Conceived and designed the experiments: IFA AR AAM YHK. Performed the experiments: IFA AR. Analyzed the data: IFA AR AAM YHK. Contributed reagents/materials/analysis tools: YHK. Wrote the paper: IFA AAM YHK.
Current address: Department of Physiology and Pharmacology, University of Western Ontario, London, Ontario, Canada
The authors have declared that no competing interests exist.
Gliomas are morbid brain tumors that are extremely resistant to available chemotherapy and radiology treatments. Some studies have suggested that calcium-activated potassium channels contribute to the high proliferative potential of tumor cells, including gliomas. However, other publications demonstrated no role for these channels or even assigned them antitumorogenic properties. In this work we characterized the expression and functional contribution to proliferation of Ca2+-activated K+ channels in human glioblastoma cells. Quantitative RT-PCR detected transcripts for the big conductance (BK), intermediate conductance (IK1), and small conductance (SK2) K+ channels in two glioblastoma-derived cell lines and a surgical sample of glioblastoma multiforme. Functional expression of BK and IK1 in U251 and U87 glioma cell lines and primary glioma cultures was verified using whole-cell electrophysiological recordings. Inhibitors of BK (paxilline and penitrem A) and IK1 channels (clotrimazole and TRAM-34) reduced U251 and U87 proliferation in an additive fashion, while the selective blocker of SK channels UCL1848 had no effect. However, the antiproliferative properties of BK and IK1 inhibitors were seen at concentrations that were higher than those necessary to inhibit channel activity. To verify specificity of pharmacological agents, we downregulated BK and IK1 channels in U251 cells using gene-specific siRNAs. Although siRNA knockdowns caused strong reductions in the BK and IK1 current densities, neither single nor double gene silencing significantly affected rates of proliferation. Taken together, these results suggest that Ca2+-activated K+ channels do not play a critical role in proliferation of glioma cells and that the effects of pharmacological inhibitors occur through their off-target actions.
Gliomas are primary brain tumors that arise from glial cells. They represent 30 to 60% of CNS primary tumors, with the incidence of 2 to 3 new cases per 100,000 persons annually
There is extensive literature suggesting that proliferation of normal and transformed animal cells requires activity of certain potassium (K+) channels (see for example
One group of K+ channels that may be relevant to proliferation of both malignant and non-malignant cells are the Ca2+-activated K+ channels (
Prior studies have implicated BK channels in the proliferation and migration of glioblastoma cells
Because of the uncertain functional significance of BK and IK channels in glioblastoma proliferation, in the present study we used a combination of pharmacological and molecular biology tools to explore their functional significance. We found that pharmacological inhibitors of both BK and IK1 strongly suppress glioma cell growth in an additive fashion. However, low concentration of the same blockers that were sufficient to inhibit channel activity had no effect on cell proliferation. To address this discrepancy, we downregulated BK and IK1 channels using gene-specific siRNAs. siRNA transfections caused strong reductions in K+ current densities but no changes in cell growth. These data argue against a critical role for BK and IK1 in GBM proliferation.
Clotrimazole, paxilline, penitrem A, 8,14-Diaza-1,7(1,4)-diquinolinacyclotetradecaphane trifluoroacetate salt (UCL1848), and all other salts and chemicals were from Sigma-Aldrich (St. Louis, MO, USA) unless otherwise noted. 1-[(2-Chlorophenyl)diphenylmethyl]-1H-pyrazole (TRAM-34) was purchased from Tocris Biosciences (Ellisville, MO, USA). Stock solutions of paxilline (20 mM), penitrem A (20 mM), clotrimazole (30 mM), and TRAM-34 (30 mM) were prepared in DMSO and stored at −30°C. UCL1848 was diluted in water at 20 mM.
Two established astrocytoma/glioblastoma cell lines, U251 MG and U87 MG, were used in this study. U251 cells were a gift of Dr. M.G. Kaplitt (Weill-Cornell Medical Center, New York, USA); their exact passage is unknown. U87 cells were obtained from the American Type Culture Collection (ATCC, Manassas, VA, USA) and used at passages 78–90. Both U251 and U87 cells were grown in Dulbecco's Modified Eagle's Medium (DMEM) supplemented with 10% fetal bovine serum (FBS), 50 U/ml penicillin, and 50 mg/ml streptomycin at 37°C in a humidified atmosphere containing 95% air and 5% CO2. All cell culture reagents are from Invitrogen-Gibco (Carlsbad, CA, USA). Culture media was changed twice a week and cells were passaged using recombinant protease TrypLE when 90–95% confluency was reached.
Primary glioblastoma cells were prepared from a surgical sample of glioblastoma multiforme. Tissue sample was obtained with written consent under a protocol approved by Albany Medical Center Institutional Review Board. Tumor tissue (∼200–300 mg) was washed twice with ice-cold Ca2+-, Mg2+-free phosphate-buffered saline (PBS, pH 7.4), minced to small pieces and treated with solution of 0.125% Trypsin/0.015% EDTA in PBS containing 250 µg/ml DNAse I. After brief digestion, tissue fragments were triturated using fire-polished glass Pasteur pipette, and the resulting cell suspension was filtered through a Nylon cell strainer (70 µm, BD Falcon, Bedford, MA, USA). Cell were grown in T75 cell culture flasks in DMEM plus 20% FBS supplemented with 100 U/ml penicillin, and 100 mg/ml streptomycin at 37°C in a humidified atmosphere containing 95% air and 5% CO2.
Expression of various Ca2+-activated K+ channels in cultured cells and the surgical GBM tissue sample was analyzed using quantitative RT-PCR. Cellular or tissue mRNA was isolated using the RNAqueous-4PCR kit (Applied Biosystems-Ambion, Austin, Texas, USA) according to the manufacturer's instructions. Cultured cells were grown in a 60-mm dish to 60–80% confluency and lyzed in 700 µl of lysis buffer. The GBM tumor sample (∼100–150 mg) was homogenized in 1 ml lysis buffer provided with the RNAqueous-4PCR kit. Concentration of mRNA in resulting samples was quantified using NanoDrop 1000 (ThermoFisher Scientific, Wilmington, DE, USA). mRNA was converted to cDNA using the iScript cDNA Synthesis kit (Bio-Rad Laboratories, Hercules, CA, USA) according to the manufacturer's instructions. One µg of mRNA was used in each 20-µl cDNA synthesis reaction mix.
Gene expression was quantified by a real-time PCR using the CFX96 Real-Time PCR Detection System (Bio-Rad) and iTaq SYBR Green Supermix kit (Bio-Rad) according to the manufacturer's instructions. One µl of cDNA product prepared as described above was used for each qPCR reaction. Gene expression was analyzed with the gene-specific QuantiTect Primer Assays (Qiagen, Hilden, Germany; see
Gene | Sequence | Manufacturer | Catalog number |
Proprietary | Qiagen | QT00089915 | |
Proprietary | Qiagen | QT01192646 | |
Proprietary | Qiagen | QT00024157 | |
Proprietary | Qiagen | QT00025375 | |
Proprietary | Qiagen | QT00016611 | |
Proprietary | Qiagen | QT00070966 | |
Proprietary | Qiagen | QT00003780 |
Protein expression of the IK1 channel was assessed by the Western blot analysis using polyclonal antibody raised against synthetic peptide corresponding to a region of human IK1 (LYDLQQNLSSSHRALEKQIDTLAGKLDALTELLSTALGPRQLPEPSQQSK, Sigma-Aldrich; cat.# AV35098). Whole cell lysates were diluted with a reducing Laemmli buffer. Proteins were separated on 10% polyacrylamide gel followed by transfer onto an Immun-Blot PDVF membrane (Bio-Rad). The membrane was blocked for 1 hr with 5% nonfat milk in Tris-Phosphate buffer containing 0.05% Tween 20 (TBS-T). It was further incubated overnight at 4°C with primary antibody (1∶500 dilution). After five washes for 5 min with TBS-T, membranes were further incubated with horseradish peroxidase-conjugated secondary antibodies (GE Healthcare/Amersham Biosciences, Piscataway, NJ; 1∶10,000 dilution), followed by four TBS-T washes. The horseradish peroxidase signal was detected and digitized using a chemiluminescence ECLplus reagent (GE Healthcare/Amersham Biosciences) and a luminescent image analyzer LAS-4000 (FujiFilm Medical Systems, Stamford, CT).
BK and IK1 currents were recorded employing conventional whole-cell patch clamp technique
BK currents were recorded from a holding potential of −80 mV in response to step pulses of −80 mV to +140 mV in 20 mV increments. BK current development and inhibition was monitored by applying repetitive (every 3 sec) depolarization ramps from −120 mV to +140 mV. Because the pipette solution is nominally Ca2+-free, other Ca2+ activated K+ channels, such as IK1 and SK1-3 are silent under these experimental conditions.
IK1 currents were recorded from a holding potential of −80 mV in response to step pulses from −120 mV to +80 mV in 20 mV increments. Development and inhibition of the Ca2+-dependent IK1 currents were monitored by applying repetitive (every 3 sec) depolarization ramps from −120 mV to +80 mV. Since these conditions also favor activation of BK currents, 2 µM paxilline was added to bath solution to isolate the IK1 currents.
Cell proliferation rates were quantified using two approaches: Coulter counter technique and colorimetric MTT proliferation assay. U251 and U87 cells were plated in 24-well cell culture plates (TPP, Trasadingen, Switzerland, Europe) at the density of 10,000 cells per well and were allowed to grow overnight. The next day, culture media was replaced with serum-containing or serum-free media including the various pharmacological inhibitors or vehicle controls. Initial cell density was quantified at this time in a separate “baseline” plate prepared in the same fashion. We used two types of culture media in pharmacological experiments: serum-containing (DMEM plus 10% FBS), or serum-free medium prepared using OptiMEM with B27 supplement (Invitrogen; 1∶50 dilution). To preclude the potential degradation of inhibitors, we replaced inhibitor-containing media with freshly prepared every 24 hrs. 48 hrs after first application of inhibitors, extracellular media were removed and cells were processed for proliferation assays.
For MTT assays, cells were briefly washed with basal solution containing (in mM): 135 NaCl, 3.8 KCl, 1.2 MgSO4, 1.3 CaCl2, 1.2 KH2PO4, 10 HEPES, and 10 glucose (pH = 7.4, adjusted with NaOH). Aliquots of basal solution containing 0.5 mg/ml thiazolyl blue tetrazolium bromide (MTT, Sigma-Aldrich) were then added to each well. After 30-min incubation at 37°C MTT solution was removed, and cells were solubilized in acidified isopropanol to dissolve newly formed MTT-formazan particles. 250-µl aliquots from each well were transferred into a 96-well plate. Absorbances at 562 nm were measured using ELx800 Absorbance microplate reader (BioTek Instruments, Winooski, VT, USA).
For Coulter counter assays, cells were detached from substrate using recombinant protease TrypLE and counted using a Z1 Series Coulter Counter (Beckman Coulter, Miami, FL, USA).
To downregulate BK and IK1, U251 cells were transfected with gene-specific siRNAs using a Nucleofector II (Lonza, Cologne, Germany) and the Amaxa Nucleofection kit T according to the manufacturer's protocol. Cells were removed from the substrate with TrypLE, counted, centrifuged at 300 g for 10 min, and mixed with nucleofection suspension buffer containing 0.5 µg of GFP cDNA (as a control of transfection efficacy) and 1 µM siRNA of choice. Cells were electroporated using program T-20 and then plated on either 24-well plates for proliferation assays or glass coverslips for patch clamp experiments. We used two to three siRNA constructs per gene that were obtained from Qiagen or Ambion. Target sequences for each siRNA are listed in
Target gene | siRNA # | Sense strand | Antisense strand | Manuf. |
Hs_KCNMA1 | BK#1 | Ambion | ||
Hs_KCNMA1 | BK#2 | Ambion | ||
Hs_KCNN4 | IK1#5 | Qiagen | ||
Hs_KCNN4 | IK1#6 | Qiagen | ||
Hs_KCNN4 | IK1#7 | Qiagen | ||
Hs_KCNK2 | TREK1#1 | Ambion | ||
N/A | NC | Proprietary nonsense siRNA (AllStars Negative Control) | Qiagen |
N/A, not applicable; NC, negative control; Manuf., manufacturer.
All data presented as mean values ±SE. Number of independent experiments per group is indicated in each figure legend. Statistical significance was determined by one-way analysis of variance (ANOVA) and
We initially evaluated the expression of Ca2+-activated K+ channels in glioma cell lines at the mRNA level. U251 and U87 are both derived from human gliomas, but demonstrate different morphological and growth characteristics as well as substantial difference in gene expression profiles
Relative levels of mRNA expression for BK, IK1 and SK1-3 channels were quantified in U251, U87, and a GBM tissue sample using reverse transcription quantitative PCR. Expression levels were normalized to the expression of two housekeeping genes, GAPDH (shown in graph) and ribosomal protein RPL13A. Mean values ±SE of gene expression in three independent cell preparations for U251 and U81, and three independent measurements for GBM are shown.
Expression of mRNA does not necessarily result in production of functional plasmalemmal channels. For example, Weaver
When U251 cells were dialyzed with a pipette solution containing no Ca2+, we registered large macroscopic K+ currents that were activated in response to depolarizing step pulses >40 mV (
(
To record IK1 currents, we increased [Ca2+] in the pipette to 750 nM and applied step pulses from –120 mV to +80 mV in 20 mV increments. To prevent concomitant activation of the BK currents, 2 µM paxilline was included in the bath solution. Under these recording conditions, we observed macroscopic K+ currents that were saturated at voltages ≥40 mV, consistent with the electrophysiological profile of IK1 channels (
(
Since immortalized cell lines frequently have different gene expression profile from cells
(
Our finding of the functional IK1 currents in all glial cell lines was in conflict with the previous observations of Weaver et al.
To explore functional involvement of BK and IK1 channels in proliferation, we tested the effects of pharmacological inhibitors of BK and IK1 using two different techniques, Coulter counter and MTT proliferation assays. Using two alternative approaches was important because MTT conversion into light-absorbing formazan is mediated by mitochondrial dehydrogenases and may be affected without changes in cell numbers or viability
Paxilline (10 µM) and clotrimazole (10 µM) were added to culture media alone or in combination, and rates of cell proliferation were determined 48 hrs later using an MTT proliferation assay. Proliferation assays were performed in standard cell culture medium (DMEM +10% FBS), serum-free OptiMEM medium supplemented with serum substitute B27, or in OptiMEM+B27 additionally containing bovine serum albumin (2 mg/ml). Quantitatively similar data were obtained using Coulter counter assay (see
Cells were grown in the serum-free OptiMEM media supplemented with serum substitute B27. The BK blocker paxilline (10 µM) and the IK1 inhibitor clotrimazole (10 µM) were added as indicated. Images of the cells were captured ∼48 hrs after addition of channel blockers using Hoffman modulation contrast optics in Olympus IX71 microscope at 10×10 magnification.
We next tested the dose dependency of the effects of paxilline and clotrimazole on U251 cell proliferation and compared their efficacy to the specific BK inhibitor penitrem A
(
The disparity between the IC50 values for inhibition of proliferation and channel activity may be explained by decomposition of tested compounds in culture media, or by nonspecific absorption of the drugs during long incubation. To address this possibility, we tested the inhibitory properties of culture media containing nominally 10 µM of either paxilline or clotrimazole after 24-hr incubation in cell proliferation assays. We diluted aliquots of the inhibitor-containing culture media 10-fold with electrophysiological bath solution and performed whole-cell patch recordings. Such dilution was necessary to minimize differences between cell culture media and bath solution. The final nominal concentrations of paxilline and clotrimazole were 1 µM. As seen in
We decided to further test the functional involvement of BK and IK1 in glioma cell proliferation using an alternative approach. For this purpose, we downregulated BK and IK1 expression using an siRNA technique. At 48 hrs after transfection, the two BK-specific siRNAs significantly downregulated BK mRNA expression levels by 62% and 58% (
(
Two out of three IK1-specific siRNAs tested by us downregulated IK1 mRNA levels by 52% and 70% 48 hrs after transfection, while the third construct proved ineffective (
Because paxilline and clotrimazole reduced cell proliferation in an additive fashion, we explored the effects of simultaneous BK and IK1 downregualtion on rate of U251 cell growth. As in the case of individual channel knockdowns, a combination of the BK siRNA#2 and the IK1 siRNA#7 (two constructs that effectively suppressed BK and IK1 current densities) was ineffective in inhibiting cell growth (
The main finding of this study is the lack of functional involvement of the Ca2+-activated K+ channels in proliferation of U251 and U87 glioma cell lines and primary GBM cells. This conclusion is at odds with some previous reports that also used pharmacological agents to assay for functional contribution of BK and/or IK1 to proliferation. Although all the inhibitors of BK and IK1 that we tested in this study suppressed glioma cell proliferation, their effects occurred at concentrations significantly exceeding those needed to block channel function. Downregulation of BK and IK1 activity using siRNA did not cause changes in cell growth. Therefore, the antiproliferative properties of the BK and IK1 channel blockers are likely mediated by off target actions.
BK channels are highly expressed in glioma cell lines and in human glioma tissue
A somewhat surprising finding of our work was the detection of functional IK1 channels in two established glioma lines, U251and U87, as well as in primary GBM cells. Although IK1 mRNA is consistently detected in immortalized glioma cell lines and human GBM samples, Weaver
The detection of the IK1 subtype of Ca2+-activated K+ channels in glioma cells may be of substantial importance. These channels have been implicated in tumorogenesis and regulation of cell proliferation in a number of cell types, including pancreatic and endometrial cancer cells, non-malignant endothelial and smooth muscle cells, and bone marrow mesenchymal cells
Overall, our data provide strong evidence against a requirement for the Ca2+-activated K+ channels BK and IK1 for proliferation of U251 cells, and likely other glioma cells. Further work is needed to define the mechanisms responsible for the antiproliferative effects of pharmacological inhibitors of BK and IK1, such as paxilline, penitrem A, clotrimazole, and TRAM-34. Our results suggest that strong caution should be taken when interpreting positive results obtained with pharmacological ion channel blockers and that such experiments should be supplemented with molecular biology studies targeting expression of proteins of interest.
The BK blocker charybdotoxin (100 nM) potently suppresses whole-cell K+ currents in U251 glioma cells. For methodological details see
(1.34 MB EPS)
U87 glioma cells express typical IK1 currents that were potently suppressed by the IK1 blocker clotrimazole (2 µM). For methodological details see
(1.19 MB EPS)
Western blot analysis of the IK1 protein expression in glioma cell lines and primary glioblastoma cells. Protein lysates (20 µg/lane) of primary glioblastoma (GBM), U87, and U251 cells were separated by sodium dodecyl sulfate polyacrylamide gel electrophoresis, transferred to the PVDF membrane and probed with polyclonal anti-IK1 antibody as described in
(0.35 MB EPS)
Effects of paxilline and clotrimazole on proliferation of U251 cells in serum-containing and serum-free media measured using two alternative cell proliferation assays. (A) Effects of paxilline (10 µM) and clotrimazole (10 µM), or their combination on proliferation of U251 in media containing 10% FBS. Cells were treated with pharmacological inhibitors for 48 hrs and rates of proliferation were quantitatively assessed using Coulter Counter (white bars) and MTT (grey bars) assay. Plating cell density is indicated by dashed line. (B) Effects of paxilline (10 µM) and clotrimazole (10 µM) or their combination on proliferation of U251 cells in serum-free media supplemented with serum substitute B27. Cell proliferation was quantified using Coulter Counter (open bars) or MTT assay (grey bars). (C) Effects of paxilline (10 µM) and clotrimazole (10 µM) or their combination on proliferation of U251 cells in serum-free media supplemented with serum substitute B27 and bovine serum albumin. Proliferation rates were measured using MTT proliferation assay. Data are the mean values ±SE of proliferation normalized to controls in the same experiments. **p<0.01 vs. control; **p<0.05 vs control;***p<0.001 vs. control.
(0.41 MB EPS)
Effects of paxilline and clotrimazole on proliferation of U87 cells in serum-containing and serum-free media. (A) Effects of paxilline (10 µM) and clotrimazole (10 µM) or their combination on proliferation of U87 in media containing 10% FBS. Cells were treated with pharmacological inhibitors for 48 hrs and rates of proliferation were quantitatively assessed using MTT assay. Data are the mean values of proliferation ±SE normalized to controls in the same experiments. (B) Effects of paxilline (10 µM) and clotrimazole (10 µM) or their combination on proliferation of U87 cells in serum-free media supplemented with serum substitute B27. Data are the mean values of proliferation ±SE normalized to controls in the same experiments. **p<0.01 vs. control; ***p<0.001 vs. control.
(0.39 MB EPS)
Representative images of U251 cells transfected with the GFP-expressing plasmid using amaxa nucleofection protocol. Images of the same field were acquired using Olympus IX71 Hoffman modulation contrast optics (A) and GFP fluorescence (B) 72 hrs after nucleofection at 10×10 magnification. Transfection efficacy varied between 60–90% in various cell preparations.
(1.24 MB EPS)
Double knockdown of BK and IK1 K+ channels does not affect proliferation of U251 glioma cells. Effect of transfection with negative control (NC), BK-specific siRNA #2 (siBK), IK1-specific siRNA#7 (siIK1) or combination of BK- and IK1-specific siRNAs (siBK+siIK1) on proliferation of U251 measured 72 hrs post transfection using MTT proliferation assay. Data are the mean values ±SE of 3 independent cell transfections.
(0.32 MB EPS)
We thank Dr. Michael G. Kaplitt for the gift of U251 cells, Drs. James A. Bennet and Thomas T. Andersen for providing U87 cells, Dr. Fassil B. Mesfin for help in preparation of GBM cells, and Dr. Mohamed Trebak for critical reading of the manuscript. We greatly appreciate the technical support of Maria C. Hyzinski.