Vol. 111, No. 5, May, 2014

cultured hiPSCs in the microchamber array under fully defined conditions. After 3 days of perfusion, no statistically significant difference in the number of cells/microchamber was observed for the non-edge positions of the array (i.e.,

Table III. Comparison of culture conditions between perfusion and static cultures. Static culture Parameter (unit) Culture vessel Culture area (mm2) Culture volume (mL) Frequency of MCs (number/day) Rate of medium change per microchamber or well (mL/day/well) Rate of medium change per unit culture area (mL/day/mm2)

Intermittent perfusion

Static culture (control)

1 MC/day

3 MCs/day

Low

High

24-well plate 200 500 1 time/day 500 2.5

Microchamber 22.4 6.4 1 time/day 6.4 0.29

3 time/day 19.2 0.86

Semi-continuous 18.8 13.4

75 53.4

MC, complete medium change.

positions 2–7 in Figs. 1b and 3c). The number of loaded cells in the microchambers at either edge of the array was less than that in microchambers in the non-edge positions. This phenomenon could be explained by the hydrodynamic filtration phenomenon (Sugiura et al., 2008; Yamada and Seki, 2005), whereby cells cannot exist in the space near the microchannel wall because of their own size. Culturing cells in a conventional culture dish requires that the medium be changed every 1–3 days to supply nutrients and remove waste products (Davis, 2002), and it is recommended to change the culture medium of hiPSCs every day (Hayashi et al., 2010; Suemori et al., 2006). Moreover, autocrine or paracrine factors or both affect proliferation and differentiation of hiPSCs (Bendall et al., 2009; Sato et al., 2003). Thus hiPSC self-renewal could be affected by an excessive perfusion flow rate as well as a lack of perfusion. To reveal the effects of medium perfusion on culturing hiPSCs, we cultured the cells under four perfusion conditions for 3 days after 1 day of static culture (Table II, Fig. 3d). As a control, the hiPSCs were also cultured in hESF9a medium on fibronectin-coated polystyrene culture dishes. The self-renewal property (i.e., undifferentiated state) of the cells was evaluated by examining the expression of a selfrenewal marker (Oct 3/4) and the growth rate, calculated as the number of cells at day 4 normalized by the number of cells loaded. Immunocytochemical staining revealed that cells under all four culture conditions expressed Oct 3/4 (Fig. 3e). The growth rate was lowest for the static culture with one complete medium change (MC)/day, intermediate for the static culture with 3 MCs/day, and highest for the low-rate (12.5 mL/h) and high-rate (50 mL/h) perfusion cultures and control cultures, which shared similar growth-rates (Fig. 3f). These results suggest that perfusion culture at a flow rate of 12.5 mL/h is suitable for maintaining the self-renewal property of hiPSCs in the microchamber array. The hiPSCs were successfully cultured for 1 week at 12.5 mL/h and expressed four self-renewal markers (TRA1-60, SSEA4, SSEA3, and Oct3/4) but did not express an early differentiation marker (SSEA1; Fig. 3g).

early differentiation into the extra-embryonic trophoblast lineage (Supplementary Fig. 2; Na et al., 2010; Vallier et al., 2009; Xu et al., 2002). 201B7 hiPSCs were loaded in hESF-9a medium into the fibronectin-coated microchamber array and cultured from day 1 to day 3 at a flow rate of 12.5 mL/h under four culture medium conditions: hESF-9a medium, hESF-6 medium (Table I), and hESF-6 medium supplemented with 10 or 50 ng/mL BMP4 (Fig. 4a). Cells were cultured as a monolayer, and spherical aggregates or embryoid bodies were not formed under any of the conditions (Fig. 4b, top). The cells cultured in hESF-6 supplemented with BMP4 became cobble-stone-like with large, flat cell bodies, and dark nuclei under phase-contrast microscopy (Fig. 4b). Immunocytochemical staining revealed that cells cultured in hESF-9a or hESF-6 medium without BMP4 were positive for the self-renewal marker, Oct 3/4. In contrast, cells cultured in hESF-6 medium plus BMP4 were positive for SSEA1 (Fig. 4b). Similar results were obtained by using 96-well plates (Supplementary Fig. 3) or another hiPSC line, Tic, with perfusion culture (Supplementary Fig. 4). The relative fluorescence intensities of Oct 3/4 and SSEA1 per cell in various culture conditions, normalized by the corresponding values for hESF-9a medium (control), showed that Oct 3/4 expression in hESF-6 medium with 10 or 50 ng/mL BMP4 and SSEA1 expression in hESF-6 medium with 50 ng/mL BMP4 were significantly different from those of the corresponding control condition (Fig. 4c). These findings suggest that the differentiation state of the hiPSCs was controlled by perfusion of defined culture media, hESF9a, and hESF-6 with or without a specific induction factor in the microchamber array chip. To test whether our system could be used for drug screening, hiPSCs were cultured in the same set of conditions and then treated with one of three antitumor drugs, actinomycin D, mitomycin C, or mithramycin (Morgan and Holguin, 2002) for 1 day (Fig. 4a). The relative numbers of cells based on fluorescence were comparable to the results obtained by using 96-well plates (Fig. 4d–f).

Discussion Differentiation and Drug Assay of hiPSCs To determine whether hiPSCs could be successfully differentiated under fully defined culture conditions in the perfusion culture microchamber array, we used BMP4, which induces

In the present study, we developed a microfluidic perfusion culture system for hiPSCs that uses a pressure-driven microchamber array chip with defined ECM (purified fibronectin) and defined culture medium conditions

Yoshimitsu et al.: Defined Microfluidic Perfusion Culture of hiPSCs Biotechnology and Bioengineering

943

Figure 4. Differentiation and drug response of hiPSCs in the microchamber array chip. (a) Schematic of experimental protocols. 201B7 hiPSCs in hESF-9a medium were loaded into the microchamber array at day 0 and allowed to adhere on the microchamber in overnight static culture. One day later, the four lanes of the array were perfused at an average flow rate of 12.5 mL/h with four types of defined culture medium: hESF-9a, hESF-6 medium without BMP4, and hESF-6 medium supplemented with 10 ng/mL BMP4 (þ10 BMP) or 50 ng/ mL BMP4 (þ50 BMP) (upper time course, differentiation assay shown in panels b and c) or hESF-9a medium (Cont), hESF-9a medium with antitumor drugs: 100 ng/mL actinomycin D with D-mannitol (þACD); 100 ng/mL mitomycin C (þMMC); or 100 ng/mL mithramycin (þMTM) (lower time course, drug assay shown in panels d and e). Immunocytochemical staining was performed at day 4 for the differentiation assay shown in panel b, and live cell staining was performed at day 2 for the antitumor drug assay shown in panel d. (b and c) Differentiation assay. (b) Microphotographs of hiPSCs at day 4 (3 days of differentiation). The top panels show phase-contrast micrographs (PhC). The other panels show immunocytochemistry of the self-renewal marker, Oct 3/4 (red), and the early differentiation marker, SSEA1 (green), nuclear staining with DAPI (blue), and the merged image. The white dotted lines represent the edges of cell culture. (c) Relative fluorescent intensity per cell for Oct 3/4 (red) and SSEA1 (green) under various conditions normalized by the corresponding value for hESF-9a (Cont) medium. Data are presented as mean  SE, n  16.  P < 0.05,   P < 0.01, Dunnetts multiple comparison against hESF-9a (Cont). (d–f) Antitumor drug assays comparing use of the microchamber array chip (d and e) and 96-well plate (f). (d) Bright field (BF) photograph of the microchamber, and fluorescent microphotographs of hiPSCs stained with 1 mM of calcein-AM for 20 min on day 2 in the microchamber (i.e., day 1 of antitumor drug treatment) (e) The number of cells per microchamber were quantified from the calcein-AM fluorescent microphotograph by using Image J software. (f) Two days after hiPSCs were plated into a 96-well plate, they were subjected to 24-h treatment with antitumor drugs. The cells were then stained with crystal violet, and the dye was eluted and quantified by absorbance at 595 nm. (e and f) Data were normalized with the control value. Data are presented as mean  SE, n ¼ 12 (e) and n ¼ 4 (f) and P-values are derived from Tukey multiple comparison.

944

Biotechnology and Bioengineering, Vol. 111, No. 5, May, 2014

(hESF-9a, hESF-6 with or without BMP4), allowing us to control the differentiation state of the hiPSCs. Because protein adsorption on PDMS is different from that on a conventional polystyrene culture dish we screened various ECMs to determine the optimal coating conditions for PDMS (Hattori et al., 2011). Using the appropriate ECM is particularly important when a defined culture medium is used, because it does not contain any ECM components (Davis, 2002; Furue et al., 2008; Hayashi et al., 2007; Ohnuma et al., 2006). Even when culturing hiPSCs under serum- and feeder-free culture conditions, undefined components such as Matrigel (BD Biosciences; Ludwig et al., 2006; Vukicevic et al., 1992) might be used. Because Matrigel is an assortment of ECM proteins extracted from Englebreth–Holm–Swarm tumors in mice, it contains a variety of growth factors including basic fibroblast growth factor and transforming growth factor beta, which affect hiPSCs (Hughes et al., 2010; Orkin et al., 1977; Vukicevic et al., 1992). Here, we found that purified fibronectin and laminin were suitable ECMs for use on PDMS surfaces with hESF-9a medium as assessed by cell attachment efficiency, proliferation, and expression of a self-renewal marker. Because fibronectin is a receptor of integrin a5b1 and avb1 and laminin is a receptor of integrin a6b1 and avb5, all of which are known to be expressed on hiPSCs, these ECMs have been successfully used for culturing hiPSCs on polystyrene culture dishes (Barczyk et al., 2010; Braam et al., 2008; Fadeev and Melkoumian, 2011; Miyazaki et al., 2008; Rowland et al., 2009), which is consistent with our results. Gelatin, which is a hydrolyzed form of collagen, is routinely used as an ECM on polystyrene culture dishes for culturing hPSC with KSR supplemented medium and feeder cells (Takahashi et al., 2007; Thomson et al., 1998; Yu et al., 2007), but here we found that both gelatin and collagen were inferior ECMs for use with PDMS and defined culture medium. We showed that perfusion is important to keep the growth rate as high as that observed for conventional static culture, because of the small volume of the microchamber. Although proliferation was slow in static culture in the microchamber array, the self-renewal marker Oct 3/4 was expressed, suggesting that the hiPSCs maintained their undifferentiated state for 3 days. The rate of medium change per unit culture area, which is proportional to the volume of medium per cell per day, for static culture in the microchamber (1 MC/day) was 0.29 mL/day/mm2. This value was about one-ninth that achieved for static culture in the 24-well plate (2.5 mL/day/ mm2). Thus, nutrients needed for proliferation might be depleted during static culture in the microchamber. Furthermore, under static culture conditions, nutrients, autocrine or paracrine factors, and wastes, as well as pH, osmolality, and temperature drastically fluctuate with the medium change every few days. These changes potentially affect the state of the cells. For example, osmolality affects cell growth (Potter and DeMarse, 2001) and sudden increases in the concentrations of nutrients (serum) induce the circadian expression of various genes (Balsalobre et al., 1998). In contrast, perfusion culture can keep the culture conditions stable by maintaining a continuous supply of fresh medium

to the cells, and thus is suitable for use in drug discovery and differentiation studies. We successfully controlled differentiation of hiPSCs in the microchamber array by using a combination of a medium perfusion system and fully defined culture conditions [with hESF-9a and hESF-6 as defined culture media and pure fibronectin as the defined ECM (Furue et al., 2008; Hayashi et al., 2010)]. The hiPSCs differentiated with addition of BMP4 to the hESF-6 medium. hESF-6 medium does not contain oleic acid, ascorbic acid, bFGF, heparin, or activin, which play a role in the self-renewal of hiPSCs and could affect the signaling of BMP4 (Furue et al., 2008; Hayashi et al., 2010; Kinehara et al., 2013; Nakanishi et al., 2009; Tabernero et al., 2001). Thus, the results indicate that the differentiation depended only on BMP4. Moreover, the rate of medium consumption in each microchamber (18.8 mL/ day/microchamber, Table III) was 1/27th and 1/11th of the standard rates used for 24-well (500 mL/day/well) and 96-well (200 mL/day/well) plates, respectively, suggesting that the use of the microchamber array was cost effective. In addition, the microchamber array designed in this study can provide four culture conditions on a single chip (Hattori et al., 2011; Sugiura et al., 2008). Therefore, our screening system is advantageous for use in drug discovery and differentiation analysis because it is highly efficient for the detection of the effects of drugs, it has parallel multiple-assay capability, and it requires low consumption of medium.

Conclusions We have designed and fabricated a PDMS microchamber array chip suitable for ECM coating, controlled cell-loading, microfluidic perfusion culture, and immunocytochemical analysis. Both fibronectin and laminin were determined to be appropriate ECMs for use on PDMS surfaces in defined culture medium, hESF-9a. We used intermittent perfusion driven by sequenced applied pressure in the microchamber array and found that the growth rate of hiPSCs under perfusion culture conditions was higher than that under static culture conditions in which the medium was replaced one or three times a day. By applying the microchamber array to the culture of hiPSCs under fully defined perfusion culture conditions, we succeeded in controlling the differentiation and drug screening of hiPSCs. We believe that the system developed here will be a platform technology for future use in drug discovery and differentiation analysis using fully defined culture media. This work was supported in part by Promotion of Independent Research Environment for Young researchers to K.O. The funding bodies had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. A part of this work was conducted at the Advanced Industrial Science and Technology Nano-Processing Facility, Tsukuba, Japan.

References Balsalobre A, Damiola F, Schibler U. 1998. A serum shock induces circadian gene expression in mammalian tissue culture cells. Cell 93(6):929–937.

Yoshimitsu et al.: Defined Microfluidic Perfusion Culture of hiPSCs Biotechnology and Bioengineering

945

Barczyk M, Carracedo S, Gullberg D. 2010. Integrins. Cell Tissue Res 339(1):269–280. Bendall SC, Hughes C, Campbell JL, Stewart MH, Pittock P, Liu S, Bonneil E, Thibault P, Bhatia M, Lajoie GA. 2009. An enhanced mass spectrometry approach reveals human embryonic stem cell growth factors in culture. Mol Cell Proteomics 8(3):421–432. Blagovic K, Kim LLY, Voldman J. 2011. Microfluidic perfusion for regulating diffusible signaling in stem cells. PLoS ONE 6(8). Braam SR, Zeinstra L, Litjens S, Ward-van Oostwaard D, van den Brink S, van Laake L, Lebrin F, Kats P, Hochstenbach R, Passier R. 2008. Recombinant vitronectin is a functionally defined substrate that supports human embryonic stem cell self-renewal via aVb5 integrin. Stem Cells 26(9):2257–2265. Davis JM. editor. 2002. Basic cell culture. 2nd edn. Oxford, England: Oxford University Press. Deng T, Wu HK, Brittain ST, Whitesides GM. 2000. Prototyping of masks, masters, and stamps/molds for soft lithography using an office printer and photographic reduction. Anal Chem 72(14):3176–3180. Duffy DC, McDonald JC, Schueller OJA, Whitesides GM. 1998. Rapid prototyping of microfluidic systems in poly(dimethylsiloxane). Anal Chem 70(23):4974–4984. Egawa N, Kitaoka S, Tsukita K, Naitoh M, Takahashi K, Yamamoto T, Adachi F, Kondo T, Okita K, Asaka I. 2012. Drug screening for ALS using patient-specific induced pluripotent stem cells. Sci Trans Med 4(145): 145ra104. El-Ali J, Sorger PK, Jensen KF. 2006. Cells on chips. Nature 442(7101):403–411. Ellison D, Munden A, Levchenko A. 2009. Computational model and microfluidic platform for the investigation of paracrine and autocrine signaling in mouse embryonic stem cells. Mol Biosyst 5(9):1004–1012. Fadeev AG, Melkoumian Z. 2011. Synthetic surfaces for human embryonic stem cell culture. Embryonic stem cells—Basic biology to bioengineering: Intech. Furue MK, Na J, Jackson JP, Okamoto T, Jones M, Baker D, Hata R, Moore HD, Sato JD, Andrews PW. 2008. Heparin promotes the growth of human embryonic stem cells in a defined serum-free medium. Proc Natl Acad Sci U S A 105(36):13409–13414. Gupta K, Kim D-H, Ellison D, Smith C, Kundu A, Tuan J, Suh K-Y, Levchenko A. 2010. Lab-on-a-chip devices as an emerging platform for stem cell biology. Lab Chip 10(16):2019–2031. Hattori K, Sugiura S, Kanamori T. 2011. Microenvironment array chip for cell culture environment screening. Lab Chip 11(2):212–214. Hayashi Y, Chan T, Warashina M, Fukuda M, Ariizumi T, Okabayashi K, Takayama N, Otsu M, Eto K, Furue MK., et al. 2010. Reduction of NGlycolylneuraminic acid in human induced pluripotent stem cells generated or cultured under feeder- and serum-free defined conditions. PLoS ONE 5(11):e14099. Hayashi Y, Furue MK, Okamoto T, Ohnuma K, Myoishi Y, Fukuhara Y, Abe T, Sato JD, Hata R, Asashima M. 2007. Integrins regulate mouse embryonic stem cell self-renewal. Stem Cells 25(12):3005–3015. Hughes CS, Postovit LM, Lajoie GA. 2010. Matrigel: A complex protein mixture required for optimal growth of cell culture. Proteomics 10(9): 1886–1890. International Stem Cell Initiative Consortium, Akopian V, Andrews PW, Beil S, Benvenisty N, Brehm J, Christie M, Ford A, Fox V, Gokhale PJ, et al. 2010. Comparison of defined culture systems for feeder cell free propagation of human embryonic stem cells. In Vitro Cell Dev Biol Anim 46(3–4):247–258. Itzhaki I, Maizels L, Huber I, Zwi-Dantsis L, Caspi O, Winterstern A, Feldman O, Gepstein A, Arbel G, Hammerman H., et al. 2011. Modelling the long QT syndrome with induced pluripotent stem cells. Nature 471(7337):225–229. Jo BH, Van Lerberghe LM, Motsegood KM, Beebe DJ. 2000. Threedimensional micro-channel fabrication in polydimethylsiloxane (PDMS) elastomer. J Microelectromech Syst 9(1):76–81. Kamei K-i, Guo S, Yu ZTF, Takahashi H, Gschweng E, Suh C, Wang X, Tang J, McLaughlin J, Witte ON, et al. 2009. An integrated microfluidic culture device for quantitative analysis of human embryonic stem cells. Lab Chip 9(4):555–563.

946

Biotechnology and Bioengineering, Vol. 111, No. 5, May, 2014

Khoury M, Bransky A, Korin N, Konak L, Enikolopov G, Tzchori I, Levenberg S. 2010. A microfluidic traps system supporting prolonged culture of human embryonic stem cells aggregates. Biomed Microdevices 12(6):1001–1008. Kim L, Vahey MD, Lee HY, Voldman J. 2006. Microfluidic arrays for logarithmically perfused embryonic stem cell culture. Lab Chip 6(3): 394–406. Kinehara M, Kawamura S, Tateyama D, Suga M, Matsumura H, Mimura S, Hirayama N, Hirata M, Uchio-Yamada K, Kohara A. 2013. Protein kinase C regulates human pluripotent stem cell self-renewal. PLoS ONE 8(1):e54122. Liu G-H, Barkho BZ, Ruiz S, Diep D, Qu J, Yang S-L, Panopoulos AD, Suzuki K, Kurian L, Walsh C. 2011. Recapitulation of premature ageing with iPSCs from Hutchinson–Gilford progeria syndrome. Nature 472(7342): 221–225. Ludwig TE, Levenstein ME, Jones JM, Berggren WT, Mitchen ER, Frane JL, Crandall LJ, Daigh CA, Conard KR, Piekarczyk MS., et al. 2006. Derivation of human embryonic stem cells in defined conditions. Nat Biotechnol 24(2):185–187. Minami I, Yamada K, Otsuji TG, Yamamoto T, Shen Y, Otsuka S, Kadota S, Morone N, Barve M, Asai Y, et al. 2012. A small molecule that promotes cardiac differentiation of human pluripotent stem cells under defined, cytokine- and xeno-free conditions. Cell Rep 2:1–13. Miyazaki T, Futaki S, Hasegawa K, Kawasaki M, Sanzen N, Hayashi M, Kawase E, Sekiguchi K, Nakatsuji N, Suemori H. 2008. Recombinant human laminin isoforms can support the undifferentiated growth of human embryonic stem cells. Biochem Biophys Res Commun 375(1): 27–32. Moledina F, Clarke G, Oskooei A, Onishi K, Günther A, Zandstra PW. 2012. Predictive microfluidic control of regulatory ligand trajectories in individual pluripotent cells. Proc Natl Acad Sci USA 109(9):3264– 3269. Morgan CD, Holguin MH. 2002. Chemotherapeutic stress mediated by certain antitumor antibiotics induces an atypical CD69þ surface phenotype in peripheral T-lymphocytes. Int Immunopharmacol 2:367–380. Na J, Furue MK, Andrews PW. 2010. Inhibition of ERK1/2 prevents neural and mesendodermal differentiation and promotes human embryonic stem cell self-renewal. Stem Cell Res 5(2):157–169. Nagata S, Toyoda M, Yamaguchi S, Hirano K, Makino H, Nishino K, Miyagawa Y, Okita H, Kiyokawa N, Nakagawa M. 2009. Efficient reprogramming of human and mouse primary extra-embryonic cells to pluripotent stem cells. Genes Cells 14(12):1395–1404. Nakanishi M, Kurisaki A, Hayashi Y, Warashina M, Ishiura S, Kusuda-Furue M, Asashima M. 2009. Directed induction of anterior and posterior primitive streak by Wnt from embryonic stem cells cultured in a chemically defined serum-free medium. FASEB J 23(1):114–122. Nishino K, Toyoda M, Yamazaki-Inoue M, Fukawatase Y, Chikazawa E, Sakaguchi H, Akutsu H, Umezawa A. 2011. DNA methylation dynamics in human induced pluripotent stem cells over time. PLoS Genet 7(5): e1002085. Ohnuma K, Hayashi Y, Furue M, Kaneko K, Asashima M. 2006. Serum-free culture conditions for serial subculture of undifferentiated PC12 cells. J Neurosci Methods 151(2):250–261. Orkin RW, Gehron P, McGoodwin EB, Martin GR, Valentine T, Swarm R. 1977. A murine tumor producing a matrix of basement membrane. J Exp Med 145(1):204–220. Potter SM, DeMarse TB. 2001. A new approach to neural cell culture for longterm studies. J Neurosci Methods 17-24. Price PJ, Goldsborough MD, Tilkins ML. 1998. Embryonic stem cell serum replacement patent WO/1998/30679. Rowland TJ, Miller LM, Blaschke AJ, Doss EL, Bonham AJ, Hikita ST, Johnson LV, Clegg DO. 2009. Roles of integrins in human induced pluripotent stem cell growth on matrigel and vitronectin. Stem Cells Dev 19(8):1231–1240. Sato N, Sanjuan IM, Heke M, Uchida M, Naef F, Brivanlou AH. 2003. Molecular signature of human embryonic stem cells and its comparison with the mouse. Dev Biol 260(2):404–413.

Suemori H, Yasuchika K, Hasegawa K, Fujioka T, Tsuneyoshi N, Nakatsuji N. 2006. Efficient establishment of human embryonic stem cell lines and long-term maintenance with stable karyotype by enzymatic bulk passage. Biochem Biophys Res Commun 345(3):926–932. Sugiura S, Edahiro J, Kikuchi K, Sumaru K, Kanamori T. 2008. Pressuredriven perfusion culture microchamber array for parallel drug cytotoxicity assay. Biotechnol Bioeng 100:1156–1165. Tabernero A, Lavado EM, Granda B, Velasco A, Medina JM. 2001. Neuronal differentiation is triggered by oleic acid synthesized and released by astrocytes. J Neurochem 79(3):606–616. Takahashi K, Tanabe K, Ohnuki M, Narita M, Ichisaka T, Tomoda K, Yamanaka S. 2007. Induction of pluripotent stem cells from adult human fibroblasts by defined factors. Cell 131(5):861–872. Thomson JA, Itskovitz-Eldor J, Shapiro SS, Waknitz MA, Swiergiel JJ, Marshall VS, Jones JM. 1998. Embryonic stem cell lines derived from human blastocysts. Science 282(5391):1145–1147. Titmarsh D, Cooper-White J. 2009. Microbioreactor array for full-factorial analysis of provision of multiple soluble factors in cellular microenvironments. Biotechnol Bioeng 104(6):1240–1244. Titmarsh D, Hidalgo A, Turner J, Wolvetang E, Cooper-White J. 2011. Optimization of flowrate for expansion of human embryonic stem cells in perfusion microbioreactors. Biotechnol Bioeng 108(12):2894–2904. Titmarsh DM, Hudson JE, Hidalgo A, Elefanty AG, Stanley EG, Wolvetang EJ, Cooper-White JJ. 2012. Microbioreactor arrays for full factorial screening of exogenous and paracrine factors in human embryonic stem cell differentiation. PLoS ONE 7(12):e52405. Toh Y-C, Blagovic K, Voldman J. 2010. Advancing stem cell research with microtechnologies: Opportunities and challenges. Integr Biol 2(7–8): 305–325. Vallier L, Touboul T, Chng Z, Brimpari M, Hannan N, Millan E, Smithers LE, Trotter M, Rugg-Gunn P, Weber A, et al. 2009. Early cell fate decisions of human embryonic stem cells and mouse epiblast stem cells are controlled by the same signalling pathways. PLoS ONE 4(6):e6082. van Noort D, Ong SM, Zhang C, Zhang SF, Arooz T, Yu H. 2009. Stem cells in microfluidics. Biotechnol Prog 25(1):52–60.

Villa-Diaz LG, Torisawa Y-s, Uchida T, Ding J, Nogueira-de-Souza NC, O’Shea KS, Takayama S, Smith GD. 2009. Microfluidic culture of single human embryonic stem cell colonies. Lab Chip 9(12):1749– 1755. Vukicevic S, Kleinman HK, Luyten FP, Roberts AB, Roche NS, Reddi AH. 1992. Identification of multiple active growth factors in basement membrane Matrigel suggests caution in interpretation of cellular activity related to extracellular matrix components. Exp Cell Res 202(1):1–8. Watanabe K, Ueno M, Kamiya D, Nishiyama A, Matsumura M, Wataya T, Takahashi JB, Nishikawa S, Muguruma K, Sasai Y. 2007. A ROCK inhibitor permits survival of dissociated human embryonic stem cells. Nat Biotechnol 25(6):681–686. Xu RH, Chen X, Li DS, Li R, Addicks GC, Glennon C, Zwaka TP, Thomson JA. 2002. BMP4 initiates human embryonic stem cell differentiation to trophoblast. Nat Biotechnol 20(12):1261–1264. Yamada M, Seki M. 2005. Hydrodynamic filtration for on-chip particle concentration and classification utilizing microfluidics. Lab Chip 5(11):1233–1239. Yang YM, Gupta SK, Kim KJ, Powers BE, Cerqueira A, Wainger BJ, Ngo HD, Rosowski KA, Schein PA, Ackeifi CA. 2013. A small molecule screen in stem-cell-derived motor neurons identifies a kinase inhibitor as a candidate therapeutic for ALS. Cell Stem Cell. Yu J, Vodyanik MA, Smuga-Otto K, Antosiewicz-Bourget J, Frane JL, Tian S, Nie J, Jonsdottir GA, Ruotti V, Stewart R, et al. 2007. Induced pluripotent stem cell lines derived from human somatic cells. Science 318(5858): 1917–1920.

Supporting Information Additional supporting information may be found in the online version of this article at the publisher’s web-site.

Yoshimitsu et al.: Defined Microfluidic Perfusion Culture of hiPSCs Biotechnology and Bioengineering

947