2.1.

Cell Culture

Human epithelial carcinoma cells (A431 cells) were cultured in Dulbecco’s modified Eagle’s medium supplemented with 10% fetal bovine serum under a humidified atmosphere of 95% air and 5% ${\mathrm{CO}}_2$ at 37°C. The cells were harvested and seeded as a monolayer in glass-bottom culture dishes for the experiments.

2.2.

Cell Membrane Perforation

Spherical protein A bound polylactic acid (PLA) microspheres of 2 μm in diameter were mixed with anti-epidermal growth factor receptor mouse monoclonal antibody (Thermo Fisher Scientific, Fremont, California). The mixture was stirred for 25 min at room temperature, and then the unbound antibody was removed by centrifugation for 10 min at 10,000 rpm. The bound PLA microspheres were resuspended in phosphate-buffered saline (PBS) and added to A431 cells. The cells were incubated for 40 min under a humidified atmosphere of 95% air and 5% ${\mathrm{CO}}_2$ at 37°C and washed three times with PBS in order to remove the unconjugated microspheres. A Ti:sapphire chirped pulse amplification laser system (Libra, Coherent, Santa Clara, California), which generates 80-fs laser pulses at an 800-nm central wavelength, was used in the experiments. The laser beam was weakly focused by using a plano-convex lens ($f=200\mathrm{mm}$) to a laser spot size of 300 μm which was determined by full-width-at-half-maximum (FWHM) of the peak intensity of the Gaussian beam. Many cells can be treated simultaneously because a number of cells were inside the laser spot. The cells were irradiated by a single or multiple shot(s) of polarized (linearly or circularly) laser pulse from the top side of the cells. The PLA microsphere works as a microlens, and the optical intensity was enhanced by 9.7 at 870 nm under the PLA microsphere calculated by the three-dimensional finite-difference time-domain method. The optical intensity distribution under the PLA microsphere is described in our previous paper.¹⁸ The fluorescein isothiocyanate (FITC)-dextran solution or the green fluorescent silica particle suspension was added to the cells just before the laser irradiation. Two minutes after the irradiation, the cells were washed three times with PBS. Fluorescent molecules delivered into the cells were observed by using a fluorescence microscope (Eclipse Ti-E, Nikon, Tokyo, Japan). The delivery rate was evaluated by counting the number of cells inside the laser spot, and it was calculated by the number of cells inside the laser spot showing fluorescence divided by the number of cells inside the laser spot. Only the cells attached to the dish were counted. All values in the figures were expressed as $\text{means}\pm \text{standard}\text{deviation}$.

2.3.

Dependence of Delivery Rate on the Ambient Temperature

Simultaneous application of mild heat may influence cell membrane permeabilization when a physical method for drug delivery was applied.¹⁹ The effect of cell membrane fluidity on the delivery rate was evaluated by changing the ambient temperature of the cells.

After the removal of the unconjugated PLA microspheres from the cells, the culture dish was placed on a thermo plate (Tokai Hit Company, Shizuoka, Japan). The temperature of the medium was controlled by the heating stage and monitored by a thermocouple thermometer. During the perforation experiment, the ambient temperature was kept at 23, 37, or 40.5°C. The FITC-dextran (4 kDa and 2 MDa; Sigma, St. Louis, Missouri) was used to evaluate the efficiency of delivery. The FITC-dextran solution (0.5 mM for 4 kDa and 0.001 mM for 2 MDa) was added to the cells just before the linearly polarized fs laser irradiation. The cells were irradiated by a single shot of fs laser pulse.

2.4.

Dependence of Delivery Rate on the Number of Irradiated Laser Pulses

The interaction between a fs laser pulse and cell membrane is governed by the peak laser intensity. Although high peak intensity is necessary for nonlinear optical interactions, excess laser intensity may lead to cell damage. To realize a cell membrane perforation with lower laser intensity, we have conducted an experiment using multiple laser pulses.

The FITC-dextran (4 kDa and 2 MDa) and the green fluorescent silica particles (plain or functionalized, 100 nm in diameter) were used to evaluate the efficiency of delivery. They were added to the cells at 23°C just before the circularly polarized fs laser irradiation (1, 2, 5, and 10 pulses). The repetition rate of the laser system was 1 kHz, and the number of shots was controlled by using a digital delay generator (DG535, Stanford Research Systems, Sunnyvale, California) to control the Pockels cell inside the laser system. The laser fluence was 1.06 or $0.53\mathrm{J}/{\mathrm{cm}}^2$ at which the corresponding peak laser intensity under the microsphere is estimated to be $1.29\times {10}^{14}$ or $6.43\times {10}^{13}\mathrm{W}/{\mathrm{cm}}^2$, respectively.¹⁸

2.5.

Dependence of Delivery Rate on the Delivered Particle Size

Different sizes of spherical particles were delivered into the cells to investigate and discuss the size of the pores formed on the cell membrane.

The green fluorescent silica particles (diameters of 30, 50, 70, 100, 200, and 500 nm) (sicastar, Micromod, Rostock-Warnemünde, Germany) were used to evaluate the size-dependent efficiency of delivery. After the removal of the unconjugated PLA microspheres, the silica particle suspension ($2.5\mathrm{mg}/\mathrm{ml}$) was added to the cells at 23°C just before the circularly polarized fs laser irradiation. The laser fluence was $1.06\mathrm{J}/{\mathrm{cm}}^2$.

2.6.

Dependence of Delivery Rate on the Charge of Delivered Particles

Since the cell membrane is negatively charged, the charge of molecules could influence the delivery rate when using physical methods to enhance the cell membrane permeability.^21,22 The charged silica particles were delivered into the cells to investigate the influence of the charge on the delivery rate by the biodegradable microsphere-mediated perforation using a fs laser.

The green fluorescent silica particles of 100 nm in diameter with functionalized surface were used to evaluate the efficiency of delivery. Aminated or carboxylated particles (sicastar, Micromod) were chosen as they are positively and negatively charged, respectively. After the removal of the unconjugated PLA microspheres from the cells, the silica particle suspension ($2.5\mathrm{mg}/\mathrm{ml}$) was added to the cells at 23°C just before the circularly polarized fs laser irradiation. The laser fluence was $1.06\mathrm{J}/{\mathrm{cm}}^2$.

3.1.

Ambient Temperature Dependence

Figure 1 shows the dependence of the delivery rate of FITC-dextran on the ambient temperature of the cells. Delivery rates with both 4 kDa and 2 MDa FITC-dextran had no statistically significant difference depending on the temperature. In this experiment, the cells were heated to the specified temperatures before and during the laser irradiation, in which the total heating time was less than 3 min. The 4-kDa FITC-dextran was delivered with higher rates compared with the FITC-dextran of 2 MDa. This result is probably related to the size of the pores formed on the cell membrane, which allowed the smaller FITC-dextran to be delivered and made it difficult for larger FITC-dextran to cross the pores. We have not investigated the effect of the temporal heating dose in this study, but it would affect the survival rate of the cells if the heating time is long. Figure 1 also shows the dependence of the delivery rate on the laser fluence. The delivery rate increased with increasing laser fluence. The peak intensity under the sphere is determined by multiplying the incident laser intensity by the enhancement factor under the PLA sphere. Thus, the figure also indirectly shows the effect of the enhancement factor.

Fig. 1

Dependence of delivery rate on cell’s ambient temperature at different laser fluences. Delivery rate was evaluated by using fluorescein isothiocyanate (FITC)-dextran of 4 kDa (green) and 2 MDa (blue) at the ambient temperature of 23°C (diamond), 37°C (square), and 40.5°C (triangle).

3.2.

Dependence the Delivery Rate of FITC-Dextran on the Number of Irradiated Pulses

Figure 2 shows the phase-contrast and fluorescence images of the cells after the laser irradiation at the laser fluence of $1.06\mathrm{J}/{\mathrm{cm}}^2$ with different numbers of pulses. Cell detachment can be observed in the center of the laser irradiated area when irradiating five or more pulses at $1.06\mathrm{J}/{\mathrm{cm}}^2$, which resulted in difficulties of the evaluation of the delivery rate. We considered that the detached cells are dead and that the laser fluence, which leads to the detachment of the cells, should not be applied to drug delivery system (DDS) applications including in vivo therapy. Several cells near the irradiated area showed fluorescence which is probably due to the low-laser intensity outside the laser spot of the Gaussian beam profile. Figure 3 shows the delivery rate of FITC-dextran with different numbers of pulses at two different laser fluences. The delivery rates for control experiments, i.e., delivery rates without laser irradiation, were $0.5\pm 0.5\%$ and $0.3\pm 0.6\%$ for FITC-dextran of 4 kDa and 2 MDa, respectively. The delivery rate increased with the increasing number of laser pulses. Cells were attached to the bottom of the dish after the irradiation of 10 pulses at $0.53\mathrm{J}/{\mathrm{cm}}^2$.

Fig. 2

(a–c) Fluorescence and (d–f) phase contrast images of A431 cells perforated by using antibody-conjugated polylactic acid (PLA) microspheres irradiated by femtosecond (fs) laser pulse(s) in the presence of 4-kDa FITC-dextran. The number of irradiated laser pulses were 1 (a, d), 2 (b, e), and 5 (c, f). The irradiated laser fluence was $1.06\mathrm{J}/{\mathrm{cm}}^2$ at the ambient temperature of 23°C. Dashed circles (300-μm diameter) indicate the laser-irradiated area.

Fig. 3

Dependence of delivery rate on the number of laser pulses. Delivery rate was evaluated by FITC-dextran of 4 kDa (diamond) and 2 MDa (circle). The irradiated laser fluence was $1.06\mathrm{J}/{\mathrm{cm}}^2$ (solid line) and $0.53\mathrm{J}/{\mathrm{cm}}^2$ (dashed line). The evaluation of the delivery rate was difficult for more than 5 pulses at $1.06\mathrm{J}/{\mathrm{cm}}^2$ or 50 pulses at $0.53\mathrm{J}/{\mathrm{cm}}^2$, since the cell detachment was observed in the center of the laser-irradiated area.

3.3.

Investigation of the Size of Pores on Cell Membrane

We investigated the size of the pores formed on the cell membrane by using different sizes of green fluorescent silica particles. An optical intensity distribution under a PLA microsphere varies depending on the polarization direction of the irradiated laser. In the case of silica substrate surface ablation, the hole fabricated by using dielectric particles had a different shape depending on the polarization of the incident laser.²³ As for the perforation of cell membrane, the delivery rate could be influenced by the polarization direction of the irradiated laser (especially for the delivery of silica particles with large diameter) which would form different shapes of pores. Therefore, we have investigated the dependence of the delivery rate on the laser polarization. Figure 4 shows the calculated optical intensity distributions under the PLA microsphere of 2-μm diameter in water irradiated by a plain 800-nm light. The optical intensity distribution of 5-nm under the microsphere and 940-nm under the microsphere, where the peak intensity is obtained, was elliptical (the long axis in the polarized direction) with the $x$-axis linear polarization [Figs. 4(a) and 4(b)] and circular with the circular polarization [Figs. 4(c) and 4(d)]. The FWHM of the optical distribution for linear polarization 940-nm under the microsphere was 610 nm along $x$-axis and 530 nm along $y$-axis, whereas that for circular polarization 940-nm under the microsphere was 570 nm along both $x$- and $y$-axes. However, as shown in Fig. 5, no significant difference in the delivery rate of FITC-dextran was obtained between circular and linear polarizations. Hereafter, we describe results by using circularly polarized laser. Figure 6 shows the dependence of delivery rate on the size of the delivering particles. The green florescent silica particles of different sizes were used in this study. Although the delivery rate had decreased as the diameters of the particles increased, silica particles with 100-nm diameter were able to be delivered into more than 20% of the cells in the laser-irradiated area.

Fig. 4

Optical intensity distributions on the $xy$ plane (a, c) 5-nm under the PLA microsphere and (b, d) on the peak intensity (940-nm under the microsphere) simulated by the three-dimensional finite-difference time-domain method. A plane wave is irradiated to the microsphere with the wave vector in the $z$-direction. The incident wave of 800 nm in wavelength is polarized (a, b) linearly and (c, d) circularly.

Fig. 5

Dependence of delivery rate on the laser polarization. The delivery rate was evaluated by FITC-dextran of 4 kDa and 2 MDa. A single shot of linearly or circularly polarized 80-fs laser pulse was irradiated at the laser fluence of $1.06\mathrm{J}/{\mathrm{cm}}^2$ at the ambient temperature of 23°C.

Fig. 6

Dependence of delivery rate on the diameters of silica particles. The delivery rate was evaluated by green fluorescent silica particles of the diameters of 30, 50, 70, 100, 200, and 500 nm. A single shot of 80-fs laser pulse was irradiated at the laser fluence of $1.06\mathrm{J}/{\mathrm{cm}}^2$ at the ambient temperature of 23°C.

3.4.

Delivery of Functionalized Green Fluorescent Silica Particles

Figure 7 shows the delivery rate of functionalized green fluorescent silica particles. The positively charged particles were able to be delivered at a higher rate than the particles without surface functionalization.

Fig. 7

Dependence of delivery rate on the surface functionalization of green fluorescent silica particles. The delivery rate was evaluated by the plain (not charged), aminated, or carboxylated green fluorescent silica particles with diameters of 100 nm. A single shot of 80-fs laser pulse was irradiated at the laser fluence of $1.06\mathrm{J}/{\mathrm{cm}}^2$. *$p<0.05$ and **$p<0.01$ versus “not charged.”

3.5.

Dependence of the Delivery Rate of Green Fluorescent Silica Particles on the Number of Irradiated Pulses

Figure 8 shows the delivery rate of functionalized green fluorescent silica particles with different number of pulses at the laser fluence of $0.53\mathrm{J}/{\mathrm{cm}}^2$. The delivery rate increased with the increasing number of irradiated pulses and was saturated at five or more pulses.

Fig. 8

Dependence of delivery rate of green fluorescent silica particles on the number of laser pulses. The delivery rate was evaluated by plain (diamond), aminated (triangle), and carboxylated (square) green fluorescent silica particles with diameters of 100 nm. The laser fluence was $0.53\mathrm{J}/{\mathrm{cm}}^2$.