2.1.

Cell Culture and Preparation

RBL-2H3 and HeLa cells (both from ATCC Company, USA) were grown in minimal essential medium (MEM) with Earle’s salt and Dulbecco’s Modified Eagles Medium (DMEM), respectively, supplemented with 10% fetal bovine serum and 2% L-glutamine (all from Gibco, USA) in an incubator with a humidified atmosphere (5% $\mathrm{C}{\mathrm{O}}_2$ ) at $37°\mathrm{C}$ . Cells were seeded on coverslips at an appropriate density $12\mathrm{h}$ before the imaging experiment.

A23187 was purchased from Sigma (Saint Louis, Missouri) and dissolved in DMSO first, then diluted with PBS to prepare the stock solution (the maximal concentration of dimethyl sulfoxide (DMSO) $<0.1\mathrm{\%}$ ). AgNP (SARPU $50\mathrm{KW}$ , 5% in water) was purchased from ABC Nanotech, Korea, and 50-fold diluted in water to prepare the $1\mathrm{mg}∕\mathrm{mL}$ of stock solution.

2.2.

Cell Volume, Surface Area, and Cross Sectional Profile Measurements with Quantitative Phase Microscopy

We have used a QPM setup, which is same as the one used in our previous paper.⁹ As is shown in the schematic diagram in Fig. 1 , it consists of a monochromatic laser source $(\lambda =633\mathrm{nm})$ , an inverted microscope system (IX51, Olympus), and a CCD (IPX-VGA210-L, Imperx Incorporated). A sample was imaged on the first image plane through a $20\times ∕0.5$ objective lens. A $110\text{-grooves}∕\mathrm{mm}$ transmission grating (G) is placed on the first image plane to create a common path interferometer utilizing the zero-order and first-order diffraction beams. High frequency conponents of the zero-order beam are filtered out with a pinhole to generate a reference light in the interferometer. The low-pass filtered zero-order beam and the first-order beam are interfered on the second image plane of the setup, and a CCD is used to take spatial interferogram data. Two lenses L2 and L3 form a 4f system with $1∕2$ magnification. The other two lenses L4 and L5 form another 4f system with fives times the magnification. The final optical magnification of a sample at the CCD plane is calibrated to be about 50 for our case. The CCD has a frame rate of $200\mathrm{fps}$ at a full resolution of $640\times 480\text{pixels}$ . The pixel-to-pixel distance of the CCD is $7.4\mu \mathrm{m}$ .

Fig. 1

Schematic diagram of QPM setup. For details, refer to the text.

The cross correlation interferogram of a cell in the spatial domain is measured. The spatial interferogram can be generally expressed as

RBL-2H3 cells on a coverslip were washed three times with PBS buffer and then sealed tightly in a special chamber filled with PBS (Gibco, Invitrogen, Carlsbad, California). With this chamber, we can take images while various chemicals are injected into the chamber with syringe pumps (KDScientific, Holliston, Massachusetts). After the chamber was mounted on the microscopy stage, images of a cell without any treatment (“blank”) were captured every $5\mathrm{s}$ for $7\min$ . Then $50\mu \mathrm{L}$ of PBS was injected into the chamber within $2\min$ , and images were taken continuously for another $7\min$ with an interval of $5\mathrm{s}$ . The same process was performed again by replacing PBS with A23187 (the final concentration of $5\mu \mathrm{g}∕\mathrm{mL}$ ) or AgNP (the final concentration of $50\mu \mathrm{g}∕\mathrm{mL}$ ) for $12\min$ if not specified. The measured volume of a cell at each time point was divided by the average volume of a “blank” period for normalization. The corresponding cell surface area was normalized with the same method. To further validate the linkage of cell volume and degranulation, we replaced PBS in the cell chamber with DPBS (Gibco) plus $5\mathrm{mM}$ of ethylenediaminetetraacetic acid tetra sodium salt dihydrate (EDTA) (Sigma, Saint Louis, Missouri), and repeated the volume measurement of RBL-2H3 cells treated with A23187 or AgNP. The transmission electron microscopy (TEM) image of AgNP used in our study was shown in Fig. 2 .

Fig. 2

TEM image of AgNP used in our experiments. This image was provided by ABC Nanotech Company with permission. The average diameter is about $15\mathrm{nm}$ .

2.3.

Fluorescent Imaging of Intracellular Calcium Concentration Changes

RBL-2H3 cells on a coverslip were washed with PBS buffer and then incubated with $10\text{-}\mu \mathrm{M}$ Fluo-3/AM (Molecular Probe, Invitrogen) at $37°\mathrm{C}$ for $30\min$ . After washing three times, these stained cells were sealed in a chamber for fluorescence microscpy measurements. Procedures to inject AgNP into a sample are the same as those used in QPM experiments. Fluorescence images of ${\left[{\mathrm{Ca}}^{2+}\right]}_{\mathrm{i}}$ were acquired by a fluorescence microscope with a $40\times$ $(\mathrm{NA}=0.60)$ objective lens (IX-51, Olympus, Japan). Blue light $\left(460\text{to}490\mathrm{nm}\right)$ filtered from an attached mercury lamp in our microscope was selected for excitation. Fluorescence signals were recorded by an attached CCD detector (Retiga EXi FAST 1394, Qimage) with a $510\text{to}540\text{-}\mathrm{nm}$ bandpass filter. In this study, we captured a series of fluorescence images of ${\left[{\mathrm{Ca}}^{2+}\right]}_{\mathrm{i}}$ every $400\mathrm{ms}$ over $2\min$ , each using an exposure time of $300\mathrm{ms}$ , and then processed with the Image-Pro Plus (version 5.1) software (Media Cybernetics, Bethesda, Maryland).

2.4.

Histamine Measurement

Histamine release from RBL-2H3 cells was measured with the method described in Ref. 4 with slight modification. Briefly, cells were seeded in a 96-well plate to reach 80% confluence and were washed with Hank’s buffer. Then, $150\mu \mathrm{L}$ of Hank’s buffer, AgNP $(50\mu \mathrm{g}∕\mathrm{ml})$ , or A23187 $(5\mu \mathrm{g}∕\mathrm{ml})$ (four wells for each group) was added, respectively. After the indicated time of treatment, supernatants were collected and reacted with a highly specific reagent, ortho-pthaldialdehyde (OPA, Sigma, Saint Louis, Missouri) ( $0.1\mathrm{mM}$ , $150\mu \mathrm{L}$ ) for another $20\min$ to determine the histamine content in supernatants (supernatant). The fluorescence intensity of the supernatant was measured at the excitation/emission wavelength of $352∕430\mathrm{nm}$ with a microplate spectrofluorometer (Spectramax XS, Molecular Devices, Silicon Valley, California). Cells from another four wells in parallel were lyzed with 0.1% Triton X-100, and the total histamine content of extracts was measured (total). The percentage of histamine released into the supernatant was calculated by using the following formula: $\text{release}\mathrm{\%}=(\text{supernatant}∕\text{total})\times 100$ .

2.5.

Statistical Analysis

Fluorescence intensities of OPA-histamine complex in each group were expressed as $\text{mean}\pm \text{standard}$ error of the mean (SEM). Statistical comparisons between treated groups and the control one were determined by using one-way analysis of variance (ANOVA), and significant difference was accepted when $p<0.05$ .

3.1.

Volume, Surface Area, and Cross Sectional Profile Changes of an RBL-2H3 Cell by A23187

We have monitored the volume of a representative RBL-2H3 cell using QPM without any treatment (blank). The volume of the RBL-2H3 cell changes very little within $\pm 3\mathrm{\%}$ of its mean value, and it shows that the speed of volume change is very slow unless it undergoes a degranulation process. Figure 3 illustrates the volume variation of the RBL-2H3 cell before and after A23187 was injected into the chamber with many RBL-2H3 cells. The cell volume is normalized by its mean value taken from $t=0\min \text{to}t=7\min$ before the cell is exposed to the degranulation stimulus A23187. A23187 is added at around $t=7\min$ , and the exact addition time point is indicated with a small arrow. The normalized volume of the RBL-2H3 cell increases sharply to the peak value (about 20% increment) just after the addition of A23187, then it decreases slightly later. Figure 3 displays changes in the surface area of the RBL-2H3 cell sample on the addition of A23187. It shows that variations in the surface area of the cell follow similar patterns to those in the cell volume shown in Fig. 3. Figure 3 shows actual phase images for the RBL-2H3 cell at $t=4$ , 7, 9, 13, 17, and $21\min$ . The total magnification for these images was $50\times$ . Cross sectional line profiles of the cell at different time points across the white dashed line in the upper left phase image of Fig. 3 are shown in Fig. 3. Although the shapes of line profiles at $t=7\min$ and $t=9\min$ look very similar, the peak height in the line profile at $t=9\min$ is a little bit higher. We can see that another peak appeared on the left side of the original peak in the line profile taken at $t=13\min$ . The distance between two arrows in the horizontal axis of the line profile plot at $t=4\min$ in Fig. 3 indicates the lateral width of a cell. These two arrows were taken by the edge detection algorithm used in our image analysis method described in Ref. 9. The width of the cell increased slightly with time in Fig. 3. Even though it is not obviously recognizable because of the diffraction limited spatial resolution of our optical imaging system, phase images in Fig. 3 imply that the cell became granulated after $7\min$ .

Fig. 3

Variations in the volume and the surface area of an RBL-2H3 cell before and after the addition of A23187. (a) Normalized cell volume and (b) normalized cell surface area with respect to time. A small arrow indicates the time when A23187 was injected. (c) Cross sectional line profiles and (d) phase images of the RBL-2H3 cell for discrete time points. Line profiles correspond to the height profiles of the sample cell. The color bars indicate the phase in radians. (Color online only.)

We did a control experiment for the degranulation process of an RBL-2H3 cell by replacing A23187 with PBS. Since PBS is a commonly used buffer solution, it is not supposed to stimulate degranulation to any cell. Figure 4 shows the volume change of a representative RBL-2H3 cell before and after PBS is injected to the cell chamber containing many RBL-2H3 cells. The vertical axis of the plot was normalized with the average volume of the cell within our observation period shown in the plot. We have followed the same volume measurement procedures used for Fig. 3. As expected, our experimental results in Fig. 4 show that there was no effect of PBS on the volume of the RBL-2H3 cell. An arrow near $t=7\min$ is the time point when PBS is injected into the chamber containing RBL-2H3 cells. Results shown in Fig. 4 indicate that PBS does not activate degranulation to the RBL-2H3 cell.

Fig. 4

Normalized cell volume for an RBL-2H3 cell as a function of time. A small arrow at $t=7\min$ indicates the time point when PBS was injected.

Next, we have used PBS and A23187 to test whether they might stimulate a degranulation process in HeLa cells. Since a HeLa cell is not a mast cell or granulocytes (neutrophils, eosinophils), it is not supposed to degranulate upon any external stimulus. Volume change in a representative HeLa cell was monitored with an interferometric QPM method, which was used in Figs. 3 and 4, while A23187 and PBS were injected into a cell chamber containing many HeLa cells. Each vertical axis in Fig. 5 was normalized by the average volume of each sample within its observation period. Figures 5 and 5 show that there is no significant cell volume change for the HeLa cell sample when it is exposed to either A23187 or PBS. Arrows near $t=7\min$ in Figs. 5 and 5 represent the time points when PBS and A23187 are injected to the chamber, respectively. We believe that small volume changes just after the addition time points in Figs. 5 and 4 are due to the fluctuations of the sample associated with the addition process. These small fluctuations last less than $2\min$ or so after the addition of PBS or A23187. The average cell volume is unchanged within the measurement error of the cell volume, which is about 2% for Fig. 5 and 3% for Fig. 4. Results shown in Fig. 5 indicate that neither PBS nor A23187 can induce significant change in HeLa cell volume.

Fig. 5

Normalized cell volume for a HeLa cell as a function of time before and after (a) PBS or (b) A23187 is injected. Small arrows at $t=7\min$ indicate the time points when PBS or A23187 were injected.

3.2.

Volume, Surface Area, and Cross Sectional Profile Changes of an RBL-2H3 Cell by AgNP

To find the effect of AgNP on the degranulation process of a mast cell, we have first monitored changes in the volume of a representative RBL-2H3 cell with QPM when it was exposed to $50\mu \mathrm{g}∕\mathrm{mL}$ of AgNP. Our measured results are displayed in Fig. 6 . The cell volume is normalized by its mean value taken for $7\min$ just before the addition of AgNP into the cell chamber. The addition time point is about $7\min$ . It shows that the volume was almost doubled immediately after AgNP was injected into the chamber, and it lasted more than $3\min$ . After $3\min$ , the cell volume was recovered to its initial volume. Similar changes in cell surface area were observed as well and are displayed in Fig. 6. Figure 6 shows actual phase images for the RBL-2H3 cell at $t=4$ , 7, 9, 11, 15, and $19\min$ . Again, the total magnification for these images was $50\times$ . Cross sectional line profiles corresponding to the phase images in Fig. 6 were obtained across a white dashed line in the top left image of Fig. 6. Figure 6 shows these cross sectional line profiles. Only one main peak was found in the cross sectional line profiles measured at $t=4$ and $7\min$ . However, several comparable peaks appeared in the line profile measured at $t=9\min$ , where the cell was swollen after AgNP was added.

Fig. 6

Effects of AgNP on an RBL-2H3 cell. Changes in (a) cell volume and (b) cell surface area after the addition of AgNP. Cross sectional line profiles (c) across the RBL-2H3 cell and (d) phase images at several different time points. The total magnification of our QPM setup was $50\times$ for these images. AgNP was injected into the cell culture chamber at $t=7\min$ . The color bars indicate the phase in radians. (Color online only.)

Moreover, both the lateral width and the height of the cell were increased as well [Fig. 6]. It clearly shows that the cell shape changed after the addition of AgNP in the time sequence of phase images shown in Fig. 6. Results shown in Fig. 6 show that volume change induced by AgNP can be clearly observed in the RBL-2H3 cell, and we need to verify that this volume change is really associated with degranulation in an RBL-2H3 cell with the measurements of ${\left[{\mathrm{Ca}}^{2+}\right]}_{\mathrm{i}}$ and histamine.

3.3.

Volume, Surface Area, and Cross Sectional Profile Changes of a HeLa Cell by AgNP

To confirm that the AgNP-induced volume change in an RBL-2H3 cell is linked with degranulation in the RBL-2H3 cell, we did the same volume measurement experiment for a HeLa cell as a control experiment. Since a HeLa cell is not a mast cell, we expect that there would be no apparent volume change in a HeLa cell when it is stimulated by AgNP. We have measured changes in the volume of a single HeLa cell with QPM before and after AgNP was injected into the cell culture chamber containing many HeLa cells. Figure 7 shows the results. After AgNP was added to the cell culture chamber, the volume of the representative HeLa cell fluctuated slightly around its average value, and no significant changes were detected. The inset in the upper part of Fig. 7 shows typical thickness fluctuations in measured data by QPM on an area without a cell. After the addition of AgNP, the standard deviation of our thickness data became as large as $2.3\mathrm{nm}$ , while it was only about $0.3\mathrm{nm}$ without AgNP. We have verified that these fluctuations are due to scattering associated intensity noise in laser light induced by AgNP. As displayed in Fig. 7, no significant changes were observed in the surface area after the addition of AgNP. Cross sectional line profiles across the white dashed line shown in the top left phase image of Fig. 7 have been shown in Fig. 7. Results shown in Fig. 7 show that there is no noticeable volume change in a HeLa cell after it is exposed to AgNP.

Fig. 7

Effect of AgNP on a HeLa cell. Changes in (a) cell volume and (b) cell surface area after the addition of AgNP at $t=7\min$ . (c) Cross sectional line profiles across the white dashed line shown in the top left image of (d) at several different time points. (d) Phase images corresponding to the time points of (c). The inset in (a) shows the height fluctuation in the area without a cell. Just after the addition of AgNP, the standard deviation of height measured by QPM was increased from $0.3\text{to}2.3\mathrm{nm}$ . The total magnification of our QPM was $50\times$ for these images. The color bars indicate the phase in radians. (Color online only.)

3.4.

Verification of ${\left[{\mathrm{Ca}}^{2+}\right]}_{\mathrm{i}}$ and Extracellular Histamine Increase Induced by AgNP

To verify that the volume change of an RBL-2H3 cell by AgNP is due to the degranulation process of the cell, we have measured ${\left[{\mathrm{Ca}}^{2+}\right]}_{\mathrm{i}}$ fluorescent images using fluorescent microscopy before and after the addition of AgNP to RBL-2H3 cells. Transmembrane fluxes of calcium are known to mediate the degranulation process in mast cells.¹ We have incubated RBL-2H3 cells with a Fluo-3/AM probe for ${\left[{\mathrm{Ca}}^{2+}\right]}_{\mathrm{i}}$ fluorescent imaging. AgNP was added at $t=0\mathrm{s}$ . A series of fluorescent images was taken after the addition and are shown in Fig. 8 . We can clearly see changes in calcium concentration by the variation of fluorescence intensity. The average intracellular calcium fluorescent intensity is calculated for each image and is plotted in Fig. 8 as a function of time. It was normalized by the initial average fluorescence intensity before the addition of AgNP. It shows that the normalized total fluorescence intensity increased quickly just after the addition of AgNP, and later it decreased slowly. The bright green curve is the averaged intracellular calcium fluorescence intensity of the RBL-2H3 cell without any treatment as a function of time. The experimental conditions were the same except for the addition of AgNP. Its intensity decreases quickly as well within a few tens of seconds. This curve works as control data representing the amount of photobleaching in our fluorescence imaging experiment. This slow decrease of the curve, which is analogous to the control data, is mostly because of photobleaching. Furthermore, the calcium in the external solution was depleted with EDTA, and the volume of RBL-2H3 cells treated with A23187 or AgNP was measured. Figures 9 and 9 show the RBL-2H3 cell volume change without the mediator of calcium. No change in the cell volume is observed in the RBL-2H3 cell treated with either A23187 or AgNP.

Fig. 8

(a) Fluorescent images of intracellular calcium in the AgNP-treated RBL-2H3 cell. (b) Normalized average intensity of ${\left[{\mathrm{Ca}}^{2+}\right]}_{\mathrm{i}}$ in the RBL-2H3 cell after AgNP treatment. (Color online only.)

Fig. 9

Effect of external calcium depletion on the cell volume. RBL-2H3 cells were sealed in the chamber filled with DPBS plus $5\mathrm{mM}$ of EDTA and treated with (a) $5\mu \mathrm{g}∕\mathrm{ml}$ of A23187 or (b) $50\mu \mathrm{g}∕\mathrm{ml}$ of AgNP. The cell volume was measured with QPM and normalized with the average volume of a blank period. Small arrows at $t=7\min$ indicate the time points when A23187 or AgNP were added.

We have also verified the degranulation in a RBL-2H3 cell due to the addition of AgNP by measuring the histamine level in cell culture wells. Three RBL-2H3 cell sample groups were prepared: one without any treatment, one with AgNP, and one with A23187. The relative histamine release (percent) for each sample group was measured by a microplate spectrofluorometer. Relative fluorescence intensity from the histamine-OPA complex can verify whether there exist degranulation processes in each cell sample well. Figure 10 shows the time course of histamine release. At each time point after $5\min$ (5, 10, and $20\min$ ) in both the A23187 and AgNP group, histamine release increased significantly compared with that of control $({}^{**}p<0.01)$ . These results show that both A23187 and AgNP activate degranulation in RBL-2H3 cells.

Fig. 10

Time course of histamine release induced by A23187 and AgNP. Measured relative histamine concentrations in three cell sample groups: Hank’s (cells without any treatment), AgNP (cells with AgNP), and A23187 (cells with A23187). At each time point after $5\min$ (5, 10, and $20\min$ ) in both the A23187 and AgNP group, histamine release increased significantly compared with that of the control $({}^{**}p<0.01)$ .