Indian Journal of Biochemistry & Biophysics Vol. 49, October 2012, pp. 329-341

Choice of cyclodextrin for cellular cholesterol depletion for vascular endothelial cell lipid raft studies: Cell membrane alterations,

cytoskeletal reorganization and cytotoxicity Adam H Hinzey†, Michelle A Kline†, Sainath R Kotha†, Sean M Sliman, Elizabeth S O’Connor Butler,

Andrew B Shelton, Travis R Gurney and Narasimham L Parinandi* Lipid Signaling, Lipidomics, and Vasculotoxicity Laboratory, Division of Pulmonary,

Allergy, Critical Care, and Sleep Medicine, Dorothy M. Davis Heart and Lung Research Institute, The Ohio State University College of Medicine, Columbus, OH, 43210, USA

Received 06 August 2012; revised 05 September 2012

The use of cyclodextrins as tools to establish the role of cholesterol rafts in cellular functions has become a widely accepted procedure. However, the adverse effects of cyclodextrins as the cholesterol-depleting agents on cellular structure and functions are not reported in detail. Therefore, in the current study, we investigated the membrane-perturbing actions and cytotoxicity of the two widely used cellular cholesterol-depleting cyclodextrins methyl-β-cyclodextrin (MβCD) and hydroxypropyl-β-cyclodextrin (HPCD) in our well-established bovine pulmonary artery endothelial cell (BPAEC) in vitro model system. BPAECs treated with different concentrations of MβCD and HPCD (2% and 5%, wt/vol.) for 15-180 min showed significant loss of membrane cholesterol, cytotoxicity, cell morphology alterations, actin cytoskeletal reorganization, alterations in cellular proteins and membrane fatty acid composition, and decrease in trans-endothelial electrical resistance (TER). MβCD induced a marked loss of cellular proteins, as compared to that caused by HPCD under identical conditions. More noticeably, MβCD caused a drastic loss of membrane lipid fatty acids in BPAECs, as compared to HPCD which failed to cause such alteration. Removal of cholesterol by cyclodextrin (especially MβCD) treatment apparently caused loss of fluidity of the cell membrane and leakage of vital cellular molecules including proteins and fatty acids, and thus caused cytotoxicity and loss of cell morphology in BPAECs. Replenishment of cells with cholesterol following its depletion by MβCD treatment significantly attenuated the depletion of cellular cholesterol, cytotoxicity and morphological alterations in BPAECs, indicating the importance of membrane cholesterol in vascular EC integrity. Also, the current study offered a safer method of cholesterol removal from membranes and lipid rafts by HPCD, suggesting its use in studies to investigate the role of lipid raft-associated cholesterol in cellular functions.

Keywords: Membrane cholesterol, Lipid rafts, Cyclodextrins, Vascular endothelial cells, Cytotoxicity, Cytoskeletal reorganization

Cholesterol stands out as an important component of cell membrane that essentially contributes to the structure and function of mammalian cells. Cellular membranes, including the plasma membrane comprise

well-organized lipid raft microdomains which act as the important mediators of diverse cellular signal transduction events1. However, lipid rafts are predominantly made up of cholesterol and sphingolipids and contain more of these lipidic species than the adjacent cell membrane from which they are derived. The lipid rafts harbor the proteins such as glycolipoproteins, crucial cell signaling enzymes and receptors. Originating from plasma membrane, lipid rafts traverse freely in phospholipid bilayer of cell membrane. Lipid raft-associated cholesterol is a pivotal player among membrane lipids in modulating cellular signaling cascades, which regulate normal cellular functions and pathophysiological events2. Although the distribution of cholesterol in various cellular membranes has been found to be heterogeneous, it has been estimated that 80-90% of the entire cellular cholesterol is localized in the plasma membrane3.

———————— *Author for correspondence: Phone: (614) 486-1974 Fax: (614) 293-4799 E-mail: narasimham.parinandi@osumc.edu †Equally contributed Abbreviations: ANOVA, analysis of variance; BPAEC, bovine pulmonary artery endothelial cell; DTT , dithiothreitol; EC, endothelial cell; ECGF, endothelial cell growth factor; ECIS, electrical cell impedance system; FBS, fetal bovine serum; GC-MS, gas chromatography-mass spectrometry; HPCD, hydroxypropyl-β-cyclodextrin; LDH, lactate dehydrogenase; MEM, minimal essential medium; MTT, 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide; MβCD, methyl-β-cyclodextrin; PAGE, polyacrylamide gel electrophoresis; PBS, phosphate buffered saline; SD, standard deviation; SDS, sodium dodecyl sulfate; TB , Tris-buffered saline; TCA, trichloroacetic acid; TER, trans-endothelial electrical resistance.

INDIAN J. BIOCHEM. BIOPHYS., VOL. 49, OCTOBER 2012

330

In order to establish the role of raft-associated cholesterol in several important cell signaling events, cyclodextrins have been used as tools for removal of cholesterol from lipid rafts of cells4. The vascular endothelium consisting of a monolayer of endothelial cells (ECs) plays a crucial role in maintaining the structure and function of blood vessel and also in regulating homeostasis of cardiovascular system. Cholesterol of vascular ECs is recognized as an important element in vascular EC signaling that dictates cellular events, including replication and cell death3,5,6. The β-cyclodextrin family of compounds have been widely used to deplete cholesterol from the plasma membrane and lipid rafts in order to investigate the role of cholesterol and cholesterol rafts in certain signaling events of vascular ECs7-9.

Although depletion of cholesterol from plasma membrane and lipid rafts of vascular ECs is successfully accomplished by the use of β-cyclodextrins (μM or μg doses), the adverse effects of cholesterol-depleting agents have not been thoroughly examined. Thus, in the present study, we have investigated the adverse effects of widely used cellular cholesterol-depleting compounds methyl-β-cyclodextrin (MβCD) and hydroxypropyl-β-cyclodextrin (HPCD) (Fig. 1) in our well-established cultured bovine pulmonary artery EC (BPAEC) model in vitro. BPAECs were treated for 15-180 min with 2% and 5% (wt/vol.) of the selected cyclodextrins MβCD and HPCD.

The results revealed that both MβCD and HPCD were effective in removing cholesterol from cells, whereas MβCD caused significantly greater extent of depletion of cholesterol, loss of cell viability, morphology alterations, loss of membrane fatty acids, alterations in proteins, barrier dysfunction and cytoskeletal reorganization, as compared to the extent of the same caused by HPCD under identical conditions in BPAECs. Also, the current study offered a safer method of removal of cholesterol by utilizing HPCD without causing extensive adverse effects as seen with the MβCD treatment in ECs for studies to investigate the role of lipid raft-associated cholesterol in cellular functions and signaling cascades. Materials and Methods Materials

Bovine pulmonary artery endothelial cells (BPAECs) (passage 4) were commercially obtained from VEC Technologies (NY, USA). Minimal essential medium (MEM), non-essential amino acids, trypsin, fetal bovine serum (FBS), penicillin/ streptomycin, phosphate-buffered saline (PBS), 12% Tris-glycine gels, SDS loading buffer, dithiothreitol (DTT) reducing agent, SeeBlue® Plus2 protein standard and Amplex red cholesterol assay kit were acquired from the Invitrogen Corporation (Grand Island, NY). Methyl-β-cyclodextrin (MβCD),

Fig. 1—Structures of β-cyclodextrins, MβCD and HPCD used for cellular cholesterol depletion

HINZEY et al: CYCLODEXTRIN VASCULOTOXICITY

331

2-hydroxypropyl-β-cyclodextrin (HPCD), 6.5% formaldehyde solution, t-octylphenoxypolyethoxye thanol (Triton X-100) and lactate dehydrogenase (LDH) assay kit was purchased from the Sigma-Aldrich Inc. (St. Louis, MO). [3H] Thymidine was procured from Perkin Elmer (Shelton, CT). Endothelial cell growth factor (ECGF) was purchased from the Upstate (Millipore) (Charlottesville, VA).

MTT [3-(4,5-dimethylthiazol-2-yl)-2,5-diphenylte trazolium bromide] and glass coverslips (18 × 18 mm) were obtained from the Fisher Scientific (Pittsburgh, PA). Rhodamine phalloidin was procured from the Molecular Probes Invitrogen Co. (Carlsbad, CA). Gold electrodes for measurement of the trans-endothelial electrical resistance (TER) on the electrical cell impedance system (ECIS) were obtained from the Applied Biophysics (Troy, NY). Polyoxyethylene sorbitan monolaurate (Tween-20), Tris-buffered saline (TBS) and Coomassie Brilliant Blue R-250 staining solution were acquired from the Bio-Rad (Hercules, CA). Heptadecanoic acid (C17:0 internal standard) was procured from the Nu-Chek Prep (Elysian, MN). Cell culture

BPAECs were grown to confluence (~95%) in MEM supplemented with 10% fetal bovine serum, 100 units/mL penicillin and streptomycin, 5 µg/mL ECGF and 1% non-essential amino acids at 37°C in a humidified environment of 95% air - 5% CO2 as reported earlier10,11. BPAECs from passages 4-19 were used in the current study. BPAECs initially cultured in the T-75 cm sterile flasks were detached with 0.05% trypsin, re-suspended in the complete EC growth medium and sub-cultured in 35-mm sterile dishes, in 100-mm sterile dishes, on sterile coverslips and on the ECIS gold electrodes to the required level of confluence under a humidified environment of 95% air - 5% CO2 at 37ºC for experiments. Membrane cholesterol depletion and determination

BPAECs cultured in 35-mm sterile dishes (~90% confluence) were treated with basal MEM alone or basal MEM containing the chosen β-cyclodextrin (2% and 5% MβCD or HPCD, wt/vol.) for 15-120 min at 37ºC in a humidified environment of 5% CO2 - 95% air. At the end of incubation, the medium was removed and 1 mL of phosphate buffered saline (PBS) was added to each dish containing the cells. The cells were then detached with a cell scraper and transferred into an Eppendorf tube and centrifuged for

10 min at 15,000 RPM. After centrifugation, the supernatant was removed without disturbing the cell pellet. The cholesterol amount in the cells was then determined spectrofluorometrically, according to the manufacturer’s recommendations (Molecular Probes - Invitrogen Detection Technologies, Grand Island, NY). Lactate dehydrogenase (LDH) assay of cytotoxicity

Release of LDH from the cells into the medium upon exposure to the β-cyclodextrins was determined spectrophotometrically as an index of cytotoxicity according to our earlier reported method12. BPAECs cultured in 35-mm sterile dishes (~90% confluence) were treated with basal MEM alone or basal MEM containing the chosen β-cyclodextrin (2% and 5% MβCD or HPCD, wt/vol.) for 15-120 min at 37 ºC in a humidified environment of 5% CO2 - 95% air. The experiment was terminated with 1 N HCl and LDH released into the medium was measured by the spectrophotometric method, according to the manufacturer’s recommendations (Sigma Chemical, St. Louis, MO). MTT Assay of cytotoxicity

MTT reduction by the mitochondria of ECs, following their treatment with β-cyclodextrins was determined as an index of mitochondrial dysfunction according our earlier reported method13. BPAECs cultured (~90% confluence) in 17.5-mm sterile dishes were treated with MEM alone or MEM containing different concentrations of either MβCD or HPCD for 15-120 min. At the end of treatment, MTT solution (10% of culture volume) was added and incubated for 3 h, following which the medium was removed and MTT solvent was added in an amount equal to original volume of the medium. The extent of MTT reduction by the cells was determined spectrophotometrically according to the manufacturer’s recommendations (Sigma Chemical Co., St. Louis, MO). [3H]Thymidine Incorporation assay for cell proliferation

Proliferation of BPAECs, following the treatment of cells with β-cyclodextrins, was determined by the [3H]thymidine incorporation into the cells according to our earlier reported method13. BPAECs grown on 35-mm sterile dishes (~70% confluence) were treated with MβCD or HPCD for the required lengths of time, following which the medium was removed and 1 ml of [3H]thymidine (1 µCi/ml of complete MEM) was added to each dish containing the cells, and then they

INDIAN J. BIOCHEM. BIOPHYS., VOL. 49, OCTOBER 2012

332

were incubated for 24 h. Following the incubation, the medium containing [3H]thymidine was removed and cells were washed with PBS. Cells were then washed with 5% trichloroacetic acid (TCA) in distilled water, and then 500 µL of 10.25 M NaOH was added to each dish for 30 min. Cells were then detached with a cell scrapper, solubilized, and the [3H] radioactivity was measured in the Packard Tri-carb 2900TR Liquid Scintillation Counter. Microscopic examination of cell morphology

Morphological alterations in BPAECs grown in 35-mm sterile dishes (~90-100% confluence), following their exposure to different concentrations of MβCD and HPCD for 15-120 min, as an index of cytotoxicity were examined under a light microscope according to our earlier reported method11. Images of cell morphology were digitally captured by using the Nikon Eclipse TE2000-S at either 10X or 100X magnification. Trans-endothelial electrical resistance (TER) determination

TER was determined as an index of loss of tight junctions and cell adherence, upon treatment with β-cyclodextrins by utilizing the ECIS equipment, according to our earlier reported method11. BPAECs grown on the sterile ECIS gold electrodes (~80-90% confluence), were treated with basal MEM and then allowed to equilibrate to the 37°C under a humidified environment of 95% air - 5% CO2 for 1 h, followed by treatments. Confocal microscopy of actin cytoskeletal reorganization

Actin stress fiber formation, as an index of the actin cytoskeletal reorganization in BPAECs, following the treatment with β-cyclodextrins was determined by confocal fluorescence microscopy according to our earlier reported method11,13. BPAECs were grown on sterile glass coverslips (~80-90% confluence), treated with MEM alone or MEM containing different concentrations of MβCD or HPCD for the desired length of time. The formation of actin stress fibers was then examined following the rhodamine-phalloidin staining with the aid of Zeiss LSM 510 Confocal/Multiphoton microscope at 543 nm excitation and 565 nm emission under 63X magnification. The images were captured digitally. Membrane fatty acid composition analysis

Fatty acid composition of the membrane lipids of BPAECs was determined according to our previously reported method14,15. BPAECs cultured in 100-mm

sterile dishes (~90% confluence) were treated with basal MEM or MEM containing different concentrations (2 and 5%) of MβCD or HPCD for 1-3 h. Following the treatments, total lipids were extracted from cells by Folch extraction (2:1 chloroform:methanol, vol./vol.). Following hydrolysis and derivatization (alkaline methanolysis) of the fatty acids in cell membrane lipids, the resulting fatty acid methyl esters were analyzed by gas chromatography-mass spectrometry (GC-MS) using the Shimadzu QP2010 GC-MS (Shimadzu Scientific Instruments, Columbia, MD, USA) equipped with Restek column. Fatty acids were normalized to mg of total cell protein utilizing the C17:0 fatty acid methyl ester as the internal standard. Membrane protein analysis

Alterations in the membrane proteins of BPAECs were analyzed by the sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) according to our previously published procedure16. BPAECs cultured in 35-mm sterile dishes (~90-100% confluence) in were treated with MEM alone or MEM containing different concentrations (2 and 5%) of MβCD or HPCD for 1, 2, and 3 h. At the end of treatment period, equal amounts of total cellular proteins from the untreated control and treated cells were subjected to SDS-PAGE analysis with appropriate protein markers for mol. wt. identification. The proteins resolved on gels were stained with Coomassie blue to examine the changes in cellular proteins following the capture of digital images of gels. Membrane cholesterol replenishment

BPAECs cultured in 35-mm sterile dishes (~90% confluence) were treated with basal MEM or MEM containing MβCD (2% & 5%) for 30-60 min, following which they were supplemented with the water-soluble cholesterol (2%) for 60 min. At the end of treatment time, total cellular cholesterol, LDH release by cells, and cell morphology were determined as described above. Statistical analysis

Experiments were conducted in triplicate and data were expressed as mean ± standard deviation (SD). Statistical analysis was performed by analysis of variance (ANOVA) by using the SigmaStat (Jandel Scientific, San Rafael, California). The level of statistical significance was set at P ≤ 0.05.

HINZEY et al: CYCLODEXTRIN VASCULOTOXICITY

333

Results β-Cyclodextrins cause cholesterol depletion and induce cytotoxicity in vascular ECs

β-Cyclodextrins have been widely used to remove or deplete cholesterol from mammalian cells, in order to establish the role of the cholesterol rafts in cell signaling processes17,18. However, removal of membrane- and/or raft-associated cholesterol by cyclodextrin was also expected to cause the membrane damage, leading to the leak of intracellular macromolecules and cytotoxicity. Along those lines, experiments in the current study revealed that both MβCD and HPCD were effective in depleting cholesterol from BPAECs in a dose- and time-dependent manner, however, HPCD removed significantly less cholesterol than MβCD. MβCD (Fig. 2A) showed significant cholesterol loss at

15 min (5% conc.) and 60 min (2% conc.), whereas HPCD (Fig. 2B) showed significant cholesterol loss at 15 min (both 2% and 5% conc.), as compared to the same in the untreated control cells. The respective amount of cholesterol removed by MβCD treatment, however, was significantly higher than that of HPCD.

MβCD caused a time- and dose-dependent release of LDH from BPAECs. MβCD (Fig. 3A) caused significant LDH leak at 15 min (both 2% and 5% concentrations). On the other hand, only HPCD at a dose of 5% (Fig. 3B) showed significant LDH leak at 120 min of treatment, as compared to the same in the untreated control cells. These results revealed that MβCD caused a greater loss of cholesterol and greater extent of cytotoxicity (LDH leak) in BPAECs in a time- and dose-dependent manner, as compared to that caused by HPCD.

Fig. 2—β-Cyclodextrins cause cholesterol depletion in vascular ECs [Confluent BPAECs (5 × 105 cells/35-mm dish) were treated for 15, 30, 60, or 120 min with basal MEM or MEM containing MβCD (A) or HPCD (B) (2% or 5% wt/vol.). The amount of cholesterol (µg) in cells following the treatments was determined spectrofluorometrically as described in ‘Materials and Methods’. Data represent mean ± SDof three independent experiments. *Significantly different at P≤0.05 as compared to the control cells treated with basal MEM alone]

Fig. 3—β-Cyclodextrins induce cytotoxicity in vascular ECs [Confluent BPAECs (5 × 105 cells/35-mm dish) were treated for 15, 30, 60, or 120 min with basal MEM or MEM containing MβCD (A) or HPCD (B) (2% or 5% wt/vol). The extent of LDH release was analyzed spectrophotometrically using the LDH Assay Kit as described in ‘Materials and Methods’. Data represent mean ± SD of three independent experiments. *Significantly different at P≤0.05 as compared to the control cells treated with basal MEM alone]

INDIAN J. BIOCHEM. BIOPHYS., VOL. 49, OCTOBER 2012

334

β-Cyclodextrins induce mitochondrial dysfunction in vascular ECs

As the earlier experiments of current study showed that β-cyclodextrins caused cytotoxicity along with cholesterol depletion in BPAECs, we further assessed the mitochondria-mediated MTT reduction in BPAECs as an index of the status of cellular mitochondrial function. The results of these studies showed that following 60 min of cholesterol depletion caused by both β-cyclodextrins at the concentrations of 2% and 5% induced a statistically significant decrease in the extent of mitochondrial MTT reduction as compared to the same in the untreated control cells (Fig. 4). Upon further analysis, however, it was seen that MβCD (both 2% and 5%) induced a markedly greater decrease in the mitochondrial function, as compared to that in the untreated control cells. In addition, MβCD (both 2% and 5% dose) also showed a markedly greater extent of suppression in MTT reduction, as compared to that caused by HPCD (both 2% and 5% dose). MβCD induces inhibition cell proliferation in vascular ECs

As the earlier experiments of current study demonstrated that both MβCD and HPCD caused cytotoxicity and loss of mitochondrial function, we investigated the response of cell proliferation in BPAECs to the treatment of both β-cyclodextrins. Proliferation of BPAECs was measured by determining their ability to incorporate [3H]thymidine into the cellular DNA. After 24 h of incubation with [3H]thymidine, MβCD (both 2% and 5%) induced a

markedly significant decline in the incorporation of [3H]thymidine by BPAECs, as compared to the same by the untreated control cells (Fig. 5). However, HPCD (both 2% and 5%) did not cause statistically significant decrease in the cellular incorporation of [3H]thymidine. In addition, at both tested concentrations of MβCD, cells showed a significantly lower ability to incorporate [3H]thymidine, as compared to the HPCD-treated BPAECs. These results demonstrated that only MβCD caused a drastic loss of the cell proliferation ability in BPAEC, as compared to HPCD. MβCD causes cell morphology alterations in vascular ECs

The earlier experiments of the current study revealed that MβCD caused a greater extent of cytotoxicity and loss of cell replication ability as compared to HPCD. Therefore, we investigated the response of cell morphology in BPAECs to the treatment of both β-cyclodextrins. Light microscopy examination of cell morphology revealed that MβCD induced a marked loss of cellular morphology in a time- and dose-dependent fashion (Fig. 6A). On the other hand, HPCD did not appear to cause any observable alterations or loss of cell morphology in BPAECs, as compared to that was caused by MβCD (Fig. 6B). β-Cyclodextrins induce barrier dysfunction of EC monolayer

The earlier experiments of current study revealed that MβCD induced greater extent of cytotoxicity, loss of mitochondrial function, decline in cell

Fig. 4—β-Cyclodextrins induce mitochondrial dysfunction in vascular ECs [Confluent BPAECs (2 × 105 cells/17.5-mm dish) were treated with basal MEM or MEM containing MβCD or HPCD (2% or 5% wt/vol) for 60 min, following which cellular MTT reduction was assayed spectrophotometrically as described in ‘Materials and Methods’. Data represent mean ± SD of three independent experiments. *Significantly different at P≤0.05 as compared to the control cells treated with basal MEM alone]

Fig. 5—MβCD induces inhibition cell proliferation in vascular ECs [Confluent BPAECs (5 × 105 cells/35-mm dish) were treated with basal MEM or MEM containing MβCD or HPCD (2% or 5%) for 60 min, following which the medium was removed and cells were incubated with [3H]thymidine (1 µCi/ml) for 24 h. The cell replication at the end of treatment was assayed by determining [3H]thymidine incorporation into the cells as outlined in ‘Materials and Methods’. Data represent mean ± SD of three independent experiments. *Significantly different at P≤0.05 as compared to the control cells treated with basal MEM alone]

HINZEY et al: CYCLODEXTRIN VASCULOTOXICITY

335

proliferation ability, and loss of cell morphology in BPAECs as compared to that caused by HPCD. Hence we studied the EC barrier function maintained by the tight junctions in BPAEC monolayer under the treatment of chosen β-cyclodextrins by determining the TER on ECIS. MβCD and HPCD (1, 2 and 5% dose) caused a dose-dependent decrease of TER in BPAEC monolayers up to 2 h of the treatment of cells with β-cyclodextrin, which indicative of the alterations of tight junctions in the cells (Fig. 7A). Noticeably, HPCD also induced a dose-dependent decline in the TER in BPAEC monolayers, but the

response was less than that was observed under MβCD treatment (Fig. 7B). Also, HPCD at any tested dose caused a marked decline of TER in BPAEC monolayers up to 10 h of treatment, but slightly to a lesser extent, as compared to that induced by MβCD. β-Cyclodextrins cause cytoskeletal reorganization in ECs

The earlier experiments of current study demonstrated that both MβCD, in particular caused cell morphology alterations and both MβCD and HPCD caused decrease in the TER in BPAEC monolayers. As the actin cytoskeletal reorganization

Fig. 6—MβCD causes cell morphology alterations in vascular ECs [Confluent BPAECs (5 × 105 cells/35-mm dish) were treated with basal MEM or MEM containing MβCD (A) or HPCD (B) (2% or 5% wt/vol.) for 15, 30, 60, or 120 min as described in ‘Materials and Methods’. At the end of the treatment time, cells were examined under light microscopy at 10X or 100X magnification. Images were captured digitally. Each micrograph is representative of three independent observations]

Fig. 7—β-Cyclodextrins induce barrier dysfunction of EC monolayer [Confluent BPAECs (grown on sterile gold electrodes) were treated with basal MEM or MEM containing MβCD (A) or HPCD (B) (1%, 2% or 5% wt/vol) while being subjected to electric cell substrate impedance sensing (ECIS). TER values were plotted every few seconds and continuous measurements made from the acquired data points as described in ‘Materials and Methods’. Each treatment was run in duplicate and averaged, and each well has 10 microelectrodes connected in parallel to average the ECIS response over a large number of cells]

INDIAN J. BIOCHEM. BIOPHYS., VOL. 49, OCTOBER 2012

336

has been observed as a pre-requisite for the downstream tight junction alterations and loss of the TER in ECs, here, we examined whether the chosen β-cyclodextrins would induce the actin cytoskeletal rearrangement (stress fiber formation) in BPAECs. Both MβCD and HPCD (2% and 5%) treatment for 1 h induced marked actin stress fiber formation, as observed from the confocal fluorescence microscopic examination. The actin cytoskeletal reorganization and cell morphology alterations were markedly worse in MβCD-treated cells (Fig. 8A), as compared to that in the HPCD-treated BPAECs (Fig. 8B). MβCD causes loss of membrane lipid fatty acids in vascular ECs

Following our earlier observations in current study that both the tested β-cyclodextrins, more so MβCD, induced cytotoxicity, loss of cell morphology, and decline in TER in BPAECs, we investigated whether MβCD and HPCD would cause alterations in the membrane lipid fatty acid composition of BPAECs. Cells treated with MEM alone or MEM containing MβCD (2% and 5%) for 1-3 h exhibited significant losses of the saturated, monounsaturated and

polyunsaturated fatty acids of the membrane lipids in a dose- and time-dependent fashion [Figs. 9A-C]. On the other hand, HPCD did not appear to cause any marked and significant loss of fatty acids of membrane lipids in BPAECs, as observed in the cells treated with MβCD under identical conditions [Figs 9D-F]. β-Cyclodextrins induce alterations of proteins in vascular ECs

Upon observing that MβCD caused the loss of fatty acids of membrane lipids in BPAECs in earlier experiments of current study, we investigated whether MβCD and HPCD could cause alterations of on cellular proteins in BPAECs. MβCD was observed to cause a robust dose- and time-dependent loss of several cellular proteins and the formation of several new protein bands which indicated MβCD-induced alterations in the protein profile of BPAECs (Fig. 10A). In contrast, HPCD did not show marked loss of cellular proteins until higher concentrations of HPCD were used at longer treatment times (Fig. 10B). These results demonstrated that both β-cyclodextrins, MβCD more drastically altered the protein profile of BPAECs.

Fig. 8—β-Cyclodextrins cause cytoskeletal reorganization in ECs [Confluent BPAECs (grown on sterile glass coverslips) were treated with basal MEM or MEM containing MβCD (A) or HPCD (B) (2% or 5% wt/vol.) and stained for actin as described in ‘Materials and Methods’. Images of the actin cytoskeletal rearrangement (stress fiber formation) were captured under the confocal fluorescence microscope at 63X magnification. Each image is a representative of three independent observations]

HINZEY et al: CYCLODEXTRIN VASCULOTOXICITY

337

Replenishment of cholesterol protects against MβCD-induced cytotoxicity in vascular ECs

We conducted experiments to establish the physiological relevance of cholesterol depletion that would be a prerequisite for the MβCD-induced cytotoxicity in BPAECs. As revealed by the earlier experiments of current study, after 60 min of

treatment, MβCD (2% and 5%) caused significant decrease in the cellular cholesterol levels, as compared to the untreated control cells (Fig. 11). Nevertheless, replenishment with water-soluble cholesterol increased the cellular cholesterol content above that was observed in the control unsupplemented cells and completely and significantly attenuated the

Fig. 9—MβCD causes loss of membrane lipid fatty acids in vascular ECs [Confluent BPAECs (grown on sterile 100-mm dishes) were treated with basal MEM or MEM containing MβCD (A, B, C) or HPCD (D, E, F) (2% or 5% wt/vol.) for 1, 2 and 3 h, respectively and then membrane lipids were extracted, methylated and subjected to GC-MS analysis as described in ‘Materials and Methods’. Data represent mean ± SD of three independent experiments. *Significantly different at P ≤ 0.05 as compared to the control cells treated with basal MEM alone Key for fatty acid designations: 16:0 = palmitic acid; 16:1 = palmitoleic acid; 18:0 = stearic acid; 18:1 = oleic acid; 18:2 = linoleic acid; 20:3 = eicosatrienoic acid; 20:4 = arachidonic acid; 20:5 = eicosapentaenoic acid; 22:4 = docosatetraenoic acid; 22:5 = docosapentaenoic acid; 22:6 = docosahexaenoic acid]

INDIAN J. BIOCHEM. BIOPHYS., VOL. 49, OCTOBER 2012

338

MβCD-induced loss of cholesterol in BPAECs. MβCD (2% and 5%) treatment for 60 min, followed by replacement with basal MEM showed a significant extent of cytotoxicity (LDH leak) at both 30 and 60 min of incubation (Fig. 12). Furthermore, MβCD (2%) depletion for 60 min, followed by the

supplementation with water-soluble cholesterol (2%) demonstrated a significant reduction in cyclodextrin- induced cytotoxicity after 60 min of cholesterol rescue. These results were also corroborated upon examination of the cellular morphology with light microscopy (Fig. 13). BPAECs treated with the water-

Fig. 10—β-Cyclodextrins induce alterations of proteins in vascular ECs [Confluent BPAECs (5 × 105 cells/35-mm sterile dish) were treated with basal MEM or MEM containing MβCD (A) or HPCD (B) ( 2% or 5% wt/vol) for 1, 2 and 3 h and then proteins were subjected to the SDS-PAGE analysis, followed by staining with Coomassie blue as described in ‘Materials and Methods’. Gels were scanned and images were compiled to compare side by side for the differences between treatments. Each gel is a representative of duplicate observations]

Fig. 11—Cholesterol removal by MβCD and replenishment by cholesterol supplementation in vascular ECs [Confluent BPAECs (5 × 105 cells/35-mm sterile dish) were treated with basal MEM or MEM containing MβCD (2% or 5%) for 60 min. The cells were then challenged with water-soluble cholesterol (2% or 5%) for 60 min to replenish β-cyclodextrin-depleted cellular cholesterol. At the end of treatment time, cellular cholesterol was determined as described in ‘Materials and Methods’. Data represent mean ± SDof three independent experiments. *Significantly different at P ≤ 0.05, as compared to the untreated control cells. **Significantly different from MβCD-treated cells at P ≤ 0.05]

Fig. 12—Replenishment of cholesterol protects against MβCD-induced cytotoxicity in vascular ECs. Confluent BPAECs (5 × 105

cells/35-mm sterile dish) were treated with basal MEM or MEM containing MβCD (2%) for 60 min, following which they were supplemented with water-soluble cholesterol (2%) for 30 and 60 min. At the end of both challenge times, LDH release into the medium was assayed spectrophotometrically as described in ‘Materials and Methods’. Data represent mean ± SD of three independent experiments. *Significantly different at P ≤ 0.05 as compared to untreated control cells. **Significantly different from MβCD-treated cells alone at P≤0.05]

HINZEY et al: CYCLODEXTRIN VASCULOTOXICITY

339

soluble cholesterol alone (2%) did not exhibit any observable cell morphology alterations. Following 60 min of cholesterol depletion with MβCD (2%), a significant loss of cell morphology was observed. However, upon cholesterol replenishment for 60 min, the MβCD-induced cell morphology alterations were reversed and the cell morphology was restored, as compared to the untreated control cells. Discussion

The current study demonstrated that cellular cholesterol, especially the lipid-raft associated type was a critical player in the vascular EC structure and viability. In order to study the role of cholesterol in cellular viability and function, two well characterized β-cyclodextrins (MβCD and HPCD) were utilized as tools to deplete cholesterol from the BPAECs in the present study. The cyclodextrins are commonly used in the manufacture of dietary preparations and have been widely utilized as the experimental tools to remove or deplete cholesterol from mammalian cells17,18. Among the β-cyclodextrins, MβCD appears to be a popular cyclodextrin that is being widely utilized in removing cholesterol from the plasma membrane and lipid rafts of the mammalian cells1. It has been reported that up to 2% (wt/vol) or 10 mM concentration of MβCD has been widely used in several studies with the cultured mammalian cell

models to remove cholesterol, especially to elucidate the role of cholesterol rafts in cellular functions19-21. Also, many of these investigations have reported that cellular models have been subjected to the MβCD treatment for extended periods of time as long as 4 h. On the other hand, MβCD has also been reported as a toxic β-cyclodextrin17. Furthermore, MβCD has been demonstrated to cause the disruption of lipid rafts and apoptosis in keratinocytes19. Taken together, studies reported so far have revealed that MβCD is cytotoxic to mammalian cells in culture.

Vascular endothelium is pivotal for the normal maintenance of vasculature and homeostasis of cardiovascular system. The critical role of cholesterol lipid rafts in vascular EC structure, functions and signaling cascades under both physiological and pathophysiological settings has been aggressively investigated7-9. Earlier, we have reported the use of MβCD and HPCD as tools to deplete cholesterol from lipid rafts for elucidation of the role of cholesterol rafts in oxidant-mediated lipid signaling in vascular ECs22. However, thorough investigations on the adverse effects of the widely used β-cyclodextrins as the cellular lipid raft-depleting compounds in vascular ECs are needed at this juncture. Hence, the current study focused on the cytotoxicity and adverse effects caused by the cholesterol-depleting cyclodextrin (MβCD) and its counterpart, HPCD in the well-established vascular EC model, BPAECs.

The current study although demonstrated that MβCD was effective in depleting cholesterol from BPAECs, its cytotoxicity was unavoidable. This cytotoxic effect was also observed through the alterations of cellular morphology. HPCD, the less often used cyclodextrin in the cellular lipid raft studies was shown in the current study to far less cytotoxic to BPAECs as compared to MβCD, while effectively removing the cellular cholesterol. The alterations in cellular morphology induced by the HPCD treatment were markedly less drastic, as compared to MβCD. However, the adverse or cytotoxic effects of MβCD as observed in the current study did exclusively arise from its cholesterol-depleting action or from its actions of perturbing the crucial EC biochemical machinery, including the loss of the membrane lipid fatty acids, alterations in cellular proteins and cytoskeletal rearrangement need to be firmly established (Scheme I).

Nevertheless, the current study revealed that both MβCD and HPCD caused the tight junction

Fig. 13—Rescue of MβCD-induced cell morphology alterations by cholesterol replenishment in vascular ECs [Confluent BPAECs (5 × 105 cells/35-mm sterile dish) were treated with basal MEM or MEM containing MβCD (2%) for 60 min, following which they were supplemented with water-soluble cholesterol (2%) for 60 min. At the end of treatment time, cellular morphology was examined by light microscopy as described in ‘Materials and Methods’ at 100X magnification. Images were captured digitally. Each micrograph is representative of three independent observations]

INDIAN J. BIOCHEM. BIOPHYS., VOL. 49, OCTOBER 2012

340

alterations, leading to the decrease in TER in BPAEC monolayer, suggesting the β-cyclodextrins did cause cellular physiological alterations, such as inducing the weakening of the cell-to-cell adherence. More strikingly, replenishing the cells with cholesterol following the MβCD-induced cholesterol depletion in BPAECs clearly showed protection against the MβCD-induced cytotoxicity, as observed in the current study. This observation suggested that cholesterol depletion from the lipid rafts caused by MβCD partly contributed to the cytotoxicity of MβCD in BPAECs. As opposed to MβCD, HPCD was observed as a safer β-cyclodextrin without causing greater extent of adverse cellular effects but as an effective cellular cholesterol-depleting compound in the vascular ECs. It could be deduced that the less polar MβCD could have contributed to its more potent adverse and cytotoxic effects on the vascular ECs, as observed in the present study.

The current study further investigated β-cyclodextrin-induced alterations of cell viability through the mitochondrial reduction of MTT, as an index of the mitochondrial dysfunction. Both MβCD

and HPCD exhibited a decrease in the MTT reduction by BPAECs, but the impact of MβCD was markedly more drastic. Also, cholesterol depletion caused by MβCD appeared to render BPAECs to lose their cell proliferation ability greater than that was caused by HPCD. Inability of BPAECs to incorporate [3H]thymidine into the cells could be attributed to their inability to successfully replicate the cellular DNA. Therefore, removal of cholesterol and leakage of other vital cellular components was more drastic when MβCD was utilized as the cholesterol-depleting compound, thus leading to a drastic decline of EC replication. The results also suggested that MβCD-mediated depletion of plasma membrane- and lipid raft-associated cholesterol in conjunction with the cytotoxic effects of MβCD might have contributed to the loss of proper energy metabolism, maintenance of the nucleotide pools and damage to the nuclear machinery, leading to the loss of cell proliferation ability in BPAECs.

In conclusion, the results of this study showed that plasma membrane- and/or lipid raft-associated cholesterol was a critical player in maintaining

Scheme I—Proposed mechanism(s) of β-cyclodextrin-induced cholesterol depletion form plasma membrane and lipid rafts of vascular ECs leading to membrane leak of LDH (cytotoxicity), decline in mitochondrial function, alterations in cell morphology, loss of ability to proliferate, loss of membrane lipid fatty acids, altered proteins, actin cytoskeletal reorganization, and barrier dysfunction.

HINZEY et al: CYCLODEXTRIN VASCULOTOXICITY

341

vascular EC morphology, viability and functions. Removal of cholesterol by β-cyclodextrin (especially MβCD) treatment apparently caused the loss of fluidity of cell membrane and leakage of vital cellular components and thus caused barrier dysfunction and loss of cell morphology in BPAECs (Scheme I). Also, the current study offered a safer method of removal of cholesterol by utilizing HPCD that did not induce extensive loss of cell viability as seen with the MβCD treatment, for studies to investigate the role of lipid raft-associated cholesterol in cellular functions. Overall, this work also emphasized the critical nature of membrane- and lipid raft-associated cholesterol in the vascular EC structural integrity and functions. Acknowledgements

Technical support provided by Valorie Ciapala, Jessica N. Mazerik and M Lakshmi Kuppusamy is grately appreciated. Financial support provided by the Dorothy M. Davis Heart and Lung Research Institute and Division of Pulmonary, Allergy, Critical care, and Sleep Medicine of The Ohio State University, the International Academy of Oral Medicine and Toxicology (IAOMT), and the National Institutes of Health (HL093463) is acknowledged. References

1 Zajchowski L D & Robbins S M (2002) Eur J Biochem 269, 737-752

2 Mishra S & Preeti G J (2007) J Neurochem 103, 135-142 3 Mukherjee S, Zha X, Tabas I & Maxfield F R (1998)

Biophys J 75, 1915-1925 4 Zidovetzki R & Levitan I (2007) Biochim Biophys Acta

1768, 1311-1324 5 Xia M, Ling W, Zhu H, Wang Q, Ma J, Hou M, Tang Z,

Li L & Ye Q (2007) Arterioscler Thromb Vasc Biol 27, 519-524

6 Xia M, Wang Q, Zhu H, Ma J, Hou M, Tang Z, Li J & Ling W (2007) Biochem Biophys Res Commun 361, 768-774

7 Bae J S & Rezaie A R (2010) Thromb Res 125(1), e9-e15 8 Betz J, Bielaszewska M, Thies A, Humpf H U, Dreisewerd

K, Karch H, Kim K S, Friedrich A W & Müthing J (2011) J Lipid Res 52, 618-634

9 Meng G, Liu Y, Lou C & Yang H (2010) Brit J Pharmacol 161, 1628-1644

10 Varadharaj S, Steinhour E, Hunter M G, Watkins T, Baran C P, Magalang U J, Kuppusamy P, Zweier J L, Marsh C B, Natarajan V & Parinandi N L (2006) Cell Signal 18, 1396-1407

11 Patel R B, Kotha R S, Sauers L A, Malireddy S, Gurney T O, Gupta N N, Elton T S, Magalang U J, Marsh C B, Haley B E & Parinandi N L (2012) Toxicol Mech Methods 22, 383-396

12 Mazerik J N, Hagele T, Sherwani S, Ciapala V, O’Connor Butler E S, Kuppusamy M L, Hunter M, Kuppusamy P, Marsh C B & Parinandi N L (2007) Int J Toxicol 26, 553-569

13 Sliman S M, Eubank T D, Kotha S R, Kuppusamy M L, Sherwani S I, Butler E S, Kuppusamy P, Roy S, Marsh C B, Stern D M & Parinandi N L (2010) Mol Cell Biochem 333, 9-26.

14 Parinandi N L, Weis B K, Natarajan V & Schmid H H (1990) Arch Biochem Biophys 280, 45-52

15 Kodavanti U P, Thomas R, Ledbetter A D, Schladweiler M C, Shannahan J H, Wallenborn J G, Lund A K, Campen M J, Butler E O, Gottipolu R R, Nyska A, Richards J E, Andrews D, Jaskot R H, McKee J, Kotha S R, Patel R B & Parinandi N L (2011) Environ Hlth Perspect 119, 312-318

16 Parinandi N L, Zwizinski C W & Schmid H H (1991) Arch Biochem Biophys 289, 118-123

17 Stella V J & He Q (2008) Toxicol Pathol 36, 30-42 18 Ostrom R S & Liu X (2007) Methods Mol Biol 400, 459-468 19 Bang B, Gniadecki R & Gajkowaska B (2005) Exp Dermatol

14, 266-272 20 Pottosin I I, Valencia-Cruz G, Bonales-Alatorre E,

Shabala S N & Dobrovin-Skya O R (2007) Eur J Physiol 454, 235-244

21 Larbi A, Douziech N, Khalil A, Dupuis G, Gherari S, Guerard K P & Fulop T Jr (2004) Exp Gerentol 39, 551-558

22 Kline M A, O'Connor Butler E S, Hinzey A, Sliman S, Kotha S R, Marsh C B, Uppu R M & Parinandi N L (2010) Methods Mol Biol 610, 201-211