Importance of Spatial Arrangement of Cardiomyocyte Network for Precise and Stable On-Chip Predictive Cardiotoxicity Measurement

Figures

• 

  Figure 1. Fabrication of three types of cardiomyocyte networks on a multielectrode chip. (A): Schematic drawing of fabrication procedure of agarose microchambers on a multi-electrode array (MEA) chip with photo-thermal agarose microfabrication technology. An MEA chip was coated with collagen and agarose layers. Then a portion of the agarose layer was melted by the spot heating of the focused infrared laser. (B): three shapes of three types of agarose microchamber structures for cardiomyocyte networks formed on an MEA chip. (a): schematic drawing of the agarose microchamber designs for cardiomyocyte network on an MEA chip. (b): a micrograph of the fabricated three shapes agarose microchambers on the one-quarter area of an MEA chip; large square sheet, small sheet, and square closed-loop-shaped sheet. (c): a micrograph of agarose microchambers on the same one-quarter area of an MEA chip two days after human embryonic stem (hES) cell-derived cardiomyocyte cultivation. Bar, 1 mm. (C): The size distribution of three types of agarose microchambers on an MEA chip. Bars and error bars represent the mean and S.D. of agarose microchamber areas, respectively. (n = 3).

• 

  Figure 2. Cultured cardiomyocyte networks in three patterns of agarose microstructures. (A): Representative micrographs of three different spatial cell arraignment types of hES cardiomyocyte networks cultured on an MEA chip for seven days. (a): cardiomyocyte network in small sheet shape agarose microchamber. (b): cardiomyocyte network in large square sheet shape agarose microchamber. (c): cardiomyocyte network in square closed-loop-shaped sheet shape agarose microchamber. Bars, 0.5 mm. (B): representative fluorescent micrographs of hES cardiomyocyte networks. B(a–c) correspond to stained of the nuclei of A(a–c), respectively. The nucleus of each cell was stained by nuclear staining fluorescent dye, DAPI, after the measurement of extracellular field potentials (FPs). Bars, 0.5 mm. (C): cell densities of hES cardiomyocyte networks in three patterns of agarose microchambers (n = 3 each). Bars and error bars represent mean values and S.D. in three patterns, respectively. No significant differences (N.S.) were observed between groups.

• 

  Figure 3. Experimental designs of extracellular field potential measurement of cardiomyocyte networks: Evaluation of hERG ion channel blocking with E-4031. (A): schematic drawing of the MEA measurement system. The extracellular field potentials (FPs) of cardiomyocytes on the microelectrodes were amplified and digitally recorded for FP waveform analysis. (B): time-course of drug administration procedure. The FPs of cardiomyocytes on a microelectrode were measured at seven different solution states in MEA chips: Free (only culture medium), Control (0.1% PBS dose), Low (1 nM E-4031 dose), Medium (10 nM E-4031 dose), High (100 nM E-4031 dose), Very high (1 μM E-4031 dose) and Wash-out (only culture medium). (C): representative FP waveforms of large square sheet cardiomyocyte network at every seven states. The red and green arrowheads indicate each FP waveform’s first and last FP peaks, respectively. The FPD prolongation, an increase of the time distance of those first and last FP peaks, was observed at the Medium state. The fibrillation-like weak and faster ISI waveform was observed at High state. An arrest of FP waveforms was recorded at Very high condition. (D): The evaluation of FPD…

• 

  Figure 4. Evaluation of hERG ion channel blocking on beating intervals in three type cardiomyocyte network patterns. (A): E-4031 concentration dependence of inter-spike intervals (ISIs) in three type cardiomyocyte networks: (a) small sheet (n = 3), (b) large square sheet (n = 3) and (c) square closed-loop-shaped sheet (n = 3). Bars, mean values. Error bars, S.D. Dunnett’s multiple comparison test, * p < 0.05, ** p < 0.01, and *** p < 0.001 vs. Control state. (B): Change in ISI in three cardiomyocyte network patterns: (a) Free to Control (n = 3 each), (b) Control to Low (n = 3 each), (c) Low to Medium (n = 3 each), (d) Medium to High (n = 3 each) and (e) Free to Wash-out (n = 3 each). Bars, mean values. Error bars, S.D. Tukey’s multiple comparison test, * p < 0.05. (C): Coefficient of variability (CV) of ISI values against E-4031 concentration. The red-filled circles, black-filled circles, and open circles indicate the mean CV values of the small sheet, large square sheet, and square closed-loop-shaped sheet group, respectively (n = 3 each). Dunnett’s multiple comparison test, ** p < 0.01 vs. Control state. (D): CV of ISI in three cardiomyocyte network patterns: (a) Control state (n = 3…

• 

  Figure 5. The beating frequency distribution of three type cardiomyocyte network samples at Control and Medium states. (A): The distribution of beating frequency (1/ISI, Hz) of three samples in Control state: (a–c) small sheet group, (d–f) large sheet group, and (g–i) closed-loop group. (B): The distribution of beating frequency in Medium state: (a–i) were the same cardiomyocyte network samples as shown in A(a–i).

• 

  Figure 6. Evaluation of hERG ion channel blocking on field potential duration time prolongation in three types of cardiomyocyte network patterns. (A): E-4031 concentration dependence of field potential duration (FPD) time: (a) small sheet (n = 3), (b) large square sheet (n = 3), and (c) large square closed-loop sheet (n = 3). Bars, mean values. Error bars, S.D. Dunnett’s multiple comparison test, * p < 0.05 and ** p < 0.01 vs. Control state. (B): Change in FPD in three cardiomyocyte network patterns: (a) Free to Control (n = 3 each), (b) Control to Low (n = 3 each), (c) Low to Medium (n = 3 each), (d) Medium to High (n = 3 each), and (e) Free to Wash-out (n = 3 each). Tukey’s multiple comparison test, * p < 0.05 and ** p < 0.01. (C): Short-term variability (STV) of FPD values against E-4031 concentration. The red, black, and open circles indicate mean STV values of the small, large square, and square closed-loop groups, respectively (n = 3 each). Dunnett’s multiple comparison test, * p < 0.05, ** p < 0.01, and *** p < 0.001 vs. Control state. (D): STV of FPD in three cardiomyocyte network patterns: (a) Control state (n = 3 each), (b) Low state (n = 3 each), (c) Medium state (n = 3 each),…

• 

  Figure 7. Cell orientation analysis of cardiomyocytes in the large square sheet and closed-loop sheet microchambers. (A): Fluorescent micrographs of hES cardiomyocytes in the large square agarose microchamber (a) and closed-loop agarose microchamber (b) on an MEA chip seven days after cultivation started. Rhodamine Phalloidin stained actin filaments in cardiomyocytes. Bar, 0.5 mm. (B): Cell orientation distribution in large square sheet (a) and square closed-loop-shaped sheet (b). Acquired images were divided into five (a) and four (b) areas, and their cell orientation distribution estimated by actin filament orientations in each cell were analyzed. (i–iv) represent the four sides of the large square sheet, and closed-loop-shaped sheet, (v) is the inner central area of the large square sheet.