| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Biochemistry and Biophysics |
Laboratory of Cell Biology, National Cancer Institute, NIH, Bethesda, Maryland 20892-4254 [C. A., J-P. A., X-J. L., C. O. C., A. A., M. M. G.]; FACS Core Facility, National Cancer Institute, NIH, Bethesda, Maryland 20892-4255 [B. T.]; and Instrumentation and Biophysics Branch, Center for Food Safety and Applied Nutrition, College Park, Maryland 20740-3835 [J. J. Y.]
 |
ABSTRACT |
|---|
 Top ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
4 h, and the protein had a half-life of 16 h. P-gp cell surface expression (measured with monoclonal antibody MRK-16) and P-gp function (measured with a fluorescent substrate, rhodamine 123) was characterized by using fluorescence-activated cell sorting. No differences in membrane potential using the fluorescent probe oxonol or in membrane "fluidity" using fluorescent anisotropy probe or electron spin resonance probe were observed in the tet-repressible P-gp-expressing cells. In contrast, several drug-selected cells that express P-gp showed an increase in membrane fluidity and membrane potential. These results suggest that expression of P-gp per se has little effect on membrane fluidity or membrane potential, and it does not have H+ pump activity. The changes in these parameters observed in drug-selected cells must reflect other host adaptations to drug selection. |
INTRODUCTION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
In the laboratory, many different cell lines have been used to demonstrate effects of P-gp on the plasma membrane, on drug resistance, on apoptosis, and on resistance to viruses in an effort to elucidate the physiological role of P-gp (6, 7, 8, 9, 10, 11, 12) . It is not surprising that results obtained from different cell lines and expression systems are not consistent with each other. The question arises whether the results obtained are because of the selection pressure on the cells, or other unknown phenomena that occur in cells that have been in culture, in some cases, for a number of years.
We have developed one mouse and one human cell line in which human MDR1 is under tetracycline control. In these cells, addition of tetracycline rapidly turns off transcription of MDR1 mRNA, and over a period of a few days P-gp disappears from the cells, allowing a comparison of the same cells with and without P-gp in the plasma membrane. In this study, we characterize the P-gp mRNA half-life, P-gp protein half-life, and measure membrane properties, such as potential and fluidity.
To assess plasma membrane biophysical characteristics, such as fluidity, of normal and tumor cells, ESR was used. ESR was used previously to measure membrane fluidity of transplanted melanoma cells (13) , human lung tumor and normal tissues (14) , and cisplatin-sensitive and -resistant cells (15) . Also, high and low metastatic Lewis lung cancer cells were shown to have different plasma membrane fluidity by using ESR (16) . Fluidity of plasma membranes was also studied using lipid soluble fluorescence probes, such as DPH (15) . Wanten and Naber (17) used DPH and TMA-DPH in their studies, which we used in our investigation here. Measurement of plasma membrane potential is indicative of the general biological status of cells. For example, alteration of membrane connected proteins changes membrane potential (18 , 19) , and antiproliferative signals induce plasma membrane depolarization in lymphocytes (20) . Using these techniques, we show that drug-selected cells frequently show changes in membrane properties, whereas P-gp expression alone does not affect these properties.
|
MATERIALS AND METHODS |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Cell Lines and Cell Culture.
All of the Tet-Off cells were cultured in DMEM high glucose (Life Technologies, Inc.) with 10% tetracycline-approved FBS (Clontech), and 2 mM L-glutamine (Quality Biological, Inc.) at 37°C with 5% CO2. HeLa Tet-Off cells (Clontech) are the parental cells of HeLa MDR-Off. HeLa P-gpon cells refer to HeLa MDR-Off cells that were grown without colchicine for at least 4 days. HeLa P-gpoff cells refer to HeLa MDR-Off cells that were grown in the presence of 2 µg/ml tetracycline and are not expressing P-gp. The nondrug-selected cells KB-3-1 (a subclone of HeLa), and drug-selected KB-8-5, KB-C1, KB-V1, KB-A1, and KB-A1rev (21
, 22)
were used in this study. Cells selected with colchicine are KB-8-5 (22)
and KB-C1 (21)
. Cells selected with vinblastine are KB-V1, and cells selected with doxorubicin are KB-A1. KB-A1rev cells are KB-A1 grown without drug selection, have lost their resistance, and are, thus, considered revertants. 77.1 cells were a generous gift from Dr. Alfred Schinkel (The Netherlands Cancer Institute, Amsterdam, The Netherlands) and are mouse embryo fibroblasts derived from mdr1a/1b knockout mice (5)
used to make the 77.1 MDR-Off cell line. 77 P-gpon refers to 77.1 MDR-Off cells grown without colchicine for at least 4 days. 77 P-gpoff refers to 77.1 MDR-Off cells grown in the presence of 2 µg/ml tetracycline for 4 days.
Stable Transfection.
The plasmid containing the human MDR1 gene, pRevTRE-MDR1 and the plasmid containing the transactivator, pRevTet-Off, were cotransfected into 77.1 cells. The pRevTRE-MDR1 plasmid was transfected into HeLa Tet-Off cells that are already stably transfected with pRevTet-Off. Using the LipofectAMINE PLUS system (Life Technologies, Inc., Rockville, MD), HeLa Tet-Off cells were incubated with 4 µg pRevTRE-MDR1, 20 µl PLUS reagent and 30 µl LipofectAMINE for 3 h in 100-mm dishes (Falcon). 77.1 cells were incubated with 4 µg pRevTRE-MDR1 and 4 µg pREVTet-Off along with the LipofectAMINE and PLUS reagent as described above. The cells were then washed and were allowed to recover overnight in DMEM supplemented with 10% tetracycline-approved FBS (Clontech) and 2 mM L-glutamine. The next day, 20 ng/ml colchicine was added to the medium for selection. Ten days later, colonies were picked and transferred to 12-well plates. Seven days later, the cells were transferred to T25 flasks, and assayed for the expression and function of P-gp.
P-gp Membrane Expression FACS Assay.
Cells were treated with 2 µg/ml tetracycline for 0–4 days. Subsequently, these treated cells were trypsinized daily from their flasks, harvested, and washed in Iscove’s Modifed Dulbecco’s Medium supplemented with 5% FBS. Cells were then resuspended in 200 µl of IMDM/5% FBS containing 5 µg of MRK16, a mAb specific for surface epitopes of human P-gp. After initial incubation at 37°C for 20 min the MRK16-treated cells were washed and then incubated with a secondary antibody, FITC antimouse IgG2a (PharMingen), for 20 min. The antibody-treated cells were washed and then analyzed on a FACSCalibur (Becton Dickinson), operated with a 15 mW argon ion laser tuned to 488 nm excitation wavelength. Fluorescence emission was collected at 525 nm. Controls used in the FACS analysis were cells that had not been incubated with antibody, cells that had been incubated with an isotype standard mouse IgG2a followed by FITC antimouse IgG2a, and cells incubated with FITC antimouse IgG2a alone. Control profiles were similar to cells alone without any antibody incubation. Triplicate samples were assayed, and their histograms were analyzed. The median fluorescence was plotted on a log scale, and the half-life of the protein was calculated by using the equation: y = 100e-0.0442x where y is percentage of expression and x is time in tetracycline. R2 = 0.9879 for this equation.
P-gp Function FACS Assay.
Cells were treated with tetracycline and trypsinized from the flasks as described above, and were resuspended in 5 ml IMEM with 5 µM cyclosporin A for 10 min at 37°C. Rhodamine 123 was added to a final concentration of 0.5 µg/ml to cells both treated and untreated with cyclosporin A for 20 min at 37°C. Cells were then washed and analyzed on a FACSCalibur (Becton Dickinson). Controls were cells not incubated with cyclosporin A or rhodamine 123. Triplicate samples were measured, and the histograms were analyzed. The median fluorescence was plotted against the time in tetracycline on a log scale. The data were presented as accumulation of rhodamine 123, which directly reflects the efflux of the substrate and, thus, the function of P-gp.
Purification of RNA.
Five x 106 cells treated with or without tetracycline were centrifuged at 1300 rpm, 4°C, for 5 min. The pellet was stored at -70°C. The Qiagen RNeasy kit was used (Qiagen, Valencia, CA) to extract RNA. The cells were lysed and homogenized with a 20-gauge syringe in the lysis buffer provided. RNA purity was confirmed by a 260 nm:280 nm ratio of 2.1 on a Pharmacia Biotech Ultraspec 3000 in 1x 10 mM Tris (pH 8)-1 mM EDTA buffer (Cambridge, United Kingdom). RNA was then electrophoresed on 1.25% agarose/formaldehyde gel. RNA was stored at -70°C at a 1 µg/µl concentration.
Western Blot.
Frozen cell pellets from cells treated with or without tetracycline were thawed and dissolved in a lysis buffer containing 10 mM Tris, 0.1% Triton X-100, 10 mM MgSO4, 2 mM CaCl2, 1% aprotinin, 1 mM 4-(2-aminoethyl) benzenesulfonylfluoride, 2 mM DTT, and 20 µg/ml DNase. Samples were incubated at 37°C for 5 min and then frozen in dry ice. Samples were then thawed and sonicated. Cell lysates were mixed with Novex Tris-Glycine SDS sample buffer (Invitrogen) and incubated at 37°C for 30 min. Protein was quantitated using a BCA assay (Pierce). Three µg of protein was loaded on a Novex 4–20% Tris-glycine gel (Invitrogen). Protein was transferred to a 0.45 µm pore size nitrocellulose membrane for 4 h at 200 milliAmps using a Mini-PROTEAN II Cell apparatus (Bio-Rad). Membranes were then blocked overnight in a 20% milk, 0.5% Tween 20, PBS buffer. Membranes were washed with 0.5% Tween 20, PBS buffer, and then incubated with the C219 primary antibody (Fujirebio Diagnostics, Inc.). C219 is a mAb that reacts with an internal membrane epitope of human P-gp, and 6C5 is a mAb against rabbit GAPDH (Research Diagnostics, Inc). A 1:2000 dilution of C219 and 1 µg/ml 6C5 were added to a 5% milk, 0.5% Tween 20, PBS buffer, and incubated with the membrane at room temperature for 2 h on a shaker. The membrane was then washed in the 0.5% Tween 20, PBS buffer. The secondary antibody is a peroxidase-labeled mouse IgG (KPL). A 1:5000 dilution of the secondary antibody was made into a 5% milk, 0.5% Tween 20, PBS solution and incubated with the membrane for 45 min at room temperature on a shaker. Enhanced chemiluminescence was performed as directed by the manufacturer (Amersham Pharmacia Biotech).
Real-Time PCR.
Real-time PCR was performed with a Light Cycler DNA SYBR Green kit (Roche Biochemicals, Indianapolis, IN). The reaction was in a 20 µl final volume with 0.5 µg of purified total RNA, 4 µl PCR mix provided by the manufacturer, 4 µl MgCl2 (25 mM), 1 µl of each primer (10 µM), 0.4 µl of enzyme mix and DEPC-H2O. Primer sequences for the genes amplified were as follows: ABCB1 (5'-GGG-ACC-GCA-ATG-GAG-GAG and 5'-GCA-CCA-ATT-CCA-CTG-TAA), ABCB1-UTR-(5'-TGA-CAT-TTA-TTC-AAA-GTT-AAA-AGC-A and 5'-TAG-ACA-CTT-TAT-GCA-AAC-ATT-TCA-A), and GAPDH (5'-CCC-TTC-ATT-GAC-CTC-AAC-TAC-AT and 5'-ACG-ATA-CCA-AAG-TTG-TCA-TGG-AT). The RT reaction was performed at 55°C for 20 min. cDNA generated by the RT step was denatured at 95°C for 20 s. Amplification of the cDNA was done in 45 cycles of 95°C for 5 s, 58°C for 10 s, and 72°C for 13 s. Fluorescence was recorded during the elongation phase at 72°C. Determination of the number of cycles it took to see a PCR product or Crossing point (Cp) was performed on the basis of the second derivative algorithm. As a control for quantitative PCR, a standard was made with increasing copy number of pRevTRE- MDR1 plasmid from 105 to 1010 copies to calculate the number of cycles needed to double the amount of RNA. The equation used was Cp = 3.9 log10 (copy number) + 49.08. The number of cycles calculated was 1.17 to double the amount of RNA in the real time RT-PCR assay. Real-time PCR of endogenous MDR1 and total MDR1 was done on the tetracycline time course samples, and they were all normalized to GAPDH. 
Cp are Cp values that have been normalized to GAPDH. The deviation of Cp of certain time points (t) from the Cp when time of tetracycline is zero (t = 0) was calculated and plotted against the time of tetracycline treatment.
Drug Resistance Assay.
Colony formation assays were used to determine the colchicine, doxorubicin, and vinblastine sensitivity of the HeLa MDR-Off cell line relative to the parental cell line, HeLa Tet-Off, after a 4-day incubation in the presence or absence of 2 µg/ml tetracycline. After their initial 37°C incubation in T25 flasks, the monolayers were trypsinized, and a Coulter Counter was used to establish the total number of cells per flask. Three hundred exponentially growing cells from each flask were seeded into 60-mm dishes containing DMEM ± 2 µg/ml of tetracycline. The cells were allowed to attach for 16 h before increasing concentrations (i.e., 0, 0.1, 0.3, 1, 3, 10, 30, 100, 300, 1,000, 3,000, 10,000, and 30,000 ng/ml) of colchicine, doxorubicin, or vinblastine were added to the existing medium. Dose response curves were generated by allowing the cells to form colonies over an additional 8–10-day incubation before staining with 0.5% methylene blue in 50% ethanol and counting them with an IS-1000 Digital Imaging System. For each cell line, the percentage of the relative cloning efficiency per drug concentration was determined by dividing the number of colonies counted after drug treatment by the number of colonies formed with zero drug treatment. Percentage points were plotted to determine the LD10 drug concentration, which reduced the cloning efficiency of the transfectants to 10% of the control without drug. Relative resistance was determined by dividing the LD10 value of the resistant line by the LD10 value of the parental line.
Membrane Potential Measurements.
Membrane potential measurements were done by flow cytometry using the negatively charged DiBaC4 (3)
oxonol dye (Molecular Probes, Eugene, OR), essentially as described earlier (23)
. Briefly, a cell suspension of 1 x 106 cells/ml was equilibrated for 1 min in PBS followed by 150 nM oxonol dye. After a 2-min equilibration at room temperature, data were collected and displayed as histograms from 104 cells/sample. Reproducibility was determined by measuring membrane potential of the cells on different days from separate cultures. Oxonol fluorescence intensity measures membrane potential when the extracellular potassium concentration is changed from 5 mM to 150 mM, resulting in an increase of the fluorescence intensity of the oxonol stained cells because of their depolarization. All of the measurements were made with a Becton Dickinson FacsCalibur flow cytometer (Becton Dickinson, Mountain View, CA), operated with a 15-mW argon ion laser tuned to 488 nm excitation wavelength. Fluorescence emission was collected at 525 nm. Results are expressed in comparative histograms of a representative series of cells.
Polarity of Fluorescent Membrane Probes.
Measurements of steady-state fluorescence polarization were done with a spectrofluorometer LS50B (Perkin-Elmer, Norwalk, CT) and the lipid soluble fluorophore, TMA-DPH (Molecular Probes). This fluorophore is known to probe membranes at the surface. TMA-DPH was dissolved in tetrahydrofurane at a concentration of 2 mM and was kept in the dark at 4°C. Cells (106/ml) were labeled with TMA-DPH at a concentration of 2 µM in PBS. After a 10-min incubation at 4°C, the cell suspension was centrifuged and washed twice in the centrifuge with PBS. After resuspension in PBS, fluorescence anisotropy was measured at 30°C. The excitation wavelength was 355 nm, and the emission was measured at 430 nm with a slit width of 5. The polarization was calculated according to Collins et al. (24)
by the equation P = (I0,0-G*I0,90)/(I0,0+G*I0,90) from the measured fluorescence intensities. Another polarization probe, AOSA, (Molecular Probes), was also used to measure polarization of cells. This probe can assess membrane fluidity six carbons deep in the plasma membrane, whereas TMA-DPH probes at the surface of the plasma membrane. The same experimental conditions were used with this polarization probe as with TMA-DPH, given above, except that 360 nm excitation and 470 nm emission wavelengths were used.
ESR Studies on P-gp Expressing and Nonexpressing Cells.
ESR studies were conducted with 5-doxyl-SA and with T-SASL probes (Molecular Probes). The 5-doxyl-SA probe intercalates to a depth of five carbons in the outer leaflet of the plasma membrane (25)
, and T-SASL probes the surface of the plasma membranes (26)
. Labeling the cells with the spin probes was done as follows: 1 mg/ml 5-doxyl-SA dissolved in ethanol and 2 x 107 cells in a 20 µl volume were mixed with 8 x 10-8 mol spin label. After 1 min of contact time, the cell suspension was transferred into a 50-µl micropipette capillary tube and sealed at the bottom with Critoseal (Syva Co., Palo Alto, CA). The micropipette with the cells was placed into the cavity of a Varian E-9 Century series spectrometer (Palo Alto, CA). ESR spectra were recorded at X-band, at 9.5 kHz, 100 field modulation, 4 G modulation amplitude, 100 G sweep range, and at 10 mW microwave power. The temperature of the probe was set to 24°C by the variable temperature accessory, using N2 gas flow. Evaluation of the obtained ESR spectra, when 5-doxyl-SA was used, was by the equation expressing the order parameter, S = 0.5407 (T11 - T1)/a0, where a0 = (T11 + 2T1)/3, and T11 and T1 are the outer and inner tensors obtained from the ESR spectra. When the T-SASL probe was used, the same instrument parameters were applied, except that the incident microwave power was 20 mW. The spectral parameters, h0 and h-1 represent spectral amplitudes and the ratio h0:h-1 defines the motional freedom of the probe according to Yin et al. (26)
. With both spin labels the ESR spectra show contributions from spin labels of restricted motion with no contribution from free-moving spin label.
Lipid Packing of P-gp Expressing and Nonexpressing Cells.
Plasma membrane lipid packing can be studied by inserting the fluorescence probe merocyanine 540 (Molecular Probes) into cell membranes and assessing the degree of insertion as determined by fluorescence intensity using flow cytometry (27
, 28)
. The experiment was according to Smith et al. (27)
with some modifications. Briefly, 106 cells suspended in 1 ml of PBS were treated with 10 µl of an MC540 stock solution at 1 mg/ml in 60% ethanol and 40% water. After a 10-min incubation at room temperature cells, were pelleted in a centrifuge, resuspended in 1 ml PBS, and fluorescence histograms were obtained. A Becton Dickson FacsCalibur flow cytometer was used at 488 nm excitation and 575 nm emission wavelength. Histograms were collected with 104 cells. Results are expressed as means of fluorescence of the histograms.
Determination of the Number of P-gp Molecules on Cell Membranes.
Determination of the number of P-gp molecules on the plasma membranes of cells was done according to the protocol of Bang Laboratories, Inc. Quantum 26 was as detailed by Weaver et al. (29)
. Standard fluorescent beads were obtained from Bang Laboratories (Fisher, IN). The supplied beads contained defined numbers of fluorescein molecules: 6318, 15877, 53589, 82914, 123338, 170473, 353992, and 437815. Fluorescence intensities of the beads were evaluated by flow cytometry using a FacsCalibur instrument. Beads were shaken vigorously before use and then mixed in pairs. Histograms were obtained by flow cytometry. Using the same instrument settings, the fluorescence intensity, means of histograms of cells labeled with MRK-16 mAb, and FITC-labeled antimouse IgG2a mAb (PharMingen, Becton Dickinson) were determined. Beads and the cells were suspended in the same PBS buffer for cytometry.
A standard curve was produced from the means of the histograms obtained with the beads, using semilogarithmic paper or by using QuickCal v2.1 data analysis software (Bangs Laboratories). Results of the two evaluations coincided. Mean fluorescence of the histograms obtained with the labeled cells was plotted on the standard curve, and numbers of P-gp molecules were read from the Y-axis. Consecutive determinations gave similar results.
|
RESULTS |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
. Fig. 1B
shows the data of the PCR done on the time course. Each point was normalized to its own GAPDH as a control for the RT-PCR reaction. The deviation of Cp at different times of tetracycline treatment from the Cp, where time of tetracycline treatment is equal to zero, was plotted against the time of tetracycline treatment to calculate the half-life of the message (Fig. 1B)
. The RNA half-life of this transcript was calculated to be 3.8 h. It should be pointed out that although these cells were selected with colchicine during their construction to select for the presence of the pRevTRE-MDR1 plasmid, the endogenous levels of MDR1 were not elevated by this selection.
|
. Percentage of rhodamine 123 accumulation is an inverse indication of P-gp function. Higher levels of rhodamine in the cells indicate decreased efflux of P-gp, or in this case decreased expression of P-gp. By the third day of tetracycline treatment, P-gp function was not detected, as indicated by the high accumulation of Rhodamine 123 in the cells. All of the cells were also treated with a P-gp inhibitor, cyclosporin A, to verify that that the assay was measuring P-gp function (data not shown).
|
. Cells treated with tetracycline for 1–4 days show a steady decline of P-gp expression. Isotype controls were used, and the fluorescence after 4 days treatment with tetracycline was indistinguishable from cells not treated with any antibody or the internal fluorescence of the cells (data not shown). The decline of P-gp expression was plotted on a log scale, and the protein half-life was calculated to be 15.86 h. Similar results are observed in the mouse 77.1 Tet-Off cell line (data not shown). Western blots were used to confirm the results observed in the FACS assays (Fig. 2C)
. HeLa P-gpon cells were treated with 2 µg/ml tetracycline for 1 day, 2 days, 3 days, and 4 days. GAPDH antibody was used as a protein loading control. On the basis of these results, for the rest of the experiments day 0 cells were not incubated with tetracycline and are denoted as HeLa P-gpon and 77.1 P-gpon. Day 4 cells were incubated with tetracycline for 4 days and are denoted as HeLa P-gpoff and 77.1 P-gpoff.
HeLa P-gpon Cells Are Multidrug Resistant.
HeLa MDR-Off cells were either grown in 20 ng/ml colchicine (HeLa P-gpcol), 2 µg/ml tetracycline (HeLa P-gpoff), or no drug (HeLa P-gpon) for at least 4 days before starting the cloning assay. The cells grown in colchicine are expressing as much P-gp as the ones grown out of drug; the nomenclature is used to specify what medium was used. Colony-forming assays with the cell lines HeLa P-gpcol, HeLa P-gpon and HeLa P-gpoff, were done, individually, with increasing concentrations of three known MDR substrates: colchicine, doxorubicin, and vinblastine (Fig. 3, A–C)
. The percentage of relative cloning efficiency was calculated and plotted against the concentration of colchicine used on a log scale (Fig. 3A)
. After the cells were treated for 4 days under different conditions, tetracycline was either present or absent (-t) from the medium for the 8–10 days that the colonies were established. These series of figures show that the resistance of the cell lines is controlled by the presence or absence of tetracycline. When P-gp is turned off (HeLa P-gpoff) before the cloning assay, it stays off in the presence of tetracycline, and the level of resistance is similar to that of the parental HeLa Tet-Off cells in all three of the drug conditions used (Fig. 3, A–C)
. This demonstrates that the period of colchicine selection of this cell line did not affect the expression of endogenous MDR1. It is interesting to note that in all three of the treatments, when HeLa P-gpoff are taken out of tetracycline for the remainder of the cloning assay (8–10 days) resistance starts to come back, indicating that P-gp expression is returning and that this system is very well regulated by tetracycline treatment (HeLa P-gpoff-t in Fig. 3, A–C
).
|
Membrane Potential of Drug Selected and Tet-repressible Cells.
A representative set of results of relative membrane potentials of P-gp nonexpressing cells (KB-3–1, KB-A1rev, 77.1 P-gpoff, and HeLa P-gpoff) and P-gp expressing cells (KB-V1, KB-A1, HeLa P-gpon, and 77.1 P-gpon) using the fluorescent membrane potential probe, oxonol, are shown in Fig. 4
. Similar results were obtained with cells grown in separate cultures (n = 3–5). There is a reproducible difference in membrane potentials between drug-selected, P-gp expressing cells, KB-V1 and KB-A1 (Fig. 4
, Lanes 2 and 4), and the P-gp nonexpressing cells, KB-3–1 and KB-A1rev (Fig. 4
, Lanes 1 and 3). Lower fluorescence of the drug-selected cells indicates a relative hyperpolarization as compared with their parental cells and revertant control cells. Contrary to the results with the drug-selected cells, 77.1 P-gpon and HeLa P-gpon cells (Fig. 4
, Lanes 6 and 8) have similar fluorescence intensity to 77.1 P-gpoff and HeLa P-gpoff (Fig. 4
, Lanes 5 and 7), suggesting that the membrane potential is similar in both conditions regardless of P-gp presence or absence.
|
, Lanes 5–8. Also, ATPase stimulation studies with partially purified membrane-containing P-gp indicated little or no stimulation with 500 nM oxonol compared with the 150 nM oxonol used to measure membrane potential (data not shown).
Motional Freedom of ESR Probes in the Plasma Membranes of P-gp Expressing and Nonexpressing Drug-selected and -transfected Cells.
The motional freedom of the ESR probe, 5-doxyl-SA, expressed as the order parameter, S, indicates the fluidity of the plasma membranes of the cells. Fig. 5
shows the calculated S values of the studied cells. Drug-selected KB-V1 (Fig. 5
, Lane 4) cells have lower S values indicating more fluid plasma membranes. The KB-A1 drug-selected cells (Fig. 5
, Lane 6) also have more fluid plasma membranes than the KB-A1revertant (Fig. 5
, Lane 5). SDs indicate statistical significance (P < 0.05; n = 3–5). Contrary to these results, the difference in the S values between the HeLa P-gpon (Fig. 5
, Lane 8) compared with HeLa P-gpoff (Fig. 5
, Lane 7), and between 77.1 P-gpon (Fig. 5
, Lane 10) and 77.1 P-gpoff (Fig. 5
, Lane 9) showed small differences that failed to reach statistical significance (P = 0.06 and 0.075, respectively).
|
Motional Freedom of Fluorescent Molecules as Determined by Polarization Experiments.
The TMA-DPH fluorescent probe was used as an alternative way to measure the fluidity of plasma membranes, because this probe is known to probe only at the surface of the plasma membrane and does not translocate deeper into membranes (24)
. Fig. 6A
shows the results of fluorescence polarization obtained with drug-selected and tet-repressible MDR1 gene-transfected cells. Whereas there is a significant difference between parental, KB-3-1 and KB-A1rev (Fig. 6A
, Lanes 1 and 3) and the drug selected, P-gp-expressing KB-V1 and KB-A1 cells (Fig. 6A
, Lanes 2 and 4), no significant difference can be detected between the tet-repressible HeLa P-gpon and 77.1 P-gpon (Fig. 6A
, Lanes 6 and 8) compared with HeLa P-gpoff and 77.1 P-gpoff cells (Fig. 6A
, Lanes 5 and 7). These results, with a cell surface probe, suggest decreased fluidity of drug-selected cells, in contrast with previous results with ESR probes, which indicated increased fluidity.
|
. This probe was used to clarify apparent differences between the ESR results (Fig. 5)
and the polarization results obtained with the TMA-DPH probe (Fig. 6A)
. Polarization numbers are lower for drug-selected P-gp-expressing KB-C1 and KB-V1 cells (Fig. 6B
, Lanes 3 and 4) than for P-gp nonexpressing KB-3-1 and KB-A1rev cells (Fig. 6B
, Lanes 1 and 5). These results parallel those obtained with both the 5-doxyl-SA (Fig. 5)
and T-SASL ESR probes. However, these results are different from the results obtained with the TMA-DPH polarization probe, which probes only the membrane surface (Fig. 6A)
. It appears, then, that motional freedom of both the polarization and ESR probes indicate similar increase in membrane fluidity inside the plasma membranes of drug-selected cells.
Membrane Lipid Packing of P-gp Expressing and Nonexpressing Cells.
Lipid packing of parental and drug-selected cells expressing different amounts of P-gp was studied to determine whether the different cell lines expressing different amounts of P-gp have different biophysical status of their plasma membranes as shown by ESR and polarization, and also insert different amounts of the lipid packing probe, merocyanin 540. Fig. 7
shows typical results of means of fluorescent histograms collected from 104 cells from each cell line (n = 3–4). Results indicate that the drug-selected P-gp-expressing cells (Fig. 7
, Lanes 2–4) incorporate less merocyanine 540 than the P-gp nonexpressing KB-3-1 cells (Fig. 7
, Lane 1), indicating that non-P-gp-expressing cells have loosely packed lipids compared with P-gp-expressing cells.
|
|
DISCUSSION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
To compare P-gp expressing and nonexpressing cells in the same genetic background, we used a tet-repressible plasmid system and achieved high-level expression of P-gp in the absence of tetracycline, which could be repressed after 4 days of tetracycline treatment. The data clearly demonstrate that the P-gp expression and MDR phenotype demonstrated in the tet-repressible cells was because of transcription from the transfected plasmid and that endogenous MDR1 mRNA levels were not affected by the tetracycline treatment or colchicine selection. It is also very clear that the total MDR1 mRNA decreases exponentially with time in tetracycline treatment. MDR1 mRNA began to decline logarithmically within a few hours of tetracycline treatment with a half-life of 3.8 h. Previous studies have reported RNA half-life for MDR1 mRNA to range between 1 and 8 h (30 , 31) , but in this case we cannot compare our data with the half-life of the endogenous MDR1 mRNA, because the MDR1 mRNA encoded by the tet-repressible vector system is not identical in its noncoding regions to the endogenous MDR1 mRNA. However, these mRNA half-life data indicate that transcription is turned off completely by tetracycline treatment, and that loss of P-gp from the cell surface and its degradation becomes the limiting determinant of P-gp function. On the basis of our analysis of cell-surface P-gp, we found a surface half-life of 15.8 h. These data concur with the estimate of 17.5 h in CHRC5 cells and 18.2 h in CEMVLBO.1 cells, which are both drug-selected cell lines (32) . Others report a P-gp protein half-life that range between 24 h and 72 h on different cell lines and under different conditions (31 , 33 , 34) . The longer half-lives in some drug-selected cells suggest the possibility that selection favors host cell alterations that extend the half-life of P-gp. There is another study that estimated the P-gp half-life to be 3 h using a pulse chase method in a vaccinia virus expression system that may be a result of the transient transfection system (35) .
Fig. 2, A and B
clearly demonstrate that as P-gp expression decreases, rhodamine 123 accumulation increases. To avoid long-term drug selection effects of colchicine, which was used as a selection to introduce the tet-repressible plasmid into cells, we kept track of cell passages and only used cells grown up to 28 passages in colchicine. Because HeLa P-gpoff cells are as sensitive to colchicine as the parental HeLa Tet-Off cells (Fig. 3A)
, the effects of the initial selection on expression must be minimal.
Membrane fluidity was studied by determining the motional freedom of ESR probes, 5-doxyl SA (Fig. 5)
and T-SASL, in a series of cells expressing increasing amounts of P-gp on their cell surface and compared with results obtained with the Tet-repressible cells. We have shown that whereas the membrane is increasingly more fluid as drug-selected cells become more resistant (and express more P-gp), the fluidity changes between HeLa P-gpon and P-gpoff, and mouse 77.1 P-gpon and P-gpoff cells are not statistically significant. Interestingly, ESR studies also using the 5-doxyl-SA probe were used by Siegfried et al. (36)
to show correlation between doxorubicin sensitivity and plasma membrane fluidity.
Results similar to those from the ESR probe were obtained when fluorescence polarization of membrane-inserted molecules were studied using 6-(9-anthroyloxy) SA (Fig. 6B)
. Again, cells of increasing drug resistance gave increasingly lower polarization numbers, indicating more fluid plasma membranes, in agreement with the ESR results. In contrast, the polarization probe TMA-DPH yielded opposite results when we compared the fluidity of KB-3-1 and KB-V1 cells (Fig. 6A)
. However, whereas 6-(9-anthryloxy) SA probes are six carbons deep in the plasma membrane, TMA-DPH probes only at the surface of the plasma membrane (24)
. We speculate that TMA-DPH may bind cell surface proteins that are expressed at higher levels in the drug-selected cells, and this binding restricts its motion, yielding higher polarization numbers, interpreted as higher rigidity of plasma membranes. No significant difference was found between polarization of TMA-DPH in the HeLa and 77.1 P-gpon and P-gpoff cells with this probe (Fig. 6A)
. This result indicates that TMA-DPH is not a substrate of P-gp and that P-gp may not be the putative surface protein or proteins to which TMA-DPH molecule binds. Lipid packing of merocyanin 540 also indicates differences among drug-selected MDR cells. This result is consistent with the established differences in the plasma membrane found by the polarization and the ESR experiments discussed above.
Membrane potential measurements also indicate that expression of P-gp alone does not alter the membrane potential of the plasma membranes, and perhaps other changes in membrane composition are responsible for these differences. Fig. 4
shows that whereas there is a substantial difference between drug-selected P-gp-expressing cells, KB-V1 and KB-A1, and nonexpressing cells, KB-3-1 and KB-A1rev, in their uptake of the membrane potential sensing dye oxonol, no such difference exists between the pairs of tet-repressible cells. The drug-selected, P-gp-expressing cells, KB-V1 and KB-A1 cells, have higher membrane potential, indicating that the cytoplasm is more negatively charged. This observation has been associated with the proton extrusion ability of P-gp expressing cells, as shown by the experiments of Landwojtowicz et al. (37)
. However, since there is no significant membrane potential difference in the human and mouse P-gp tet-repressible cells, one can argue that P-gp is not the proton pump responsible for the membrane potential differences observed in drug-selected cells (Fig. 4)
.
What changes in drug-selected cells might be responsible for the changes in membrane fluidity and membrane potential? Most likely, there are other cellular alterations that help confer drug resistance in the presence of P-gp. These could be changes in other membrane proteins, membrane-anchoring proteins (19) , or in the composition of membrane lipids themselves (19) . However, because the cell lines KB-C1 and KB-V1 can have their resistance completely reversed by inhibition of P-gp, it seems likely that these putative membrane changes that affect fluidity and potential do not confer resistance by themselves, but work to enhance the effect of P-gp. For example, increased membrane fluidity would enhance delivery of xenobiotics to P-gp for excretion; increased membrane potential would help keep negatively charged organic cations out of cells after they are pumped. It is possible that alterations in the plasma membrane biophysical properties may help the functionality of P-gp. Callaghan et al. (38) showed that incorporation of fatty lipids into the plasma membrane, which made the membranes less fluid, suppressed the function of the expressed P-gp molecules. Finally, it is also possible that sustained expression of P-gp at the cell surface has a deleterious effect on cell growth, and the changes in biophysical properties of the membrane may help ameliorate these negative effects of P-gp. The tet-repressible expression system would not uncover such evolutionary adaptations seen in cells subject to drug selection over long periods of time.
|
FOOTNOTES |
|---|
2 To whom requests for reprints should be addressed, at Laboratory of Cell Biology, National Cancer Institute, NIH, Building 37, Room 1A09, 37 Convent Drive, MSC 4254, Bethesda, MD 20892-4254. 
3 The abbreviations used are: P-gp, P-glycoprotein; MDR, multidrug resistance; ESR, electron spin resonance; DPH, 1,6-diphenyl-1,3,5-hexatriene; TMA, trimethyl-ammonium; FBS, fetal bovine serum; FACS, fluorescence-activated cell sorter; GAPDH, glyceraldehyde-3-phosphate dehydrogenase; UTR, untranslated region; RT, reverse transcription-PCR; Cp, crossing point; AOSA, 6-(9-anthroyloxy)stearic acid; T-SASL, 2,2,6,6-tetramethyl piperidin-1-oxyl-4-yl-octadecenoate; 5-doxyl-SA, 5-doxyl stearic acid; mAb, monoclonal antibody.
Received 10/22/02. Accepted 4/15/03.
|
REFERENCES |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
interferon: biologic correlation with antiproliferative signal. Biochem. Biophys. Res. Commun., 146: 300-306, 1987.[Medline]
This article has been cited by other articles:
|
|
 |
|
  I. Bottova, A. B. Hehl, S. Stefanic, G. Fabrias, J. Casas, E. Schraner, J. Pieters, and S. Sonda Host Cell P-glycoprotein Is Essential for Cholesterol Uptake and Replication of Toxoplasma gondii J. Biol. Chem., June 26, 2009; 284(26): 17438 - 17448. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. S. Carey, M. Gleason-Guzman, V. Gokhale, and L. H. Hurley Psorospermin structural requirements for P-glycoprotein resistance reversal Mol. Cancer Ther., November 1, 2008; 7(11): 3617 - 3623. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 W. Le Goff, M. Settle, D. J. Greene, R. E. Morton, and J. D. Smith Reevaluation of the role of the multidrug-resistant P-glycoprotein in cellular cholesterol homeostasis J. Lipid Res., January 1, 2006; 47(1): 51 - 58. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. Y. Frank, A. Margaryan, Y. Huang, T. Schatton, A. M. Waaga-Gasser, M. Gasser, M. H. Sayegh, W. Sadee, and M. H. Frank ABCB5-Mediated Doxorubicin Transport and Chemoresistance in Human Malignant Melanoma Cancer Res., May 15, 2005; 65(10): 4320 - 4333. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
N. Y. Frank, S. S. Pendse, P. H. Lapchak, A. Margaryan, D. Shlain, C. Doeing, M. H. Sayegh, and M. H. Frank Regulation of Progenitor Cell Fusion by ABCB5 P-glycoprotein, a Novel Human ATP-binding Cassette Transporter J. Biol. Chem., November 21, 2003; 278(47): 47156 - 47165. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Cancer Research | Clinical Cancer Research |
| Cancer Epidemiology Biomarkers & Prevention | Molecular Cancer Therapeutics |
| Molecular Cancer Research | Cancer Prevention Research |
| Cancer Prevention Journals Portal | Cancer Reviews Online |
| Annual Meeting Education Book | Meeting Abstracts Online |
) and HeLa MDR-Off cells with different treatments were assayed for resistance to increasing amounts of (A) colchicine, (B) doxorubicin, and (C) vinblastine. HeLa P-gpcol (
) were cultured in colchicine, HeLa P-gpoff (
) were cultured in 2 µg/ml tetracycline for the 8–10-day cloning assay, and HeLa P-gpon (
) were cultured in the absence of all drugs 4 days before the cloning assay. HeLa P-gpoff-t (
) cells were cultured in 2 µg/ml tetracycline before the cloning assay; for the duration of the cloning assay, tetracycline was removed; bars, ±SD.
