**2. Electrical stimulation of EBs with cell-nonadhesive electrode substrates**

In this section, we summarize the methods and the results of ensemble electrical stimulation of P19 EBs with the microfabricated electrode substrates, as we reported before (Takayama et al. 2009).

The experimental procedures for fabricating microcavity-array electrode substrate are described as follows. A glass substrate with a transparent conductive layer (ITO; Indiumtin-oxide, Sanyo Vacuum) was cleaned with acetone (Wako) and isopropyl alcohol (IPA, Wako). Then, SU-8 3050 negative photoresist (Microchem) was spin-coated onto an ITO substrate for 60 s at 1000 rpm with a thickness of 100 m. The coated substrates were baked on a hotplate for 50 min at 95 oC. The substrates were exposed to UV through a custommade photomask. A reduced projection exposure system (MM-505, Nanometric Technology) was used for fabricating the photomasks. The photomask featured sixteen microcavities with a diameter of 200 or 500 m aligned in 4 x 4 matrix patterns. After baking on a hotplate for 10 min at 95 oC, the substrates were developed in SU-8 developer (Microchem) and rinsed with IPA. Because SU-8 resist have good insulating properties, the bottom of each microcavity, where the ITO layer was exposed, acted as the stimulation electrode. After fabrication of the microcavity-array pattern, a glass ring of 1 cm height was mounted onto a substrate with a silicone elastomer (KE-103, Shin-etsu Silicones). A stranded wire of cupper was also mounted onto a substrate with an electroconductive paste (Dotite FC-415, Fujikura Kasei). The ITO regions inside a glass ring, except the bottom surfaces of the microcavity, were insulated by coating with a silicone elastomer. The protocols for fabricating the microcavity-array electrode substrate were shown in Fig. 1.

In this research, we used P19 cells as a model system for a stem cell. P19 is a mouse embryonal carcinoma cell line and can differentiate into three germ layers (Bain et al. 1994). For its easy handling and high neuronal differentiation rate, we have used P19 cells as a stem cell model. P19 cells were rountinely cultured and passaged every two days in alpha minimum essential medium (-MEM, Invitrogen) containing 10 % fetal bovine serum (FBS, HyClone) and 5 – 40 U/ml penicillin-streptomycin (Sigma-Aldrich). To produce uniform size EBs, P19 cells were collected and replated in a 96-well low cellattachment plate (Sumilon, Sumitomo Bakelite). Due to its spheroid bottom structure, a single EB was formed in each well. The obtained EBs were collected and transferred into

that applying physical stimulation could affect exogenous factors and also differentiation processes of stem cells. Although the experimental methods to regulate the endogenous or exogenous signalling have been proposed as above, increase in differentiation efficiency is inadequate for regeneration therapy, and combination of the two approaches is not

Based on these problems, we have tried to propose a precise control method for cell differentiation by combining EB size treatment and electrical stimulation technique. To do this, we developed microcavity-array device with embedded electrodes (Takayama et al., 2009). Uniform-sized EBs of P19 cells were prepared and aligned in the electrode substrate, and were stimulated uniformly and simultaneously. However, difficulty in handling EBs and limited number of stimulating EBs hindered further analysis. In this chapter, we describe an alternative method to stimulate more number of EBs based on soft-lithography technique and recent attempt to analyse the effects of EB size treatment and electrical

**2. Electrical stimulation of EBs with cell-nonadhesive electrode substrates**  In this section, we summarize the methods and the results of ensemble electrical stimulation of P19 EBs with the microfabricated electrode substrates, as we reported before (Takayama

The experimental procedures for fabricating microcavity-array electrode substrate are described as follows. A glass substrate with a transparent conductive layer (ITO; Indiumtin-oxide, Sanyo Vacuum) was cleaned with acetone (Wako) and isopropyl alcohol (IPA, Wako). Then, SU-8 3050 negative photoresist (Microchem) was spin-coated onto an ITO substrate for 60 s at 1000 rpm with a thickness of 100 m. The coated substrates were baked on a hotplate for 50 min at 95 oC. The substrates were exposed to UV through a custommade photomask. A reduced projection exposure system (MM-505, Nanometric Technology) was used for fabricating the photomasks. The photomask featured sixteen microcavities with a diameter of 200 or 500 m aligned in 4 x 4 matrix patterns. After baking on a hotplate for 10 min at 95 oC, the substrates were developed in SU-8 developer (Microchem) and rinsed with IPA. Because SU-8 resist have good insulating properties, the bottom of each microcavity, where the ITO layer was exposed, acted as the stimulation electrode. After fabrication of the microcavity-array pattern, a glass ring of 1 cm height was mounted onto a substrate with a silicone elastomer (KE-103, Shin-etsu Silicones). A stranded wire of cupper was also mounted onto a substrate with an electroconductive paste (Dotite FC-415, Fujikura Kasei). The ITO regions inside a glass ring, except the bottom surfaces of the microcavity, were insulated by coating with a silicone elastomer. The protocols for

fabricating the microcavity-array electrode substrate were shown in Fig. 1.

In this research, we used P19 cells as a model system for a stem cell. P19 is a mouse embryonal carcinoma cell line and can differentiate into three germ layers (Bain et al. 1994). For its easy handling and high neuronal differentiation rate, we have used P19 cells as a stem cell model. P19 cells were rountinely cultured and passaged every two days in alpha minimum essential medium (-MEM, Invitrogen) containing 10 % fetal bovine serum (FBS, HyClone) and 5 – 40 U/ml penicillin-streptomycin (Sigma-Aldrich). To produce uniform size EBs, P19 cells were collected and replated in a 96-well low cellattachment plate (Sumilon, Sumitomo Bakelite). Due to its spheroid bottom structure, a single EB was formed in each well. The obtained EBs were collected and transferred into

reported.

et al. 2009).

stimulation on the gene expression of P19 cells.

each microcavity of the electrode substrates. To our experience, initial plating of 500 cells resulted in a EB of 200 m diameter and 4000 cells in a EB of 500 m diameter at next day of plating. The microcavity-array electrode substrates were previously coated with MPC (2-methacryloyloxyethyl phosphorylcholine) polymer (Lipidure CM5206E, NOF corp.) to prevent nonspecific cell adhesion (Ishihara et al., 1999) and to localize EBs into each microcavity.

Fig. 1. Schematic of procedures for fabricating the microcavity-array electrode substrate using SU-8 negative photoresist.

Fig. 2. Schematic diagram of the experimental system. Size-controlled EBs were simultaneously stimulated through the microfabricated ITO substrates.

Toward the Precise Control of Cell Differentiation Processes by Using Micro and Soft Lithography 39

most EBs in the same observing field showed similar responses. In 200 m diameter EBs, most of cells in an EB showed calcium transients responded to electrical pulse. In 500 m diameter EBs, cells in outer region of EB primarily showed calcium transients. In all stimulation experiments, we confirmed that similar results were obtained (n=5 for 200 m diameter patterns and n=3 for 500 m diameter patterns). The results indicated that the microcavityarray electrode device could stimulate size-controlled EBs simulataneously and uniformally.

Fig. 4. EBs of P19 cells inserted in the microcavity-array structures (phase-contrast image) and stimulus-evoked intracellular calcium transients (fluorescence image). (Takayama et al.,

In this experiment, by using SU-8 negative thick photoresist, we could fabricate microcavityarray structures onto electrode substrate with enough depth (100 m) to trap and localize P19 EBs. By using the electrode substrate, we could stimulate P19 EBs simultaneously. Although several studies have carried out electrical stimulation to EBs, such as a field pacing (Sauer et al, 1999 and 2005; Yamada et al., 2007), there have been no studies that observed cell activity in a large number of EBs simultaneously. This may be due to difficulty in observing randomly floating EBs under suspension culture condition. Thus, localizing EBs in specific regions and detecting cell activity of EBs simultaneously by using microcavity-array structures are important bases for further electrical stimulation

2009)

experiments and analysis.

After insertion of P19 EBs, constant voltage stimulation was applied to the trapped EBs via the bottom surfaces of microcavities with an electrical stimulator (SEN-8203, Nihon Kohden) and an isolator (SS203J, Nihon Kohden). A platinum (Pt) electrode ( 1 mm) was used as the counter electrode during electrical stimulation. A single negative-first biphasic pulse with intensity of 5 V and duration of 1 ms was used for stimulation. Stimulation-induced responses of EBs were visualized by calcium imaging technique. The EBs were labelled with a calcium indicator Fluo-4AM (Molecular Probes) and fluorescence signals were detected with a cooled CCD camera (C8800-21C, Hamamatsu Photonics) mounted on an inverted microscope (IX-71, Olympus). The frame rate of 0.5 frame/s was used. The recording solution contained 148 mM NaCl, 2.8 mM KCl, 2 mM CaCl2, 1 mM MgCl2, 10 mM HEPES and 10 mM glucose. The overview of the experimental system was shown in Fig. 2.

Figure 3 showed photographs of the fabricated substrate and magnified phase-contrast images of the microcavity-array region and the trapped EBs. These microcavity-array patterns were fabricated with fine reproducibility. The EBs of P19 cells prepared by using spheroid-bottom plates also showed high uniformity in size and morphology. Figure 3-c showed the EBs inserted in the microcavities of 500 m diameter. The EBs were successfully trapped within each microcavity with 100 m depth. We confirmed that pre-coating with the MPC polymer inhibited EB adhesion to the device surface over several hours. Thus, suspension culture condition was maintained during electrical stimulation experiments. We also confirmed that pre-coating with the MPC polymer did not affect the electrical properties (impedance, phase) of the device by observing with an LCR meter.

Fig. 3. A microcavity-array dish for ensemble stimulation. (Takayama et al., 2009)

We then applied electrical pulses to the trapped EBs and recorded evoked response by calcium imaging. Figure 4 showed phase-contrast images of EBs in the microcavities and corresponding fluorescence images after applying a single biphasic pulse. The fluorescence images represent the normalized difference ratios in fluorescent intensity between pre- and post-stimulus. Spontaneous calcium transients in cells of P19 EBs were rarely observed before stimulation. By applying electrical stimulation, in contrast, significant elevations of intracellular calcium concentration were observed. For both of microcavity-array patterns,

After insertion of P19 EBs, constant voltage stimulation was applied to the trapped EBs via the bottom surfaces of microcavities with an electrical stimulator (SEN-8203, Nihon Kohden) and an isolator (SS203J, Nihon Kohden). A platinum (Pt) electrode ( 1 mm) was used as the counter electrode during electrical stimulation. A single negative-first biphasic pulse with intensity of 5 V and duration of 1 ms was used for stimulation. Stimulation-induced responses of EBs were visualized by calcium imaging technique. The EBs were labelled with a calcium indicator Fluo-4AM (Molecular Probes) and fluorescence signals were detected with a cooled CCD camera (C8800-21C, Hamamatsu Photonics) mounted on an inverted microscope (IX-71, Olympus). The frame rate of 0.5 frame/s was used. The recording solution contained 148 mM NaCl, 2.8 mM KCl, 2 mM CaCl2, 1 mM MgCl2, 10 mM HEPES

and 10 mM glucose. The overview of the experimental system was shown in Fig. 2.

properties (impedance, phase) of the device by observing with an LCR meter.

Fig. 3. A microcavity-array dish for ensemble stimulation. (Takayama et al., 2009)

We then applied electrical pulses to the trapped EBs and recorded evoked response by calcium imaging. Figure 4 showed phase-contrast images of EBs in the microcavities and corresponding fluorescence images after applying a single biphasic pulse. The fluorescence images represent the normalized difference ratios in fluorescent intensity between pre- and post-stimulus. Spontaneous calcium transients in cells of P19 EBs were rarely observed before stimulation. By applying electrical stimulation, in contrast, significant elevations of intracellular calcium concentration were observed. For both of microcavity-array patterns,

Figure 3 showed photographs of the fabricated substrate and magnified phase-contrast images of the microcavity-array region and the trapped EBs. These microcavity-array patterns were fabricated with fine reproducibility. The EBs of P19 cells prepared by using spheroid-bottom plates also showed high uniformity in size and morphology. Figure 3-c showed the EBs inserted in the microcavities of 500 m diameter. The EBs were successfully trapped within each microcavity with 100 m depth. We confirmed that pre-coating with the MPC polymer inhibited EB adhesion to the device surface over several hours. Thus, suspension culture condition was maintained during electrical stimulation experiments. We also confirmed that pre-coating with the MPC polymer did not affect the electrical most EBs in the same observing field showed similar responses. In 200 m diameter EBs, most of cells in an EB showed calcium transients responded to electrical pulse. In 500 m diameter EBs, cells in outer region of EB primarily showed calcium transients. In all stimulation experiments, we confirmed that similar results were obtained (n=5 for 200 m diameter patterns and n=3 for 500 m diameter patterns). The results indicated that the microcavityarray electrode device could stimulate size-controlled EBs simulataneously and uniformally.

Fig. 4. EBs of P19 cells inserted in the microcavity-array structures (phase-contrast image) and stimulus-evoked intracellular calcium transients (fluorescence image). (Takayama et al., 2009)

In this experiment, by using SU-8 negative thick photoresist, we could fabricate microcavityarray structures onto electrode substrate with enough depth (100 m) to trap and localize P19 EBs. By using the electrode substrate, we could stimulate P19 EBs simultaneously. Although several studies have carried out electrical stimulation to EBs, such as a field pacing (Sauer et al, 1999 and 2005; Yamada et al., 2007), there have been no studies that observed cell activity in a large number of EBs simultaneously. This may be due to difficulty in observing randomly floating EBs under suspension culture condition. Thus, localizing EBs in specific regions and detecting cell activity of EBs simultaneously by using microcavity-array structures are important bases for further electrical stimulation experiments and analysis.

Toward the Precise Control of Cell Differentiation Processes by Using Micro and Soft Lithography 41

(a) Cell-nonadhesive electrode substrate using SU-8 photoresist. (b) Cell-adhesive electrode substrate using PDMS elastomer.

Fig. 5. Structures of the electrode sustrates for stimulating EBs

rpm with a thickness of 100 m. After exposure and development with the photomasks which had inverted patterns as used in section 2, column-array structures were formed on a glass substrate. In this experiment, 289 (17 17 matrix pattern) and 64 columns (8 8 matrix pattern) were fabricated for 200 and 500 m diameter microcavities, respectively. Before pouring PDMS, a thin layer of photoresist (OFPR-800, positive photoresist, Tokyo Ohka) with a thickness of 4 m was formed onto the substrate by spin-coating for 40 s at 2000 rpm and then baking for 30 min at 80 oC. The thin photoresist layer worked as a sacrifice layer to release a cured PDMS sheet from the substrate (Nam et al., 2006). Then, the mixture of PDMS pre-polymer and catalyst (10:1 ratio, Silpot 184, Dow Corning) was spin-coated on the substrate for 80 s at 2500 rpm and 40 s at 2000 rpm, for 200 and 500 m diameter microcavites, respectively. The PDMS coated substrate was put on a hotplate and then a premanufactured thick PDMS annulus (20 mm square and 3 mm thickness) which had a 15 mm diameter hole was immediately attached on the substrate. The PDMS layer and annulus

It is widely recognized that intracellular calcium dynamics play important roles in regulation of cell differentiation processes (Dolmetsch et al., 1998; Yamada et al., 2007). Particularly, Spitzer et al. reported that spatio-temporal patterns of intracellular calcium transients regulated neuronal differentiation processes, such as extension of neurite and selection of neurotransmitter (Spitzer et al., 2004). Thus, inducing and controlling intracellular calcium transients in undifferentiated stem cells by applying electrical stimulation, which parameters are flexibly determined, is promising method to affect cell differentiation processes artificially. Optimizing parameters of electrical stimulation to induce desirable spatio-temporal patterns of intracellular calcium transients will be next step for controlling cell differentiation processes.

There are several problems, however, with the cell-nonadhesive electrode substrate using SU-8 thick photoresist. In the experiments, we manually transferred EBs one by one from 96 well culture plates into each microcavity using micropipettes. Furthermore, due to the cellnonadhesiveness of the electrode substrate by the MPC polymer coating, mild mechanical perturbations by moving the electrode substrate easily remove EBs from the microcavities. These problems make it difficult to stimulate more than 20 EBs simultaneously with the SU-8-based cell non-adhesive electrode substrates and to perform subsequent gene expression analysis and culture test after stimulation. On the basis of these disadvantages, we attempt to fabricate PDMS (poly(dimethylsiloxane))-based cell-adhesive electrode substrate to stimulate more number of EBs. The experimental procedures and present results for the celladhesive electrode substrate are summarized in a next section.
