Page 110 - IJB-4-1
P. 110

Formation of cell spheroids using Standing Surface Acoustic Wave (SSAW)

            Logan, UT, USA) containing 10% fetal bovine serum (FBS,   and compared. It is assumed that all individual cells were
            Gibco, Waltham, MA, USA) and 1% antibiotic-antimycotic   distributed uniformly across the PDMS cavity and do not
            solution, including 10,000 units/mL of penicillin, 10,000   gather before reaching the pressure node. In this simulation,
            µg/mL of streptomycin, and 25 µg/mL of amphotericin B   the  motion  of  cells  and  the  time  required  to  reach  the
            (Gibco), in a cell culture flask (t75, ThermoFisher Scientific).   equilibrium state are highly dependent on the equivalent
            The cells were incubated at 37 °C in a humidified incubator   force applied to them and their initial location. It is found
            (Heracell 150i, ThermoFisher Scientific) under the condition   that the trajectory motion of cells in the SSAW field can
            of 5% CO 2 . The culture medium was changed every two   be fitted by an exponential rise curve and the rising rate is
            or three days depending on the initial seeding. Achieving   dependent on the initial distance to the pressure node and
            80%  confluence,  the  cell  was  dissociated  using  0.25%   acoustic  operating  parameters,  such  as  the  excitation
            Trypsin 1 mM EDTA.4Na (Lonza, Basel, Switzerland),   frequency and power (see Figure 2). The correlationbetween
            centrifuged at 1,000 RPM (SL 8 small benchtop centrifuge,   acoustics  parameters  (e.g.,  excitation  frequency,  power
            ThermoFisher Scientific) for 5 min at room temperature,   output) and cell motion by SSAW was listed in Table 2.
            and subsequently re-suspended in the culture medium in   Firstly, the cell motion across the cavity by either low- and
                                6
            a concentration of 210  cells/mL and a volume of about   high-frequency SSAW at different initial positions is shown
            400 µL. Cell density was estimated using hemocytometer   in Figures 2A and B. It is clear that using the high-
            (Hausser scientific hemocytometer, ThermoFisher Scientific).   frequency  excitation  could  accumulate  the  cells  much
            Live/dead  cell  viability  assays  (L3224,  L/D  kit  for   more quickly. The effects of output power and cell diameter
            mammalian cells, ThermoFisher Scientific) consisting of   on the trajectory motion of cell were also investigated if
            calcein-AM and ethidium homodimer-1 were used to stain
            the cells. The samples in 5 random areas were captured   the distance between the initial position and pressure node
            by the optical microscope and processed with ImageJ using   is fixed as 42 μm which is one-quarter of wavelength or
            the established protocols [43,44]  to count the live and dead   the distance from anti-pressure node to adjacent pressure
            cells  stained  in  green  and  red,  respectively.  The  cell   node at the high-frequency excitation. Referring to Eq.2,
            spheroids were then cultured in ultralow attachable culture   acoustic radiation force is proportional to the volume of
                            ®
            dish (#3262 Corning , Thermo Fisher Scientific) to minimize   the cell (or cube of cell diameter in the shape of a sphere)
            the cell attachment. The spheroid size and cell viability were   and the power (or square of acoustic pressure). Large cells
            measured daily for 7 days [36,45] .                reach the pressure node in a short time because of large
                                                               acoustic radiation force applied to them (see Figure 2C).
            3. Results and Discussion                          At the high-frequency excitation, the cells in a diameter of 8
                                                               μm, 10 μm, and 15 μm at the acoustic excitation power of
            3.1 Numerical Simulation of Cell Motion by SSAW    1.0 W reach the pressure node after 6.26 s, 4.01 s, and 1.78 s,
            Using a network analyzer (HP8510B, Agilent Technologies,   respectively. In comparison, the corresponding values at
            Santa  Clara,  CA,  USA),  the  S 12   frequency  response  of   the low-frequency excitation are 13.58 s, 8.70 s, and 3.87 s,
            IDTs (transmission coefficient) shows several peaks [42] .   respectively, almost twice as those at the high frequency. In
            Trajectories of biological cells excited by low-frequency   addition, the motion time required to reach the pressure
            (10.4 MHz) and high-frequency (23.8 MHz) were simulated   node also decreases with the output power (see Figure 2D).

                                                                (B)               IDTs


                      (A)                                 Individual
                                                             cells
                           Function
                           Generator   Power Amplifier   Device
                                                                         PDMS




                                                           Cell
                                                         spheroid

            Figure 1. (A) Schematic diagram of experimental setup of forming cell spheroids by SSAW and (B) zoomed photo showing two
            pairs of interdigital transducers (IDTs) and PDMS cavity.

            4                               International Journal of Bioprinting (2018)–Volume 4, Issue 1
   105   106   107   108   109   110   111   112   113   114   115