| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Department of Physiology and Pharmacology, Sackler Faculty of Medicine, Tel Aviv University, Tel Aviv, Israel
Correspondence: Address reprint requests to Prof. Rafi Korenstein, Dept. of Physiology and Pharmacology, Sackler School of Medicine, Tel Aviv University, Tel Aviv 69978, Israel. Tel.: 972-3-640-6042; Fax: 972-3-640-8982; E-mail: korens{at}post.tau.ac.il.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
200 mV, has been associated mostly with the phenomenon of electroporation (Neumann and Rosenheck, 1972
; Zimmermann et al., 1974; Kinosita and Tsong, 1977). Its employment has gained wide use in the areas of cell biology and biotechnology, since it supplied an efficient physical tool to permeabilize cells by opening hydrophilic pathways in the cell membrane, through which molecules could diffuse along their electrochemical gradients (for recent reviews, see Teissie, 2002; Weaver, 2003; Gehl, 2003). Exposure of cells to extremely low electric and magnetic fields, which can induce a transmembrane potential difference <1 mV, have been associated with environmental exposures of humans to low-frequency electromagnetic fields on the one hand, and, on the other, with possible application of medical therapeutic devices (for reviews, see Lacy-Hulbert et al., 1998; Adair, 1999; Ahlbom et al., 2001). The underlying mechanisms of these very weak electric fields are still ill understood and highly debated (Adair, 1999; Weaver et al., 1999, 2000; Foster, 2003). The intermediate domain of low electric fields, which may lead to the induction of transmembrane potential alteration in the range of 1–100 mV, has been routinely applied in electrophysiological studies of ion transport through channels. It was realized quite early that physiological electric fields in tissues are at the lower end of this range (Jaffe, 1966; Nuccitelli and Jaffe, 1974; Borgens et al., 1977). These endogenous electric fields were associated with processes of development, regeneration and wound healing (Nuccitelli, 2003). An attempt was made to understand the role of these electric fields under situations imitating physiological and pathological conditions by applying low DC electric fields and examining the phenomenon of electrophoresis of charged membrane components in the plane of the cell membrane. Lateral electrophoretic displacements of charged membrane proteins and lipids resulted in segregation of these components at the cell surface (Poo and Robinson, 1977; Poo et al., 1979; Poo, 1981). Recent studies of the effect of low DC electric fields on altering basic cellular activities such as cell cycle, directional growth, and motility explored the underlying cellular pathways involved in inducing these changes (e.g., Zhao et al., 2002; Song et al., 2002; Wang et al., 2003).
An additional manifestation of the possibility to affect natural cellular functions by low electric fields emerges from our previous observation that exposure of cells to pulsed low electric fields can stimulate endocytic activity (Rosemberg and Korenstein, 1997). We could demonstrate the enhanced uptake of macromolecules (polysaccharides in the range of 1–2000 kD and ß-galactosidase) via stimulating endocytic-like processes by exposing cells to a train of pulsed low electric fields (LEF). Recently, we have extended this study by exploring the basic spatial and time-dependent characteristics of LEF-induced uptake via adsorptive and fluid-phase uptake (Antov et al., 2004).
In this study we explore the steady-state and temporal LEF-induced changes in the adsorption properties of the cell surface and the interrelationship of these changes with the subsequently enhanced uptake process.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
Exposure of cells to pulsed electric fields
Cells in suspension were exposed to a train of low intensity unipolar rectangular voltage pulses by employing an electric pulse generator (Grass S44 Stimulator, Grass-Telefactor, West Warwick, RI). The exposure was performed in a plastic cuvette by placing the cell suspension between two parallel stainless steel electrodes separated by 0.5 cm, yielding a uniform electric field. The electric field parameters were monitored on line by recording voltage and current (using a wide-band current transformer, Rearson Electronics, Palo Alto, CA) on a Tektronix 2430A oscilloscope (Tektronix, Beaverton, OR). Experiments were run at different temperatures in the range of 4–37°C. In routine experiments, 0.5 ml of cell suspension (0.5–10 x 106 cells/ml) in DMEM-H was exposed to a train of electric field pulses of 20 V/cm. The train consisted of unipolar rectangular pulses of 180 µs duration, frequency of 500 Hz, and total exposure time of 1 min. The application of this train of pulses resulted in small polarization of the electrodes and the appearance of a residual low DC component (
2 V) at the highest electric field. The temperature of the solutions during exposure was measured using fiberoptic temperature sensors (FISO Technologies, Quebec, Canada). The readings were recorded every 2.2 s with averaging 1 s/point. The typical transient temperature rise measured at the end of 1 min exposure in DMEM-H medium to LEF of 20 V/cm at 4°C, 24°C, and 37°C was 1.4 ± 0.6°C.
Molecular probes
Dextran conjugated to fluorescein-5-isothiocyanate (dextran-FITC, 2000 kDa, 0.009 mol FITC/mole glucose) was used at a final concentration of 0.2 µM. Bovine serum albumin conjugated to fluorescein-5-isothiocyanate (BSA-FITC) contained 12 mol FITC per mole albumin (66 kDa). Both FITC-conjugated probes were dialyzed before use against the exposure medium. Propidium iodide (PI) was used at the final concentration of 30 µM for the detection of damaged cells. These markers were purchased from Sigma Chemicals (Rehovot, Israel). No change was detected in the fluorescent spectra of the different probes at the specified concentration used, suggesting that no probe aggregation took place under the conditions employed.
Measurement of adsorption and uptake by flow cytometry, fluorimetry, and confocal fluorescence microscopy
Cells (usually 3 x 106 ml–1) were exposed to low electric fields in the presence of a molecular probe. In some cases the probes were added after the exposure of the cells to LEF. After exposure and incubation with the probe, cells were washed twice in DMEM-H medium. Since BSA-FITC was found to be significantly adsorbed to the cell membrane, especially after exposure to LEF, we differentiated between internalized and adsorbed fractions of the probe by subjecting the cells to 0.01% trypsin in phosphate-buffered solution for 5 min at 37°C (Glogauer, 1992). The trypsin-digested BSA-FITC represents the amount of BSA adsorbed to the cell population, whereas the amount of the probe measured per single cell is attributed to the internalized amount of the probe. The efficiency of removal of the adsorbed BSA-FITC probe from the cell surface by trypsin was validated by confocal fluorescence microscopy (LSM 410, Zeiss, Jena, Germany). To visualize the cellular location of the probe the exposed cell suspensions and controls were washed twice in DMEM-H and kept at the chosen temperature, until observation at the appropriate time by confocal microscopy. Computer-generated images of 0.5-µm optical sections were obtained at the approximate geometric center of the cell as determined by repeated optical sectioning.
Flow cytometry analysis was carried out by FACSort (Becton Dickinson, San Jose, CA), employing 488-nm argon laser excitation. The green fluorescence of FITC was measured via 530/30-nm filter whereas the red fluorescence of PI was detected via 585/42-nm filter. To eliminate signals due to cellular fragments, only those events with forward and side light scattering comparable to whole cells were analyzed. The side light scattering was also employed to detect changes in the formation or disappearance of intracellular vesicles. All fluorescence signals were logarithmically displayed. Ten thousand cells were run for each sample and data were collected in the list mode. The analysis of flow cytometry data was performed using WINMDI 2.8 flow cytometry application software. Control samples exhibited a rise in fluorescence when incubated with fluorescent probes at 24°C due to constitutive uptake. Cells exposed to LEF without any probe also revealed an increased fluorescence as compared to untreated control cells due to a rise in autofluorescence after exposure. This background signal was subtracted from the fluorescence of all cells exposed to electric field. Results were expressed in terms of cell quantities at 95% confidence interval or as the geometric mean of cell population. The geometric mean (Gm) was used for log-amplified data as it takes into account the weighting of data distribution. The combined LEF-induced adsorption and uptake of the probe was determined by flow cytometry of cells that were not subjected to the trypsinization step (Fig. 1 J, histogram 3). In contrast, the uptake of the probe was determined by flow cytometry of cells after being subjected to trypsinization (Fig. 1 J, histogram 4). Efficiencies of probe adsorption and uptake were characterized by fold of induction (FI) which was calculated as a ratio of the geometric mean of fluorescence in an exposed sample to that of a control unexposed one.
|
ex = 485/20, em = 530/25). The readings were normalized to the fraction removed from the unexposed samples. Cell viability was assessed by the propidium iodide exclusion test. For flow cytometry and confocal microscopy, PI (30 µM) was added to exposed and control cells 5 min before evaluation.
Binding and uptake studies
The adsorption of albumin to the plasma membrane of COS 5-7 cells was characterized using BSA-FITC as a probe. The cells were washed three times at 4°C with DMEM-H (210 x g; RT6000D, Sorvall) for 5 min and resuspended, usually at 3 x 106/ml, in cold (4°C) DMEM-H medium containing different BSA-FITC concentrations (0.1–15 µM). Control and exposed cells were incubated with BSA-FITC for a total period of 5 min at 4°C. Since similar adsorption of BSA-FITC was obtained for 30 min at 4°C, it may be concluded that we reached a steady state. The cells were exposed for 1 min to LEF (20 V/cm, 180 µs, 500 Hz) at 4°C after 4 min of preincubation with the probe at the same temperature. Then the cells were washed three times in cold DMEM-H medium to remove unbound BSA-FITC. Analysis of adsorbed BSA-FITC was carried out by flow cytometry. At 4°C the BSA-FITC binds to the plasma membrane but is not internalized. Nonspecific binding of BSA-FITC was determined by incubating the cells in a 1000-fold excess of unlabeled albumin, with consequent washing. The analysis of the adsorption isotherms was based on a single binding site scheme: B = (Bmax x C)/(Kd + C), where C is the probe's concentration, B is the probe adsorption (in terms of Gm), Bmax is the maximal probe adsorption at high probe concentrations, and Kd indicates the probe-receptor dissociation constant. Curve fitting (of the above expression as well as of Eadie-Hofstee and Hill plots) was performed by the least squares method with Microcal Origin 6 software (Microcal Software, Northampton, MA) using direct weighting to avoid errors at small concentrations.
Uptake by COS 5-7 cells was initiated by suspending the cells in DMEM-H medium containing different concentrations of BSA-FITC. For each uptake assay the probe was in the cells' suspension for a total of 5 min at 4°C (preincubated for 4 min and exposed for 1 min to LEF at 4°C). Then cells were washed once at 4°C and further incubated in the absence of BSA-FITC for an additional 25 min at 24°C. At the end of the incubation cells were treated with 0.01% trypsin for 5 min at 37°C. The efficiency of the removal of BSA-FITC from the cell surface was validated by microscopic visualization. The extent of nonspecific uptake of BSA-FITC was determined by incubating the LEF-exposed and control cells, in a 1000-fold excess of unlabeled albumin, for 15 min at 4°C.
To study the extent of LEF-induced adsorption and uptake of BSA-FITC as a function of time interval after exposure of cells to LEF, the probe was either present during the electric treatment or added to the cell suspension at various times after termination of exposure from zero to 10 min. To differentiate between LEF-induced adsorption and uptake kinetics experiments were performed at 4°C and 24°C.
To determine the dependence of LEF-induced binding and uptake of BSA-FITC on temperature the cells were incubated during and after exposure at different temperatures in the range of 4–37°C, using the same protocols as described above.
Dependence of uptake on electrical parameters and medium composition
The dependence of LEF-induced uptake of BSA-FITC by COS 5-7 cells on the solution's conductance characteristics was studied by performing uptake experiments in solutions of different compositions. DMEM-H medium was mixed with 0.3 M sucrose supplemented with 25 mM HEPES at different proportions (0, 10, 25, 50, 75, and 100%). Cells (3 x 106/ml) were resuspended in these solutions in the presence of 6.8 µM BSA-FITC and then exposed to LEF for 1 min at 24°C under different conditions:
Adsorption of DNA and DNA-complex to cell membrane after exposure
A 4.7-kb plasmid (GFP S65T, Clontech, Palo Alto, CA) carrying green fluorescent protein was prepared from Escherichia coli using Midiprep DNA purification system (Promega, Madison, WI). The plasmid was stochiometrically stained with the DNA intercalating dye PI. The staining was performed with 7.5 x 10–5 M dye at DNA concentration of 1 µg/µl for 60 min at 4°C. This concentration yields an average base pair/dye ratio of 4–5 (Golzio et al., 2002). Stained DNA was added to the cell suspension as a single component or as a complex with either diethylaminoethyl-dextran (DEAE-dextran, 500 kDa), or polyethylenimine (PEI, 2000 Da). The DEAE-dextran-DNA complex was prepared by incubating the labeled DNA plasmid with 1 mg/ml DEAE-dextran for 30 min at 24°C. DNA-PEI complex was prepared by incubating stained DNA with 6 µM of PEI for 30 min at 4°C.
The study of the LEF-induced adsorption of DNA to the cell membrane of COS 5-7 cells was performed by resuspending the cells in DMEM-H medium at 4°C in the presence of PI-labeled plasmid or plasmid complexes. One-half milliliter of cell suspension containing 106 cells and 10 µg of labeled plasmid or DNA complex with either DEAE or PEI was exposed to LEF of 20 V/cm for 1 min at 4°C. Due to its high quantum yield the PI-labeled plasmid was used for the quantification of DNA adsorption. Quantitative analysis of DNA binding to cell surface was performed by flow cytometry.
Determination of LEF-induced cell-cell aggregation
To study electric field-induced cell-cell aggregation COS 5-7 cells were suspended in DMEM-H medium at five different cell concentrations: 0.3 x 106; 0.75 x 106, 1.5 x 106, 3 x 106, and 6 x 106 cells/ml. They were exposed for 1 min to LEF of 10 V/cm (pulse duration of 180 µs, 500 Hz). After exposure the 0.5 ml of cell suspension was diluted 100-fold in DMEM-H to minimize spontaneous aggregation.
The time dependence of LEF-induced aggregation was estimated by exposing low concentration of 0.5 x 106 cells/ml to electric stimulation of 10 V/cm (180 µs, 500 Hz) for 1 min. After exposure 0.5 ml of cell suspension was transferred to a large volume (50 ml) of DMEM-H. At different times after exposure in the range of 0–2 h, cells in both control and exposed samples were sedimented down (210 x g for 3 min) so that they formed a loose pellet of cells, leading to enhancement of cell-cell aggregation. Cell pellets were gently resuspended by a standardized procedure. Detection of cell aggregation was performed using a Coulter counter (Coulter Multisizer, Coulter Electronics, Miami, FL) employing a 100-µm orifice. For each sample, 50–70 x 103 events were counted and data were collected in a list mode. The extent of LEF-induced cell-cell aggregation was calculated in terms of the percentage of aggregates consisting of two or more cells using the modified procedure of McLean and Hause (1981). First, size distribution parameters were determined for a diluted 0.3 x 106 cells/ml suspension. Then the gating value of single cells' population mean diameter multiplied by 1.2 was chosen for the selection of aggregates.
Statistical analysis
Experiments were conducted at least in triplicate and the mean and standard deviation (SD) were calculated. Significances of differences between groups were assessed by unpaired Student's t-test.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
|
Dependence of LEF-induced adsorption and uptake on BSA-FITC concentration
The dependence of LEF-induced binding on probe concentration was studied by exposing COS 5-7 cells to LEF in the presence of different concentrations of BSA-FITC in the range of 0–15 µM at 4°C. The total binding of the BSA-FITC probe to the cell surface as well as the nonspecific one (after incubation with 1000-fold excess of unlabeled BSA) were determined by flow cytometry (Fig. 3 A). The total binding of BSA-FITC both in LEF-treated and unexposed cells is concentration-dependent. Analyses of the binding data reveal that respective Eadie-Hofstee plots (not shown) for both the constitutive and LEF-induced total binding are curved and may be clearly divided into two linear regions, one with high affinity and low capacity, and another one with low affinity and high capacity. In contrast, the specific binding of BSA-FITC to control and LEF-treated cells is well described as a one-binding site reaction. Taking into account the large differences between high- and low-affinity binding sites, we assume that, in the concentration range studied, the total binding is a sum of specific and nonspecific binding, where the latter linearly increases with concentration. The exact dissociation constants and maximal binding capacities were obtained from nonlinear fitting of the adsorption isotherms (Fig. 3 A) as it gives more precise values compared to linear Eadie-Hofstee or Scatchard plots.
|
5 µM. The fitting to one binding site model gives half-maximum saturation concentration of 
(value ± fitting error) for unexposed control samples, and 
for the exposed ones. The binding capacity of specific sites on the plasma membrane was significantly (6.5-fold) higher in exposed cells where 
a.u. as compared to controls, where 
a.u.. The specific binding of BSA-FITC to control cells lacks cooperativity (Hill plot slope of 1.07 was not significantly different from 1), while the specific binding of the probe to LEF-treated cells exhibits a weak negative cooperativity with a slope of 0.78 for the Hill plot.
The nonspecific binding of BSA-FITC was found to be linearly dependent on BSA-FITC concentration in the range of 0–15 µM both for control and LEF-treated COS 5-7 cells (R2 = 0.78 and 0.98, respectively). The nonspecific binding of BSA-FITC to control cells is very weak as characterized by a slope 
of 0.4 ± 0.09 a.u./µM, while in LEF-treated cells it is 33-fold steeper, possessing 
of 13.2 ± 0.6 a.u./µM.
The dependence of LEF-induced uptake on the probe's concentration was examined simultaneously with the binding studies. This was performed by determining LEF-induced internalization into COS 5-7 of different BSA-FITC concentrations in the range of 0.1–15 µM at 24°C. To differentiate uptake via specific binding sites from the total uptake, the cells were incubated at 4°C in the presence of three different BSA-FITC concentrations of 6.8 µM, 10 µM, and 15 µM, and were exposed to LEF, again at 4°C. An excess of unlabeled BSA was then added to exposed and unexposed cell suspensions, and the temperature was raised to 24°C, resulting in an uptake of BSA-FITC only via nonspecific binding sites. The results were analyzed as performed for the binding experiments. The uptake as a function of BSA concentration, normalized to the maximal total uptake, is shown in Fig. 3 B. The constitutive process taking place in control cells shows a specific uptake with saturation at concentrations above 3 µM, possessing apparent 
and 
a.u. The nonspecific uptake rises linearly (R2 = 0.99) with BSA-FITC concentration yielding a slope 
of 1.45 ± 0.05 a.u./µM. The LEF-induced uptake via specific sites reaches saturation at concentrations above 6 µM. The apparent 
and 
for LEF-stimulated specific uptake amount to 1.31 ± 0.16 µM and 71.2 ± 3.0 a.u., respectively. The nonspecific LEF-stimulated uptake fits a linear dependence (R2 = 0.96) on BSA-FITC concentration having a slope 
of 7.60 ± 0.51 a.u./µM. The total uptake showed no saturation within the studied concentration range either in exposed or in unexposed cells.
Temperature dependence of electric field-induced adsorption and uptake
Temperature dependence of LEF-induced adsorption (Fig. 4 B) and uptake (Fig. 4 A) of BSA-FITC (6.8 µM) was investigated by flow cytometry, fluorimetry, and confocal microscopy. We performed LEF-induced internalization of BSA-FITC at different temperatures in the range of 4–37°C. Though the LEF-induced internalization was lowest at 4°C, it could not be completely abolished because of the trypsinization stage, which had to be performed at 37°C. Therefore, the formation of endocytic vesicles was not completely inhibited at 4°C, though further fusion events between vesicles were totally suppressed below 15°C. Thus, BSA-FITC was found to be located in the immediate vicinity of the membrane in both exposed and unexposed cells. However, exposure to LEF at 24°C in the presence of BSA-FITC revealed, by confocal microscopy, the formation of massively patched fluorescent patterns in the cytosol. LEF-induced uptake of BSA-FITC was found to be exponentially dependent on temperature (Fig. 4 A). The energy of activation, calculated for the constitutive uptake, was 2.9 ± 0.6 kcal/mol (R2 = 0.65) in the temperature range of 4–15°C and 8.1 ± 1.9 kcal/mol (R2 = 0.60) in the temperature range of 15–37°C. However, LEF-induced uptake at all temperatures between 4°C and 37°C was best fitted to a single line reflecting activation energy of 12 ± 0.6 kcal/mole (R2 = 0.96).
|
In contrast to LEF-induced uptake of BSA-FITC, the LEF-induced adsorption of the probe was found to be independent of temperature. LEF-induced adsorption of the BSA-FITC probe was determined by analyzing the fraction of probe that was removed from the cell surface after the proteolytic action of trypsin. Fig. 4 B shows the LEF-induced adsorption determined at 4°C, 15°C, 24°C, and 37°C, compared to control unexposed cells at each temperature. The obtained results demonstrate that there is no statistically significant change in the level of BSA adsorption over the whole temperature range of 4–37°C.
Dependence of adsorption and uptake on time after exposure to LEF
The time course of LEF-induced changes of cell surface leading to increased adsorption of BSA-FITC was examined by adding the probe before exposure or after exposure termination (at different time points in the range of 0–10 min). The study of the temporal adsorption of BSA-FITC to COS 5-7 cells, performed at 4°C (Fig. 5 A), shows a maximal 10.2-fold increase of adsorption when the probe was present during LEF treatment. When the probe was added to the exposed cells at different times after exposure, the total binding of the probe to the plasma membrane was only threefold higher than that of controls, a level that was preserved for up to 30 min after the termination of LEF, at 4°C.
|
The possibility that LEF induces changes in the probe's adsorption characteristics to the cell membrane was examined by exposing BSA-FITC only to LEF and then adding it to the cells. The adsorption of exposed BSA-FITC to unexposed cells was negligible and was not different from the adsorption of the unexposed probe to unexposed cells. Small but significant adsorption was detected when the exposed probe was added to exposed cells, yielding adsorption similar to that of the unexposed probe to exposed cells. These findings suggest that LEF-enhanced adsorption changes can be attributed to changes in the cell's surface rather than to conceivable electrochemical modification of the probe.
The decay characteristics of BSA-FITC uptake at 24°C were similar to those of BSA-FITC adsorption. The uptake was maximal when the probe was present during exposure and was 4.9- and 6.2-fold higher for COS 5-7 and HaCaT cells, respectively. Uptake increase of only 1.7- and 2-fold was obtained when BSA-FITC was added to COS 5-7 and HaCaT cells immediately after exposure to LEF. Similar to adsorption, the uptake persisted for up to 10 min after exposure to LEF.
Dependence of adsorption, uptake, and viability on electric field strength
The dependence of adsorption and uptake on electric field strength was examined by exposing cell suspensions of COS 5-7 and HaCaT cells in DMEM-H medium to different electric fields in the range of 2.5–20 V/cm (pulse duration 180 µs, 500 Hz, 1 min total time of exposure), in the presence of 6.8 µM BSA-FITC. Both adsorption and uptake demonstrate a linear dependence of LEF-induced uptake by COS 5-7 cells (Fig. 6 A) on electric field strength (correlation coefficients of 0.98 and 0.99 for adsorption and uptake, respectively). Similar results were obtained for HaCaT cells (Fig. 6 B). The slopes of the dependence of LEF-induced adsorption on electric field strength relative to that of LEF-induced uptake are 2.2- and 1.5-fold higher for COS 5-7 and HaCaT cells, respectively. The viability of COS 5-7 and HaCaT cells decreased linearly with the electric field in the range of 2.5–20 V/cm (Fig. 6, C and D). After exposure to the maximal electric field of 20 V/cm, the viability of COS 5-7 and HaCaT cells was 82% and 85%, respectively.
|
20–30 cells can be observed in the exposed cells compared to controls, where spontaneous aggregation occurs to a much lower extent, consisting mostly of two or three cells. The level of aggregation of COS 5-7 cells at different cell concentrations after exposure to LEF of 20 V/cm (frequency of 180 µs, pulse duration of 500 Hz) for 1 min was determined by a Coulter counter (Fig. 7 C). When cell concentration in suspension increased, elevated spontaneous formation of cell-cell aggregates took place. Increase of cell concentration from 0.3 x 106 to 3 x 106 in unexposed suspensions resulted in elevation of cell-cell aggregation from 1.5 ± 0.7% to 8.2 ± 4.5%. Exposure of cell suspensions to LEF of 20 V/cm leads to a much larger aggregation at all cell concentrations. Thus, increase in cell concentration from 0.3 x 106 to 3 x 106 resulted in elevation of cell-cell aggregation from 5% to 40% after exposure. At the higher concentration of 6 x 106 cells/ml 60% of cells appeared as aggregates. It should be stressed that due to the limit of the 100-µm orifice used in the Coulter counter instrument, large cell aggregates could not be measured. Thus our data represent a low estimation of the extent of LEF-induced aggregation.
|
33% is maintained provided the cells are brought into contact by centrifugation within 20 min after exposure. However, 30 min after exposure the ability of cells to form aggregates decreases to that of exposed but not centrifuged cells. The extent of cell-cell aggregation 60 and 120 min after exposure does not differ from controls. These findings suggest that LEF induces a long-lasting modification of cell surface which contributes to the increased "sticky" character of the cells toward each other.
|
|
Effect of pulse duration on BSA-FITC uptake
The dependence of LEF-induced uptake of BSA-FITC on pulse duration was studied by exposing COS 5-7 cells in DMEM-H medium for 1 min at 24°C while keeping all other parameters constant (electric field strength of 20 V/cm, current of 140 mA, frequency of 500 Hz). Increasing pulse duration from 50 to 250 µs resulted in a corresponding increase in the uptake of BSA-FITC into cells by up to fivefold (Fig. 9 B). The LEF-induced uptake showed a sigmoidal dependence on pulse duration (Fig. 9 F). Viability of COS 5-7 cells exposed to LEF of 20 V/cm decreased linearly down to 85% upon lengthening pulse duration up to 250 µs.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
); and 2), in situ "lateral electrophoresis" of charged proteins and lipids in the plane of the cell membrane (Poo, 1981).
The exposure of a cell to external electric field leads to the induction of a transmembrane polarization (
) given by (Schwan, 1957, Farkas et al., 1984)
 | (1) |
is the angle between the radius vector to any point in the plasma membrane and the vector of the electric field. Thus, exposure of a cell possessing radius of 
8 µm to the electric field in a range of 2.5–20 V/cm (used in our studies when employing highly conductive media) will result in an induced potential difference 3–24 mV across the plasma membrane. Therefore, the membrane region facing the anode will be hyperpolarized by 3–24 mV; that facing the cathode will be depolarized to the same extent. Though these changes in the transmembrane potential do not lead to electroporation (Rosemberg and Korenstein, 1997; Antov et al., 2004), the possible involvement of the induced transmembrane potential in the uptake process cannot be ruled out. It may alter ion fluxes across the membrane, either through modulation of ion channels or via electric field-induced conformational changes of transport systems. The latter has been suggested by numerous studies that formulated an electroconformational coupling model (Chen and Tsong, 1994, Xie et al., 1997, Tsong, 2002, Tsong and Chang, 2003). The model implies that low electric fields can drive oscillations or fluctuations of enzyme conformation between two states. However, since changes in transmembrane potential difference have not been previously reported to directly cause cell surface alterations, they cannot explain the observed LEF-induced enhancement of BSA binding to the cell surface, which occurs even at low temperatures, or the measured increase in cell aggregation.
The other possible primary effect induced in cells exposed to low electric fields is that of electrophoretic induced segregation of labile charged lipids and proteins in the plane of the cell membrane (Fig. 10). This phenomenon was previously explored both theoretically and experimentally (e.g., Jaffe, 1977; Poo and Robinson, 1977; Poo et al., 1979; Poo, 1981; McLaughlin and Poo, 1981; Sowers and Hackenbrock, 1981). It has been pointed out that the external electric field tangential to the cell surface (E) is the driving force (Eq. 2):
 | (2) |
will induce electrophoretic mobility toward the anodic or cathodic sides of the cell, either by direct electrophoretic mobility of the negatively charged components or by electroosmosis, respectively (McLaughlin and Poo, 1981). Though most studies of lateral electrophoretic segregation of charged membrane components have been carried out on adherent cells employing low DC electric fields (e.g., Poo, 1981; Wang et al., 2003), it was shown that a similar segregation phenomenon could be induced in photosynthetic membrane vesicles of a size comparable to that of cells (Brumfeld et al., 1989; Miller et al., 1994), when exposed in suspension to a train of unipolar pulsed electric fields of duration much shorter than the rotational time of the vesicles. Under these conditions the membrane vesicle can be considered to be in a rotational stationary state, at which time the lateral electrophoretic displacement takes place. Similarly, taking into account that the mean average radius of a COS 5-7 cell is 8 µm, its rotational time constant, according to the Stocks-Einstein equation, is 7.5 min. Thus, the exposure to LEF in all our experiments was restricted to 1 min so that it would be much shorter than the rotational diffusion relaxation time of the cell. The short exposure time to LEF is another feature that sets our study apart from previous ones. Exposure to DC electric field of 10 V/cm for 45 min to 3 h was shown to induce asymmetric distribution of acetylcholine receptors in embryonic muscle cells (Orida and Poo, 1978). Similar exposure to DC electric fields of 10 V/cm applied for 30 min induced asymmetry in the distribution of Con A receptors in Xenopus muscle cells (McLaughlin and Poo, 1981). In addition, cathode-directed accumulation of epidermal growth factor receptors in corneal epithelial cells was shown to occur after exposure to 3 V/cm for 20 min, whereas similar accumulation of fibroblast and transforming growth factor receptors occurred when exposed to a DC electric field of 1.5 V/cm for 12 h (Zhao et al., 1999). It emerges from these studies that global segregation of membrane receptors, reflected by their asymmetric distribution between the cell hemispheres facing the cathode's or anode's sides, occurs only after relatively long exposure time. Therefore, in the case of a brief exposure we may expect the induction of limited short-range electrophoretic mobility leading to localized segregation of membrane proteins and lipids. It can be proposed that the formation of small receptor aggregates formed after short exposures is reversible differing from the irreversible formation of large aggregates created by prolonged exposure to electric fields (Poo, 1981). This proposal is in agreement with our finding of the transient ability of the cells to form aggregates after LEF termination.
|
LEF stimulates adsorption of macromolecules to the cell surface
This study demonstrates that a short exposure to LEF leads to electric field-driven enhancement of adsorption of macromolecular solutes of different chemical nature to the cell membrane. Employment of three different types of macromolecules, BSA-FITC (66 kDa), dextran-FITC (2000 kDa), and DNA (4.7 kb; 3102 kDa) leads to their LEF-enhanced adsorption to the cell surface. Though BSA-FITC and dextran-FITC possess different molecular weights, dissimilar negative charges (42 and 222, respectively, per molecule at pH 7.4), various shapes, and different hydrophilic/hydrophobic characteristics, they show comparable LEF-induced adsorption to the cell surface of 10- and 10.2-fold higher, respectively, than that in controls. However, the LEF-stimulated adsorption of the negative fluorescent DNA-PI complex or its positive tertiary complexes DNA-PI-DEAE and DNA-PI-PEI is only 1.4, 1.9, and 1.6-fold higher, respectively, than that of controls. This lower adsorption of DNA as compared with that of dextran of a similar molecular weight may be attributed to the higher surface charge density of DNA and its lower flexibility. These findings suggest that LEF-enhanced adsorption to the cell surface is a characteristic common to different macromolecules. Furthermore, the adsorption of a molecule to the cell surface is determined by many different variables, such as electrostatic and Van der Waals forces, hydrophobicity, molecular geometry of the interacting molecule, the heterogeneous character of the cell surface, and the existence of specific binding sites. Thus, due to the complexity of this system it is not yet possible to quantitatively discuss the relationship between molecular characteristics and the extent of LEF-induced adsorption to the cell surface of a target macromolecule.
The enhancement of adsorption by LEF can be ascribed to two main processes. The first one is the LEF-induced increase in the velocity and rate of collision between the macromolecular solutes and cells. The second process is the LEF-induced alteration of the cell surface, probably through electrophoretic lateral segregation of membrane components. Discrimination between these two processes emerges from examining the kinetic patterns of BSA adsorption to the cell surface (Fig. 5). A threefold higher adsorption is obtained when BSA is added to cells before their exposure to LEF as compared to BSA addition immediately after the termination of exposure. This increase may be attributed to the difference between the electrophoretic mobilities of cells and macromolecular solutes as well as to their accumulation near the anode, which is expected to increase the collision rate between them during exposure, thereby leading to a higher extent of adsorption. At the same time LEF induces changes in the cell membrane, resulting in a cell surface more sticky toward macromolecules. Although elevated collision rate takes place only during LEF treatment, cell surface alteration also persists after LEF termination. Thus, comparing the adsorption levels in the presence and absence of electrophoretic forces (when BSA is added before or after exposure, respectively) allows us to roughly estimate that 70% of the increase in the total adsorption is attributed to the electrophoretically enhanced collision rate between the altered surface of the cells and the solute macromolecules. The remaining 30% of the total adsorption is attributed to a diffusion rate-limited adsorption of the macromolecules to the electrically modified surface of the cells.
The stimulation of cell surface alteration by LEF seems to be temperature-independent as reflected by similar levels of induced adsorption of BSA-FITC at 4°C and 24°C immediately after the termination of exposure (compare Fig. 5, A and B). This finding is in agreement with the weak dependence (Q10 of 1.2) of membrane receptor segregation by DC electric fields on temperature (Poo, 1981). However, reversibility of the LEF-induced surface alterations is temperature-dependent. The LEF-induced elevated adsorption of BSA-FITC decays to half maximal value within 2 min at 24°C, whereas at 4°C it is maintained for at least 30 min (Fig. 5). Thus it should be pointed out that the LEF-enhanced adsorption taking place at 4°C can be considered to occur under equilibrium conditions, whereas that occurring at elevated temperatures is dominated by kinetic aspects.
Additional evidence for the LEF-induced alteration of the cell membrane stems from the observation of cell-cell aggregation. (Figs. 7 and 8). This aggregation takes place both in the absence and presence of macromolecular probes in the external medium, suggesting that the cell membranes themselves, and not the probes, are primarily modified by LEF. Thus, the dependence of LEF-induced cell-cell aggregation on cell density originates from both enhanced adsorptive properties of the cell surface and the diffusion-limited collision rate. Naturally, if the cell population possesses heterogeneous values of electrophoretic mobilities, an additional electrophoretic contribution to cell aggregation is expected to be proportional to the width of the distribution histogram of the electrophoretic mobilities. Moreover, cells treated by LEF in dilute suspensions (5 x 105 cells/ml), when brought in close contact by centrifugation, preserve their ability to form aggregates for up to 15–20 min after the exposure, at 24°C. However, with time the level of cell aggregation diminishes. The LEF induction of cell-cell aggregation decays to its half-maximal value within 30 min at 24°C, demonstrating again that the LEF-induced surface modification is a reversible one.
Electric field-induced changes, leading to increased adsorption properties of the cell surface, have been previously reported when employing very high electric fields. The exposure of suspension of cells and large unilamellar vesicles to high short electric pulse (100 µs of 3–4 kV/cm) led to increased cell-liposome adhesion (Chernomordik et al., 1991). This electric field-induced higher affinity for liposomes was found to possess a lifetime of several minutes, similar to the LEF-induced adsorption changes, suggesting that both LEF and high electric fields may share a common feature in modifying the cell surface.
LEF enhances specific and nonspecific adsorption of BSA to the cell surface
More insight into the LEF-driven adsorption of macromolecules to cells was gained by an extended study of the characteristics of LEF-enhanced adsorption of BSA to COS 5-7 cells. Analysis of the adsorption isotherms of BSA-FITC to the surface of COS cells at 4°C, assuming they occur at equilibrium, shows that the total adsorption is characterized by specific and nonspecific binding in both cells subjected to LEF and in unexposed control cells. The specific adsorption may be described by a classical one-binding site model with saturation at BSA concentration 
5 µM. The nonspecific binding fits a linear rise with concentration. The affinity of specific binding does not change significantly after exposure to LEF possessing 
and before and after exposure, respectively. However, the number of specific binding sites (
) increases dramatically by 6.5-fold after LEF treatment. We can speculate as to the identity of the specific binding sites for BSA in COS 5-7 cells by considering the characteristic albumin receptors that exist in similar cells. Taking into account the kidney origin of the COS 5-7 cell line, these receptors could possibly be identified as cubilin and megalin (Christensen and Birn, 2002), though the presence of other receptor types such as gp60, gp31, and gp18 cannot be excluded (Tiruppathi et al., 1997; Schnitzer and Bravo, 1993). The reported Kd of these receptors for albumin is in the range of 0.3–0.6 µM (Gekle et al., 1996; Birn et al., 2000;Yammani et al., 2002), which is in good agreement with the value of 0.51 µM obtained for unexposed COS 5-7 cells. The absence of significant changes in the Kd values after exposure to LEF suggests that membrane receptors do not undergo LEF-induced substantial structural change sufficient to alter their affinities. However, the significant elevation of the capacity of the cell surface for specific binding after exposure suggests that LEF treatment of cells leads to exposure of binding sites on the cell surface that were unavailable for binding before the LEF treatment. The possibility that exposure to LEF brings about the incorporation of additional receptors from the cytoplasm (LEF-stimulated exocytosis) can be discarded, since it is not plausible for this process to take place at 4°C. An attractive explanation may be put forward in view of recent findings showing strong changes in local dipole potential at the cell membrane interface due to human serum albumin binding (O'Shea, 2003). Treatment of cells with cholesterol or 6-ketocholestanol was shown to lead to increased membrane dipole potential and augmentation of binding reaction with albumin via higher binding capacity without major changes in the affinity (Asawakarn et al., 2001; O'Shea, 2003). Since the employed LEF is sufficient to alter the local dipole potential at the membrane-water interface, it may cause an increase in the number of receptors available for BSA binding. Another possible explanation may be attributed to segregation of charged lipids and membrane proteins along the cell surface taking place upon exposure to electric fields (Jaffe, 1977; Poo, 1981). This notion is in agreement with the findings that cationized ferritin substantially increases binding of albumin to cultured aortic smooth muscle cells (Sprague et al., 1985). The effect was attributed to cationized ferritin-triggered segregation of surface anionic sites resulting in intervening areas essentially devoid of negative charge, thus reducing the electrostatic repulsion between negatively charged human serum albumin and cell surface. Since the possible segregation of plasma membrane receptors is a nonselective process we would like to speculate that the observed phenomena is a general one, not confined to BSA receptors only.
The nonspecific binding of BSA-FITC is also dramatically altered by LEF. The nonspecific components of binding for both control and exposed samples seem to rise linearly with BSA-FITC concentration. However, the slope of the LEF-induced nonspecific binding is 33-fold higher than that of control. It may be proposed that this low affinity and high capacity of BSA binding to the cell surface is associated with hydrophobic interactions of lipids in the outer leaflet of the plasma membrane with albumin molecules. There are 11 hydrophobic pockets known in human serum albumin that are responsible for the binding of fatty acids (Hamilton, 2002). Albumin was reported to preferentially adsorb onto hydrophobic surfaces (Ying et al., 2003; Martins et al., 2003) and to neutral liposomes (Yokoyama et al., 2002). It may be argued that decreased repulsion and increased binding between BSA and the cell surface will be attained when the negatively charged molecular components, exposed to the extracellular milieu, are unevenly distributed along the cell surface due to lateral electrophoresis in the membrane plane. Electric fields of 10–30 V/cm, similar to the ones used in this study, were shown to induce lipid segregation when applied tangentially to a supported bilayer consisting of charged and uncharged lipids (Groves et al., 1997). The measured velocity of charged lipids in the supported lipid bilayer was 6 µm/min (Groves et al., 1998), which is comparable to a mean radius of 8 µm for cells used in our study. External inhomogeneous electric fields applied to cholesterol-phospholipid monolayers were shown to lead to phase separations under appropriate conditions (Radhakrishnan and McConnell, 2000). Thus, LEF-induced segregation between charged and uncharged components on the cell surface is expected to increase nonspecific BSA binding to lipid areas depleted of the anionic charged components.
Low electric fields enhance uptake
On the relationship between LEF-induced adsorption and uptake
The stimulation of uptake via adsorptive endocytic pathway(s) requires, at the first stage, enhanced adsorption to the cell surface. To be able to differentiate between LEF-stimulated processes of adsorption and uptake we made use of the finding that whereas adsorption is independent of temperature, uptake is a temperature-dependent process. We demonstrate that LEF-enhanced adsorption carried out at 4°C is consequently followed by enhanced uptake upon increasing the temperature to 24°C (Fig. 1, A–H), suggesting that the two processes are consecutive. Furthermore, the association of these two processes is compatible with the finding that both processes decay at similar rates at 24°C (Fig. 5, B and C). These similar kinetics do not depend on the cell type and were demonstrat
