Biomed Microdevices (2007) 9:335–343 DOI 10.1007/s10544-006-9038-y

Electrokinetic measurements of dielectric properties of membrane for apoptotic HL-60 cells on chip-based device Chengjun Huang · Ailiang Chen · Lei Wang · Min Guo · Jun Yu

Published online: 30 December 2006 C Springer Science + Business Media, LLC 2007


Abstract The specific membrane capacitance and conductance of mammalian cells reflect the surface morphological complexities and barrier functions of cell membrane, respectively, and could potentially respond to cell physiological and pathological changes in a measurable manner. In this study, an electrokinetic system was developed by using negative dielectrophoretic force (nDEP force) assisted positioning and electroroation (ROT) measurement. Numerical simulations regarding the geometric model of the electrode were performed primarily for the electric field analysis. The dielectric responses of membrane for apoptotic HL-60 cells induced by bufalin were detected. The membrane capacitance of the cells was found to fall from an initial value of 15.6 ± 0.9 mF/cm2 to 6.4 ± 0.6 mF/cm2 after a 48 h treatment with 10 nM bufalin. However, the membrane conductance remained almost constant at (2.25 ± 1.1) × 103 S/m2 during the first 12 h of bufalin treatment and then increased C. Huang · J. Yu (
) Department of Electronic Science and Technology, Huazhong University of Science and Technology, Wuhan 430074, China e-mail: [email protected], [email protected] A. Chen · L. Wang Department of Biological Sciences and Biotechnology, Tsinghua University, Beijing 100084, China A. Chen · L. Wang Medical Systems Biology Research Center, Tsinghua University School of Medicine, Beijing 100084, China C. Huang · A. Chen · L. Wang · M. Guo · J. Yu National Engineering Research Center for Beijing Biochip Technology, 18 Life Science Parkway, Changping District, Beijing 102206, China

distinctly to (4.2 ± 1.3) × 103 S/m2 thereafter. Scan electron microscopy (SEM) studies of the cells revealed a decreased complexity in cell membrane morphology following bufalin treatments, suggesting that the observed changes in the membrane capacitance was dominated by the alterations of cell surface structures. The results demonstrate that the ROT technique gives a quantitative analysis of the toxic damage by chemicals to cells and can be exploited in the testing and development of new pharmaceuticals and active cell agents. Keywords Cell positioning . Dielectrophoretic force . Electrorotation (ROT) . Apoptosis . Membrane capacitance . Membrane conductance . Bufalin

1 Introduction In recent years, apoptosis has attracted increasing attention, and its study has led to the development of new strategies for cancer chemotherapy (Dive and Hickman, 1991). One area of intense interest is early detection of the apoptotic cells. There are several techniques to detect and measure cell apoptosis, including flow cytometry, propidium iodide (PI) staining, annexin V assay, DNA fragmentation ladder and microscopic analysis, etc. Each of these methods detects different morphological or biochemical features. However, these methods are usually invasive and time-consuming (O’Brien et al., 1997; Alcouffe et al., 1999; Lund et al., 2001; Micoud et al., 2001). There is considerable interest among basic and clinical researchers in developing novel drugs with activity against leukemia. The vast history of experience of traditional Chinese medicine (TCM) may facilitate the identification of novel anti-leukemic compounds. Bufalin is one of the maSpringer

336

jor active components prepared from toad venom extracts of Chan Su, a widely used TCM. Bufalin has been reported to selectively inhibit the growth of various lines of human cancer cells (Akiyama et al., 1999; Kawazoe et al., 1999; Chen et al., 2001; Yeh et al., 2003), especially of leukemic cells (Chen et al., 2001; Xu et al., 2001; Shaulian and Karin, 2002; Addya et al., 2004). It can induce leukemic cells such as HL-60, U937, ML1 and THP-1 into apoptosis (Zhang et al., 1992; Jing et al., 1994; Masuda et al., 1995; Watabe et al., 1996). In recent years, electrorotation (ROT) of particles caused by the interaction between rotating non-uniform alternating current (AC) electric field and field-induced polarization of particles has been extensively studied for microscale preparative uses (Arnold et al., 1988). It has widely potential applications for non-invasive manipulation and characterization of individual biological cells, and provides an opportunity for dielectric characterization of different cell types or cells in different physiological states within a cell mixture. The study of cellular ROT responses as a function of frequency of the applied electric field allows cell membrane capacitance and conductance to be deduced at single cell level. It has been shown previously that these membrane dielectric properties are highly characteristic of cellular physiological activities, and alterations in physiological activities and that the induction of pathologic states rapidly alters the dielectric properties of the cell (Arnold et al., 1988; Hu et al., 1990; Holzel and Lamprecht, 1992; Huang et al., 1992; Becker et al., 1995; Gimsa et al., 1996; Huang et al., 1996; Sukhorukov and Zimmermann, 1996). Some of the different cell types were investigated by this method, including rabbit oocytes (Arnold et al., 1988), yeast (Holzel and Lamprecht, 1992; Huang et al., 1992), murine and human lymphocytes (Hu et al., 1990; Yang et al., 1999), human erythrocytes (Gimsa et al., 1996; Sukhorukov and Zimmermann, 1996), and a number of cancer cell lines (Becker et al., 1995; Huang et al., 1996). However, the previous ROT devices were usually fabricated by gluing an O-ring over a glass/silicon substrate that supported two pairs of micro-electrodes (Arnold et al., 1988; Hu et al., 1990; Holzel and Lamprecht, 1992; Huang et al., 1992; Becker et al., 1995; Gimsa et al., 1996; Huang et al., 1996; Sukhorukov and Zimmermann, 1996). Due to only a single working unit, the throughput of the devices was very low and the experiments usually cost a lot of time and labor; meanwhile, a frequent practical problem with the previous ROT measurements was that the particle was seldom positioned in the exact center of the inter-electrode space. Also, groups of particles could not all be in the exact center of the space, so many potential spectra would be lost, as the uniformity of the field could not be guaranteed over the whole region of the electrode chamber (Dalton et al., 2001).

Springer

Biomed Microdevices (2007) 9:335–343

In an attempt to overcome these problems, we constructed a high throughput ROT chip-based device with a 12 × 2 array of working electrodes which could rapidly detect the dielectric responses in cells, and we also developed a more efficient method of negative dielectrophoretic force (nDEP force) assisted positioning cells before ROT measurements, which could avoid the translational motion of cells causing by the non-uniformity of electric field in the previous ROT measurements. By using the device, we examined the changes of dielectric properties in cell membrane accompanying apoptosis of HL-60 promyelocytic leukemic cells induced by bufalin, which might help to quantitatively analyze toxic chemical effect during identification of novel anti-leukemic compounds from TCM. 2 Principle and design 2.1 Dielectrophoresis When a particle is exposed to a spatially non-uniform electric field, a DEP force will be exerted on the particle, due to the interaction between the electric field and the field-induced dipole. The DEP force, FDEP , acting on a spherical particle is given by Wang et al. (1997): 2 FDEP = 2π εmr 3 Re[ f CM (ω)]∇ E rms

(1)

where r represents the radius of the particle, ω is the angular frequency of the exterior electric field. Re means real part; fCM (ω) represents the dielectric polarization factor (Clausius-Mossotti factor) of the particle, defined as f CM = (εp∗ − εm∗ )/(εp∗ + 2εm∗ ), and εp∗ and εm∗ are the complex permittivities of the particle and the suspending medium, respectively, defined as ε∗ = ε − jσ/ω with ε the permittivity, σ the conductivity and j = (−1)1/2 . The Clausius-Mossotti factor reflects the polarization ability of the particle to the suspending medium. If Re[fCM (ω)] < 0, an nDEP force will occur, which push the particle to the region where the field is 2 reflects the magnitude minimal. The DEP force factor ∇ E rms of the nDEP force acted on the particle. 2.2 Electrorotation Consider a particle suspended in the center of a quadrature electrode system with each electrode energized by a sine voltage 90◦ out of phase with its adjacent electrodes. In this situation, the resultant electric field will rotate at an angular velocity equal to the angular frequency, ω of the applied sine voltages. The particle in the field will experience a rotational torque,(ω). The direction and the magnitude of the rotational torque can be expressed as (Wang et al., 1997; Dalton et al., 2001):

Biomed Microdevices (2007) 9:335–343

337

Fig. 1 The ROT chip and the PMMA cover. The device was constructed by cover a PMMA cover piece with pre-fabricated microstructures against a glass substrate whose upper surface was plated with a 12 × 2 array of circular-micro-electrodes with 200 µm inner radius. The electrode array was comprised of plain parallel elements connected alternately to bus lines on either side of a printed circuit board (PCB) substrate 2 (ω) = −4πr 3 εm Im [ f CM (ω)] E rms

(2)

where Im[fCM (ω)] means the imaginary part of the Clausius– Mossotti factor. In a steady state, the rotational torque is balanced by the oppositely directed torque arising from the frictional force. Then, the resultantly rotational velocity of the particle is given by Wang et al. (1997) and Dalton et al. 2001): R(ω) = −εm E 2 Im [ f CM (ω)] /2η

(3)

where η is the medium dynamic viscosity. When the particle and the medium keep constant, R(ω) is a function of the field frequency, which called “the ROT spectra of cells”. From Eq. (1) and (2), it can be seen that the cellular electrokinetic responses importantly depend on the Clausius-Mossotti factor, Re[fCM (ω)] or Im[fCM (ω)]. 2.3 Positioning electric field and Ratating electric field Cellular electrokinetic responses induced by a rotating AC field usually included translational motion (induced by DEP force) and rotation (induced by rotational torque), depending on the strength and spatial variation of the electric field (Wang et al., 1997; Dalton et al., 2001). In order to avoid the effect of translational motion of cells caused by DEP force, we designed and developed a circular quadrature electrode system with 200-µm inner radius, as shown in Fig. 1. The variation of the non-uniform electric field and the DEP force 2 , generated by this electrode system were simfactor, ∇ E rms ulated and the results were showed in Fig. 2. When four sine

Fig. 2 The spatial variation of ∇ Er2ms When four sine signals (Vpp = 8 V, f = 20 kHz) with opposite phase were applied on the adjacent electrodes, the spatial variation of the DEP force factor was calculated in the x–y plan at z = 0 µm. The values of ∇ Er2ms decreased radially to zero at the center of the inter-electrode space. These meant that the nDEP force would push cells towards to the central region until cells were captured to the center, thus forming an nDEP trap

voltage signals with a special frequency were applied to the adjacent electrodes with phase shifted 180◦ , an nDEP force would be generated in the inter-electrode space. The DEP 2 decreased radially to zero at the center, force factor ∇ E rms as shown in Fig. 2. These meant that the nDEP force would push cells towards to the central region until the cells were captured to the center, thus forming an nDEP trap. Then, if the electric field was switched to “ROT mode” by energizing the four electrodes with sine voltages in phase quadrature, the resultant rotating electric field was shown in Fig. 3. It could be seen from Fig. 3(b) that the nearer to the center of the electrodes, the smaller of the variation of the electric field, and the field vectors in the central region within an 80 µm-radius circle almost did not change their magnitude, as shown by the three curves near x axis in Fig. 3(b). Thus, the nDEP force acted on cells which caused by the non-uniform electric field could be ignored in this region, as shown in Fig. 3(a), and the cells would be well positioned in this central region, without translational motion during ROT measurements. This indicated that the shape of our electrodes could provide a roughly uniform electric field in the central region and reduce the unwanted DEP force that cells might experience. 3 Materials and methods 3.1 Cell culture and chemicals Human promyelocytic leukemic cell line HL-60 was obtained from ATCC (Rockville, MD) and was grown in suspension culture in RPMI 1640 supplemented with 10% Springer

338

Biomed Microdevices (2007) 9:335–343

tration of 10 nM and then incubating and sampling for 6, 12, 24, and 48 h. All the treated cultures were maintained alongside control samples in an incubator and aliquots of each were withdrawn as required for measurements. 3.2 Cell positioning and ROT measurement

Fig. 3 (a) The distribution of the electric field vectors within a circularmicro-electrode. (b) The electric field strength at different places along the path as shown by the arrow in (a). The different colors of the curves stand for the different distances between the detected positions and the center, with ( ) 20 µm, ( ) 50 µm,( ) 80 µm,( ) 100 µm,( ) 120 µm and ( ) 150 µm. The x axis is the electric field strength (kV/m) and the y axis is the time of a period ( × 10−5 s). The vectors in the central region within an 80-µm-radius circle almost did not change their magnitude and provide a roughly uniform electric field in the central region as shown in (a)

FCS (HyClone, Logan, UT), penicillin and streptomycin (100 U/mL and 100 µg/mL, respectively), and L-glutamate (2 mM/L final concentration) at 37◦ C in a humidified atmosphere of 5% CO2 , 95% air. The cells were maintained in logarithmic growth by splitting cells the day prior to induction in 10% FBS supplemented with the described RPMI medium at a density of 5 × 105 cells/mL in culture dishes (BD Biosciences, Bedford, MA). Cell viability was above 98% as judged by trypan blue dye exclusion. Bufalin was donated by Dr. De’an Guo in Beijing University and was dissolved in dimethyl sulfoxide (DMSO) at 100 µM concentration, and stored at − 80◦ C. Then, 10 µM stock solution of bufalin in DMSO was prepared in phosphate-buffered saline (PBS). To induce apoptosis, a time-course experiment was performed by adding 10 µM stock solution to the cell culture solution to a final concenSpringer

The biochip-based device was constructed by covering a ploy (methyl methacrylate) (PMMA) cover piece with premolded microstructures against a glass substrate whose upper surface was plated with a 12 × 2 array of circular-microelectrodes fabricated using microelectromechanical system (MEMS) technology, as shown in Fig. 1. The electrode array was comprised of plain parallel elements connected alternately to bus lines on either side of a printed circuit board (PCB) substrate. The cell samples were washed two times with isotonic (280 mOs/kg) buffer composed of 8.5% (w/v) sucrose + 0.3% (w/v) dextrose solution, which had a conductivity of 355 µS/cm and a relative permittivity of 78, respectively, and re-suspended in the buffer to yield a cell suspension containing approximately 104 cell/mL for measurement. For each experiment, 50 µL of this suspension was pipetted into the concave edge of the PMMA cover and was absorbed to the effective area by capillary action, forming a 50-µm-thick liquid film. An nDEP force was used to position the cells towards the central region of the electrodes by applying four sine voltages (Vpp = 8 V, f = 20 kHz) with opposite phases (DEP Power B, Capitalbio, Beijing, China) on the adjacent electrodes, as we have described previously (Yu et al., 2004). Only those cells lying within 80-µm-radius at the central region of the electrodes and differing from each other at least three cell diameters were measured to ensure that the translational motion and the interaction of the cells could be neglected. Under these conditions, the cells did not experience significant changes in their local electric field during the course of the ROT determinations. To make the ROT measurement, the electrodes were then switched by applying four sine voltages (Vpp = 8 V) in phase quadrature with a frequency range of 100 Hz–25 MHz on the adjacent electrodes. Typically four points per decade were recorded after the cells were well positioned and a complete spectrum was obtained in 20 min. Cells motion was visualized using phase contrast microscopy (Leica DMRE, Leica Microsystems, Germany) with a color CCD (Panasonic WVGP410, Japan) and recorded to video (Panasonic NV-HD 500 video recorder, Japan) for later analysis. Experimental data points were taken at frequencies chosen in random order. These procedures ensured that the measurements were taken on cells in the most homogeneous region of the applied field so that systematic errors caused by time-dependent changes in cells could be avoided. The averaged results were obtained for at least five cells at each point. The procedure of the cell

Biomed Microdevices (2007) 9:335–343

339

positioning and the ROT measurement could be continuously and rapidly performed in different working unit of a 12 × 2 array. The diameter of each cell calibrated against a micrometer was determined from its image on the TV monitor and timing the rotation speed by a stopwatch. The temperature of the measurement chamber was maintained at 21 ± 0.1◦ C.

last, the cells were examined after coating with gold in a scanning electron microscope (Fei Quanta 200, Philips Ltd., Netherlands) at a magnification of 8000 at 10 KV.

3.3 Scanning electron microscopy

4.1 Positioning efficiency

Cells from control and treated cultures were washed in serum-free RPMI and re-suspended in 8.75% (w/v) sucrose solution (280 mOs/kg) for 15 min. Then, cells were fixed in a 3% solution of glutaraldehyde in PBS for 1 h, rinsed in PBS and exposed to 1% osmium tetroxide for a further hour. The cells were dehydrated by sequential immersion in 30%, 50%, 70%, 90% and 2 × 100% ethanol, followed by a 50/50 ethanol/acetone solution and then 100% acetone. At

Based on the simulation results above, the nDEP assisted positioning and the ROT measurements of HL-60 cells were performed. The experimental result well confirmed the simulation prediction by positioning several cells in the central region in 5 s, as shown in Fig. 4. The electric field was applied at a trial frequency (typical 20 kHz) and the induced nDEP force pushed the cells towards the central region of the circular electrodes. The farther cells lay from the electrode edges, the slower their motion became. Once the cells fell down into the trapping area bounded by a circle with an 80 µm-radius, as shown in Fig. 4(b), they would be trapped quickly and accurately in the medium to form a predetermined pattern in the chamber. Even if the signal frequencies were shifted during the ROT measurements (about 20 min), the cells were still well positioned in the center of the electrodes, without obviously translational motion.

4 Results

4.2 ROT spectra and data analysis Typical ROT spectra were shown in Fig. 5 for the untreated HL-60 cells and the HL-60 cells treated by bufalin. Spectra for the control samples without bufalin treatments incubated for 0, 6, 12, 24 and 48 h were found to be almost identical, indicating a reasonably high degree of homogeneity within

Fig. 4 Positioning the HL-60 cells by the nDEP force. Initially, the cells were distributed in the chamber randomly (Fig. 4(a)). When the signals were applied, the cells were well defined in the central region bounded by a circle with an 80 µm radius after several seconds (Fig. 4(b))

Fig. 5 Typical ROT spectra of the HL-60 cells on time-dependency. The cells were treated for 0 h ( ), 6 h ( ), 12 h ( ), 24 h ( ) and 48 h ( ) with 10 nM bufalin. The cells were suspended in an isotonic sucrose/dextrose medium with a conductivity of 355 µS/cm (dot line for experimental data, and real line for theoretical data)

Springer

Figure 7 shows the SEM photographs of the untreated HL-60 cells and the cells following 10 nM bufalin treatments at 6, 12, 24 and 48 h as apoptosis progression, the untreated cells appeared to be homogeneous in both size ( ∼6 µm in diameter) and surface morphology. The surfaces of the untreated Table 1 The membrane capacitance (Cmem ) and the membrane conductance (Gmem ) of the untreated HL-60 cells and the apoptotic HL-60 cells induced by bufalin

Springer

6

2

17 15

5

13

4

11

3

9 2

7 5

0

6 12 24 Treatment Time (h)

Membrane Capacitance

50

1

Membrane Conductance

Fig. 6 The changes of dielectric properties in cell membrane of the apoptotic HL-60 cells

cells were covered with abundant microvilli (Fig. 7(a)). After 6 h treatment of bufalin, the number of microvilli was already decreased (Fig. 7(b)), and after 48 h, most of the cell surfaces were predominantly smooth with few microvilli (Fig. 7(e)). Furthermore, cytotoxic effects were detectable after the exposure of cells to bufalin. After the exposure time >12 h, cell shrinkage and blebs formation were observed. This phenomenon was progressively accompanied by an increase in the number of blebs (Fig. 7(c), (d), (e)). It was associated with the increases in smooth areas of the cell surface and loss of microvilli, suggesting that microvilli and other membrane features were lost to the blebs. The loss of cell surface area would accompany the decreasing morphological complexity. This observation was consistent with the observed decrease in membrane capacitance during the progression of bufalin induced apoptosis of HL-60 cells, as shown in Figs. 6 and 7.

5 Discussion

4.3 SEM

0 h (control) 6h 12 h 24 h 48 h

2

the untreated cell population. We have previously described the general features of the ROT spectra of biological cells (Wu et al., 2005). As shown in Fig. 5, the cell rotations occurred in the opposite direction to the applied rotating field (anti-field rotation) below 3 MHz, and in the same direction to the field (co-field rotation) above that frequency at a suspension conductivity of 355 µS/cm. The untreated and treated cells both exhibited two relaxation peaks with the anti-field rotation peak and co-field rotation peak. The shapes of spectral curves of HL-60 cells were similar to some other types of mammalian cells observed previously (Wang et al., 1994; Sukhorukov and Zimmermann, 1996; Kurschner et al., 1998; Yang et al., 1999a). However, there were distinct differences within the different spectra obtained from the untreated and the apoptotic HL-60 cells. The anti-field rotation peaks of the ROT spectra decreased gradually as the time of bufalin-treatment increased, and the peak frequencies of the ROT spectra shifted progressively towards lower frequencies following the increased time of the celltreatment with bufalin. For example, the anti-field rotation peak for the untreated cells reached 0.12/s · V2 , which occurred at ∼150 kHz. However, it fell to 0.052/s · V2 with the peak frequency ∼70 kHz by a 48 h treatment with 10 nM bufalin. The characteristic shape of the ROT for mammalian cells could be well described by a single shell dielectric model (Arnold and Zimmermann, 1988; Huang et al., 1992; Wang et al., 1994). The corresponding dielectric parameters derived from the best fit curves to the ROT spectra according to the model were summarized in Table 1 for the apoptotic HL-60 cells induced over time by bufalin. Figure 6 shows the alterations in the membrane capacitance (Cmem ) and the membrane conductance (Gmem ) over the time course of inducing cell apoptosis.

Gmem (S/m )

Biomed Microdevices (2007) 9:335–343

Cmem (mF/m )

340

Cmem (mF/m2 )

Gmem (S/m2 )

15.6 ± 0.9 12.8 ± 0.8 10.4 ± 0.7 7.5 ± 0.8 6.4 ± 0.6

(2.25 (2.25 (2.67 (3.50 (4.20

± ± ± ± ±

1.1) 1.0) 1.1) 1.2) 1.3)

× × × × ×

103 103 103 103 103

Our study combined the nDEP force assisted positioning and the ROT techniques together, and obtained more accurate ROT spectra, due to overcoming the effect of translational motions of the cells caused by non-uniform electric field. The ROT measurements have revealed that significant differences exist in the dielectric properties of the cell membrane in the different physiological states. These findings demonstrate that cell dielectric characteristics are closely associated with biologic function and collectively constitute a physical component of cell phenotype. This dielectric phenotype provides a new additional criterion for the classification of cells. The experimental results revealed that the membrane capacitance of the untreated HL-60 cells was approximately 15.6 ± 0.9 mF/m2 and which reduced to 6.4 ± 0.6 mF/m2 after a 48 h treatment with 10 nM bufalin; meanwhile, the membrane conductance increased from an initial value of

Biomed Microdevices (2007) 9:335–343

Fig. 7 The SEM photographs of HL-60 cells at a magnification of × 8000 after treatments for 0 h (a), 6 h (b), 12 h (c), 24 h (d) and 48 h (e) with 10 nM bufalin. Following the bufalin treatments, the

341

cells revealed progressive changes in the cell membrane morphology as apoptosis developed

Springer

342

(2.25 ± 1.1) × 103 S/m2 to (4.2 ± 1.3) × 103 S/m2 . These results were comparable with earlier findings which determined the dielectric properties of HL-60 cells by DEP crossover frequency method (Ratanachoo et al., 2002; Wang et al., 2002). It indicated that the ROT method used by us was valid for the measurement of the dielectric responses of cell membrane. Figure 6 shows the membrane responses in Cmen and Gmen resulted from bufalin treatments in a timedependent manner. The changes in membrane capacitance were up to about 60% between the initial value and the value after a 48 h treatment of bufalin. It also could be seen from Table 1 and Fig. 6 that the membrane conductance increased in a time-dependent manner, but these changes occurred later than the observed changes in membrane capacitance. During the first 12 h of treatment, the membrane conductance was (2.67 ± 1.1) × 103 S/m2 and was little altered compared with the untreated cells ((2.25 ± 1.1) × 103 S/m2 ), but after a 48 h treatment, the membrane conductance reached (4.2 ± 1.3) × 103 , which was almost twice of that of the untreated cells. Moreover, the membrane capacitance value of 15.6 ± 0.9 mF/m2 for the untreated HL-60 cells also fell within the range of values determined by the ROT method for T-cells (10.5 ± 3.1 mF/m2 ), and B-cells (12.6 ± 3.5 mF/m2 ) (Yang et al., 1999). However, the standard deviations obtained in our experiments (0.9 mF/m2 ) were less than those (3.1–3.5 mF/m2 ) detected by the previous ROT method. This may partially due to the fact that cell positioning procedure used in our experiments decreased the effect of non-uniformity of the electric field and decreased the systematic errors caused by cell location. The SEM results confirmed that a reduction in the density of complex surface morphology occurred during the apoptosis progression of HL-60 cells and that the time course of these changes mirrored that of the detectable dielectric changes by the ROT method. The magnitude of the membrane capacitance (defined as the capacitance per unit area of the cytoplasmic membrane (Zimmermann and Neil, 1996)) can be taken as a measure of the total surface area of the cell and the extent to which this is enhanced by membrane features, such as microvilli, folds and blebs. During apoptosis progressing of HL-60 cells, there is a distinct loss of the surface features, which decrease the net surface area of the cell membrane, as shown in Fig. 7. We believe that over the time course of apoptosis these changes in membrane morphology mirror the measured changes in membrane capacitance determined by the ROT method; meanwhile, the continuous shrinking of cells can also cause a decreased membrane capacitance if the reduced surface area and the possible delivery of membrane-bound water are taken into account. In addition, the alterations in membrane-molecule polarization may also be one of the reasons of the decreased membrane capacitance. The observed decrease of membrane capacitance is

Springer

Biomed Microdevices (2007) 9:335–343

probably a complex process that originates from more than one event (Dopp et al., 2000). On the other hand, due to the near-insulating nature of the membrane lipid bilayer, the membrane conductance mainly reflects the net transport of ionic species across the plasma membrane through the charge carrier transportation system (pores, ion carriers, channels and pumps) under the influence of the applied electric field (Wang et al., 2002). In our experiments, the alterations in membrane conductance occurred later than the changes observed in membrane capacitance (see Table 1 and Fig. 6). It is well known that cells undergoing apoptosis can maintain the barrier function of their plasma membranes for several hours after initiating nuclear damage (Martin et al., 1995; Darzynkiewicz et al., 1997). The increasing membrane conductance might result from injury to genome and gene transcription system and the membrane barrier function may change because of looser packing of phospholipids or increased field-dependent ion migration through ion channels during apoptosis. Thus, the membrane conductance can remain almost constant for about 12 h when cells undergo apoptosis and then increase independently until the membrane barrier function breaks down completely. Previous studies have shown that the alterations of membrane morphology and membrane barrier function of cells were important events in the apoptotic process (Savill et al., 1993; Wyllie 1993; Woodle and Kulkami, 1998) and appeared to dominate the differences in the dielectric properties (Huang et al., 1996; Dopp et al., 2000; Wang et al., 2002; Ratanachoo et al., 2002). However, to date, the mechanism to clearly explain this phenomenon has not been identified. For example, the morphological changes alone can account for the dielectric alterations demonstrated here, with a possible minor contribution from membrane compositional modifications. Also, at the plasma membrane level, the alterations in ion transportation, are very important to the subsequent events in the apoptotic sequence (Lang et al., 1998; Wang et al., 2002). It would be of interest and crucial to correlate such specific processes with the dielectric changes of the cell membrane we have observed by the ROT measurement here. And our findings suggest that further investigation is needed of the plasma membrane as a possible site for early and obligatory participation in apoptosis.

6 Conclusions A high throughput electrokinetic experimental system has been developed that allows a rapid and high throughput detection of the dielectric properties of the apoptotic HL-60 cells induced by bufalin. The method of ROT measurement after positioning the cells at the central region of the interelectrode space, made it possible to acquire the ROT spectra

Biomed Microdevices (2007) 9:335–343

of the cells at low frequency. The effect of the various time of bufalin to induce HL-60 cells apoptosis have been investigated using this apparatus. Our results showed that the dielectric properties of the cell membrane clearly reflect the biophysical alterations to the cell membrane that appeared while undergoing apoptosis induced by bufalin. The results indicated that the ROT method might be useful for the quantitative analysis of toxic chemicals effects in terms of the kinetics of the development of cell damage. The work also suggested further potential biotechnological and biomedical application, as for example in the monitoring of cell viability during exposure to pharmacological agents, toxic chemicals and drug sensitivity testing during drug discovery from TCM. Acknowledgments We thank the financial support from Chinese National 863 plan (Project code: 2002AA2Z2011), National Engineering Research Center for Beijing Biochip Technology (Beijing, China) and Capitalbio Corp. (Beijing, China).

References S. Addya, M.A. Keller, K. Delgrosso, C.M. Ponte, R. Vadigepalli, G.E. Gonye, and S. Surrey, Physiological Genomics 19, 117–130 (2004). M. Akiyama, M. Ogura, M. Iwai, M. Iijima, S. Numazawa, and T. Yoshida, Human Cell 12, 205–209 (1999). J. Alcouffe, S.C. Bauguil, V. Garcia, R. Salvayre, M. Thomsen, and H. Benoist, Journal of Lipid Research 40, 1200–1210 (1999). W.M. Arnold, U. Zimmermann, W. Heiden, and J. Ahlers, Biochimca et Biophysica Acta 942, 96–106 (1988). W.M. Arnold and U. Zimmermann, Journal of Electrostatics 21, 151–191 (1988). F.F. Becker, X.B. Wang, Y. Huang, R. Pethig, J. Vykoukal, and P.R. Gascoyne, Proceedings of the National Academy of Science of USA 92, 860–864 (1995). X.Y. Chen, W.L. Hu, R.C. Xu, L. Chen, and J. Qian, Chinese Journal of Pharmacology and Toxicology 15, 293–296 (2001). C. Dalton, A.D. Goater, J. Drysdale, and R. Pethig, Colloids and Surfaces A: Physicochemical and Engineering Aspects 195, 263–268 (2001). E. Dopp, L. Jonas, B. Nebe, A. Budde, and E. Knippel, Environmental Health Perspectives 108, 153–158 (2000). Z. Darzynkiewicz, G. Juan, X. Li, W. Gorczyca, T. Murakami, and F. Traganos, Cytometry 27, 1–20 (1997). C. Dive and J.A. Hickman, British Journal of Cancer 64, 192–196 (1991). J. Gimsa, T. Muller, T. Schnelle, and G. Fuhr, Biophysical Journal 71, 495–506 (1996). R. Holzel and I. Lamprecht, Biochimica et Biophysica Acta 1104, 195–200 (1992). X. Hu, W.M. Arnold, and U. Zimmermann, Biochimica et Biophysica Acta 1021, 191–200 (1990).

343 Y. Huang, R. Holzel, R. Pethig, and X.B. Wang, Physics in Medicine and Biology 37, 1499–1517 (1992). Y. Huang, X.B. Wang, F.F. Becker, and P.R. Gascoyne, Biochimica et Biophysica Acta 1282, 76–84 (1996). Y. Jing, H. Ohizumi, N. Kawazoe, S. Hashimoto, Y. Masuda, S. Nakajo, T. Yoshida, Y. Kuroiwa, and K. Nakaya, Japanese Journal of Cancer Research 85, 645–651 (1994). N. Kawazoe, T. Aiuchi, Y. Masuda, S. Nakajo, and K. Nakaya, Journal of Biochemistry 126, 278–286 (1999). M. Kurschner, K. Nielsen, C. Andersen, V.L. Sukhorukov, W.A. Schenk, R. Benz, and U. Zimmermann, Biophysical Journal 74, 3031–3043 (1998). F. Lang, A.L. Wienhues, M. Paulmich, I. Szabo, D. Siemen, and E. Gulbins, Cell Physiological Biochemistry 8, 285– 292 (1998). P.K. Lund, A.B. Westvik, G.B. Joo, R. Ovstebo, K.B. Haug, and P. Kierulf, Journal of Immunological Methods 252, 45–55 (2001). S.J. Martin, C.P. Reutelingsperger, A.J. McGahon, J.A. Rader, R.C. van Schie, D.M. Laface, and D.R. Green, Journal of Experimental Medicine 182, 1545–1556 (1995). Y. Masuda, N. Kawazoe, S. Nakajo, T. Yoshida, Y. Kuroiwa, and K. Nakaya, Leukemia Research 19, 549–556 (1995). F. Micoud, B. Mandrand, and C. Malcus-Vocanson, Cell Proliferation 34, 99–113 (2001). I.E. O’Brien, C.P. Reutelingsperger, and K.M. Holdaway, Cytometry 29, 28–33 (1997). K. Ratanachoo, P.R. Gascoyne, and M. Ruchirawat, Biochimica et Biophysica Acta 1564, 449–458 (2002). J. Savill, V. Fadok, P. Hension, and C. Haslett, Immunology Today 14, 131–136 (1993). E. Shaulian and M. Karin, Nature Cell Biology 4, 131–136 (2002). V.L. Sukhorukov and U. Zimmermann, Journal of Membrane Biology 153, 161–169 (1996). X. Wang, F.F. Becker, and P.R. Gascoyne, Biochimica et Biophysica Acta 1564, 412–420 (2002). X.J. Wang, X.B. Wang, and P.R. Gascoyne, Journal of Electrostatics 1997, 277–295 (1997). X.B. Wang, Y. Huang, P.R. Gascoyne, F.F. Becker, R. Holzel, and R. Pethig, Biochimica et Biophysica Acta 119, 330–344 (1994). M. Watabe, Y. Masuda, S. Nakajo, T. Yoshida, Y. Kuroiwa, and K. Nakaya, Journal of Biological Chemistry 271, 14067–14072 (1996). E.S. Woodle and S. Kulkami, Transplantation 66, 681–691 (1998). Y.F. Wu, C.J. Huang, L. Wang, X.L. Miao, W.L. Xing, and J. Cheng, Colloids and Surfaces A: Physicochemical and Engineering Aspects 262, 57–64 (2005). A.H. Wyllie, British Journal of Cancer 7, 205–208 (1993). R.C. Xu, X.Y. Chen, L. Chen, and J. Qian, Zhongguo Zhong Yao Za Zhi 26, 59–61 (2001). J. Yang, Y. Huang, X. Wang, X.B. Wang, F.F. Becker, and P.R. Gascoyne, Biophysical Journal 76, 3307–3314 (1999). J.Y. Yeh, W.J. Huang, S.F. Kan, and P.S. Wang, Prostate 54, 112–124 (2003). Z. Yu, G.X. Xiang, L.B. Pan, L.H. Huang, Z.Y. Yu, W.L. Xing, and J. Cheng, Biomedical Microdevices 6, 311–324 (2004). L. Zhang, K. Nakaya, T. Yoshida, and Y. Kuroiwa, Cancer Research 52, 4634–4641 (1992). U. Zimmermann and G.A. Neil, Electromanipulation of Cell (CRC Press, Boca Raton, 1996).

Springer