Brain Research, 597 (1992) 108-113 © 1992 Elsevier Science Publishers B.V. All rights reserved 0006-8993/92/$05.00
108
BRES 18293
Galactocerebroside mediates
Ca 2+
signaling in cultured glioma cells
Preeti G. Joshi a n d S u d h a M i s h r a Department of Biophysics, National h~stitute of Mental Health and Neuro Sciences, Bangalore (India) (Accepted 30 June 1992)
Key words: Galactocerebroside; Glioma ceil; Anti-galactocerebroside; Transmembrane signaling; Intraceilular calcium; Fara-2
Expression of galactocerebroside (GaiC) was detected in human glioma cell line (U-87 MG). Exposure of cells to antibody against GalC and fluoresceinated second antibody showed i,.tcnse fluorescence on the plasma membrane. Possible ir~voivement of GalC in receptor-mediated transmembrane signaling was explored m this cell line. Antibodies raised against GalC were used as iigands. Binding of anti-GalC to these cells caused a transient increase in intrac¢llular free calcium ([Ca`'+ ]i). The response was observed bo~.h in the presence and absence of extraceilular calcium demonstrating that the rise in [Ca 2+ ], induced by anti-GaIC was due to an influx of Ca 2+ through plasma membrane as well as the release of Ca "~ from intracellular pools. Ca 2+ influx was blocked by verapamil, indicating that influx is mediated by voltage-sensitive channels. O,r ~esults suggest that GaIC can play a role in transmembrane signaling by modulation of voltage-sensitive Ca 2÷ channels.
INTRODUCTION
Giycosphingolipids are ubiquitous components of plasma membrane of mammalian cells and are especially enriched in nervous tissue. Several studies have implicated their involvement in a variety of cellula: functions such as cell growth, differentiation, immune response and recognition5,1t'.~s, however, the mechanisms underlying these effects are not clear. Recently some reports have appeared which suggest a role for glycosphingolipids in cellular signal transduction. Ganglioside GM1 stimulates neuritogenesis in cells with differentiation potential 28-3°. GM1 has been shown to modulate cyclic AMP levels, breakdown of phosphoinositides and a variety of kinase activities ~9,~6,as. Wu and Ledeen demonstrated that gangliosidc GM1 stimulates neuritogenesis in cultured neuroblastoma cells by a mechanism involving modulation of Ca 2+ fluxesa°. Transmembrane signaling events induced by binding of cholera toxin to GM1 in lymphocytes are well studied 14. Dixon et al. ~ have shown that the B subunit of cholera toxin which binds exclusively to GM1 ganglioside on cell surface stimulates proliferation of rat thymocytes which is mediated by an increase in intracellular calcium [Ca 2 + ]i" Recently Harouse et al. ~7 have suggested a role for ga!actocereb.,-osi~e (GalC) as a receptor for human
immunodeficiency virus in the nervous system. Antibodies to GalC inhibited the uptake and infection of HIV-1 in neural cell lines U-373 MG and SK-N-MC. Bhat et al. 3 have further demonstrated that the viral receptor binding protein gpl20 specifically binds to GalC. Dyer et al. i°-~2 have studied the function of GalC in oligodendroglial cells. It was found that the interaction of antibodies to GalC and sulfatide with cultured oligodendroglial cells initiates a sequence of events leading to dramatic changes in cell morphology and myelination. Subsequently they showed that binding of anti-GalC to oligodendroglial cells caused a sustained increase in intracellular calcium concentration, leading to disassembly of microtubules and suggest~,d a role for GaIC in transmembrane signaling~2. In the present study we have found that the clonal human glioma cell line U-87 MG expresses GalC. Interaction of GaiC-specific antibodies with these cells was found to stimulate C a 2+ mobilization. Our results demonstrate that GaIC participates in regulation of v o l t a g e - s e n s i t i v e C a 2+ channels in U-87 MG cells. MATERIALS AND METHODS
Materials Eagle's minimum essential medium (EMEM) and phosphatebuffered saline (PBS) were purchased from Hi Media Bombay (India), fetal calf serum from Northumbria Biological Ltd., Cramlington
CJrrespondence: P.G. Joshi, Department of Biophysics, NIMHANS, Hosur Road, Bangalare-560 029, India.
109 (UK), sodium pyruvate, sulfinpyrazone and digitonin from Sigma Chemical Company, St. Louis. MO (USA) and dispase from Boehringer Mannheim GmbH, Mannheim (FRG). Fura-2 acet o , m e t h y l ester and pluronic were purchased from Molecular Probes Inc., Eugene, OR (USA). Other chemicals were obtained from commercial sources. Bovine serum and normal goat serum were prepared in our laboratory. U-87 MG cell line was obtained from American type culture collection, Rockviile, MD (USA). Monoclonal and polyclonal antibodies against GalC (IgG), monoclonal antibody against sulfatide (IgM), rhodamine-conjugated goat anti-mouse lgM were gifts from Dr. Joyce A. Benjamins. Monoclonal antibodies were isolated from hybridoma cell lines whereas polyclonal antibody was raised in rabbit 1"2'l°. Fluorescein-conjugated second antibodies, goat anti-rabbit (GAR-FITC) IgG, goat anti-mouse (GAM-FITC) IgG were obtained from Bangalore Genie, Bangalore (India).
Cell culture U-87 MG cells were cultured in 25 cm 2 plastic culture flasks (Nunc). The flasks were seeded at a density of 1.6 × 104 cells/cm 2 in EMEM containing 1 mM sodium pyruvate, 5% fetal calf serum, 5% bovine serum, 2.2 mg/l HEPES, 50,000 units/I benzyl penicillin, 3500 units/! streptomycin and 2.2 rag/! nystatin. Cells were grown at 37°C. Medium was changed once after 72 h of growth. Experiments were performed after 96 h of growth.
Immunocytochemistry Immunostaining procedure was used to detect the presence of GaiC and sulfatide in U-87 MG cells. Cells were harvested with 1% dispase containing 0.02% glucose and 0.03% EGTA. Cell smears were prepared on glass slides fixed with 4% paraformaldehyde, washed with PBS and treated with 10% normal goat serum for 30 min. Cells were then exposed to 1:15 and 1:40 dilutions of polyclonal or monoclonal anti GaIC IgG or 1 : 15 dilution of anti-sulfatide for l h at room temperature. After exposure to primary antibodies
cells were washed with PBS and stained for 30 min with fluoresceinconjugated goat anti-rabbit IgG (GAR, 1:200), goat anti-mouse IgG (GAM, l:100), or rhodamine-conjugated goat anti-mouse IgM (1 : 100). Control cultures were treated with IgG isolated from rabbit prebleed serum, The whole procedure was performed a~ ~oom tern. perature.
Measurement of intracelhdar calcium Intracellular calcium concentration was measured using fluorescent Ca 2+ indicator dye fura-2. Dye loading was performed in monolayers of cells grown in plastic culture flasks. After 96 h of growth the monolayers were washed with buffer containing 130 mM NaCI, 5 mM KCI, 1 mM MgC! 2, 1 mM CaCI 2, 5 mM glucose and 20 mM Tris (pH 7.4). Cells were then incubated with the same buffer containing 2 p,M fura-2 acetoxy methyl ester (fura-2/AM). Fura2 / A M was thoroughly mixed with pluronic (2% final concentration) before addition to cells. Incubation was done in a shaker bath maintained at 37°C. After 15 min incubation, monolayers were washed 2 times with buffer and cells were released using 1% dispase. The cell suspension was washed again and resuspended in buffer without Ca 2+. In initial experiments we noticed that loaded dye was leaking out of the cells at a very fast rate. We therefore used sulfinpyrazone, an inhibitor of anion transport s, in the loading buffer to reduce the leakage of dye. We tried different concentrations of sulfinpyrazone and found that 500 p,M was the optimal concentration required to block the leakage of dye. Consequently 500 p,m sulfinpyrazone was present in the buffer during and after dye loading. Fluorescence measurements were performed with a SLM 8000C spectrofluorometer. Cells were suspended at a density of 0.5 x 106/mi and fluorescence intensities at 340 and 380 nm excitation wavelengths were measured keeping the emission wavelength fixed at 500 nm. We noticed that upon addition of 1 mM EGTA to cell suspension the fluorescence intensity, instantaneously reduced by 5-8%. This observation indicated that a small amount of fura-2 was bound to the plasma membrane of U-87 MG cells. [Ca 2 + ]i was estimated as described by Grynkiewicz et al? s. Calibration was performed by
Fig. 1. lmmunostaining of cultured human glioma U-87 MG cells with anti-GalC. Cells were fixed prior to incubation with polyclonal anti-GalC (1:40 dilution) and fluorescei1,.~ted secondary antibody.
110 measuring the fluorescence in cells permeabilized with 1% digitonin in the presence of 1 mM CaCI 2 or 1 mM EGTA. A macro progla,;, was used to calculate the 340/380 nm intensity ratio and [Ca `'+ ]i simultaneously during the fluorescence measurement and the K d value used was 225 riM. Fluorescence from control samples not loaded with fura-2 was subtracted before calculating 340/380 intensity ratio. While calculating [Ca 2 + ]i we have not applied any correction for the fluorescence intensity associated with membrane-bound fura-2. Changes in [Ca -'+ ]i induced by antibodies were determined by measuring the time course of fluorescence intensities at 340 and 380 nm. During the measurement cell suspension was continuously stirred in cuvette with a magnetic stirrer. Antibodies and other reagents were added through an injection port on the lid of the sample compartment of fluorometer. All fluorescence measurements were performed at room temperature (25°0.
7.0 G.0
815
555
•.-- 5-0 0 al)
357 £ ,,
108
BRES 18293
Galactocerebroside mediates
Ca 2+
signaling in cultured glioma cells
Preeti G. Joshi a n d S u d h a M i s h r a Department of Biophysics, National h~stitute of Mental Health and Neuro Sciences, Bangalore (India) (Accepted 30 June 1992)
Key words: Galactocerebroside; Glioma ceil; Anti-galactocerebroside; Transmembrane signaling; Intraceilular calcium; Fara-2
Expression of galactocerebroside (GaiC) was detected in human glioma cell line (U-87 MG). Exposure of cells to antibody against GalC and fluoresceinated second antibody showed i,.tcnse fluorescence on the plasma membrane. Possible ir~voivement of GalC in receptor-mediated transmembrane signaling was explored m this cell line. Antibodies raised against GalC were used as iigands. Binding of anti-GalC to these cells caused a transient increase in intrac¢llular free calcium ([Ca`'+ ]i). The response was observed bo~.h in the presence and absence of extraceilular calcium demonstrating that the rise in [Ca 2+ ], induced by anti-GaIC was due to an influx of Ca 2+ through plasma membrane as well as the release of Ca "~ from intracellular pools. Ca 2+ influx was blocked by verapamil, indicating that influx is mediated by voltage-sensitive channels. O,r ~esults suggest that GaIC can play a role in transmembrane signaling by modulation of voltage-sensitive Ca 2÷ channels.
INTRODUCTION
Giycosphingolipids are ubiquitous components of plasma membrane of mammalian cells and are especially enriched in nervous tissue. Several studies have implicated their involvement in a variety of cellula: functions such as cell growth, differentiation, immune response and recognition5,1t'.~s, however, the mechanisms underlying these effects are not clear. Recently some reports have appeared which suggest a role for glycosphingolipids in cellular signal transduction. Ganglioside GM1 stimulates neuritogenesis in cells with differentiation potential 28-3°. GM1 has been shown to modulate cyclic AMP levels, breakdown of phosphoinositides and a variety of kinase activities ~9,~6,as. Wu and Ledeen demonstrated that gangliosidc GM1 stimulates neuritogenesis in cultured neuroblastoma cells by a mechanism involving modulation of Ca 2+ fluxesa°. Transmembrane signaling events induced by binding of cholera toxin to GM1 in lymphocytes are well studied 14. Dixon et al. ~ have shown that the B subunit of cholera toxin which binds exclusively to GM1 ganglioside on cell surface stimulates proliferation of rat thymocytes which is mediated by an increase in intracellular calcium [Ca 2 + ]i" Recently Harouse et al. ~7 have suggested a role for ga!actocereb.,-osi~e (GalC) as a receptor for human
immunodeficiency virus in the nervous system. Antibodies to GalC inhibited the uptake and infection of HIV-1 in neural cell lines U-373 MG and SK-N-MC. Bhat et al. 3 have further demonstrated that the viral receptor binding protein gpl20 specifically binds to GalC. Dyer et al. i°-~2 have studied the function of GalC in oligodendroglial cells. It was found that the interaction of antibodies to GalC and sulfatide with cultured oligodendroglial cells initiates a sequence of events leading to dramatic changes in cell morphology and myelination. Subsequently they showed that binding of anti-GalC to oligodendroglial cells caused a sustained increase in intracellular calcium concentration, leading to disassembly of microtubules and suggest~,d a role for GaIC in transmembrane signaling~2. In the present study we have found that the clonal human glioma cell line U-87 MG expresses GalC. Interaction of GaiC-specific antibodies with these cells was found to stimulate C a 2+ mobilization. Our results demonstrate that GaIC participates in regulation of v o l t a g e - s e n s i t i v e C a 2+ channels in U-87 MG cells. MATERIALS AND METHODS
Materials Eagle's minimum essential medium (EMEM) and phosphatebuffered saline (PBS) were purchased from Hi Media Bombay (India), fetal calf serum from Northumbria Biological Ltd., Cramlington
CJrrespondence: P.G. Joshi, Department of Biophysics, NIMHANS, Hosur Road, Bangalare-560 029, India.
109 (UK), sodium pyruvate, sulfinpyrazone and digitonin from Sigma Chemical Company, St. Louis. MO (USA) and dispase from Boehringer Mannheim GmbH, Mannheim (FRG). Fura-2 acet o , m e t h y l ester and pluronic were purchased from Molecular Probes Inc., Eugene, OR (USA). Other chemicals were obtained from commercial sources. Bovine serum and normal goat serum were prepared in our laboratory. U-87 MG cell line was obtained from American type culture collection, Rockviile, MD (USA). Monoclonal and polyclonal antibodies against GalC (IgG), monoclonal antibody against sulfatide (IgM), rhodamine-conjugated goat anti-mouse lgM were gifts from Dr. Joyce A. Benjamins. Monoclonal antibodies were isolated from hybridoma cell lines whereas polyclonal antibody was raised in rabbit 1"2'l°. Fluorescein-conjugated second antibodies, goat anti-rabbit (GAR-FITC) IgG, goat anti-mouse (GAM-FITC) IgG were obtained from Bangalore Genie, Bangalore (India).
Cell culture U-87 MG cells were cultured in 25 cm 2 plastic culture flasks (Nunc). The flasks were seeded at a density of 1.6 × 104 cells/cm 2 in EMEM containing 1 mM sodium pyruvate, 5% fetal calf serum, 5% bovine serum, 2.2 mg/l HEPES, 50,000 units/I benzyl penicillin, 3500 units/! streptomycin and 2.2 rag/! nystatin. Cells were grown at 37°C. Medium was changed once after 72 h of growth. Experiments were performed after 96 h of growth.
Immunocytochemistry Immunostaining procedure was used to detect the presence of GaiC and sulfatide in U-87 MG cells. Cells were harvested with 1% dispase containing 0.02% glucose and 0.03% EGTA. Cell smears were prepared on glass slides fixed with 4% paraformaldehyde, washed with PBS and treated with 10% normal goat serum for 30 min. Cells were then exposed to 1:15 and 1:40 dilutions of polyclonal or monoclonal anti GaIC IgG or 1 : 15 dilution of anti-sulfatide for l h at room temperature. After exposure to primary antibodies
cells were washed with PBS and stained for 30 min with fluoresceinconjugated goat anti-rabbit IgG (GAR, 1:200), goat anti-mouse IgG (GAM, l:100), or rhodamine-conjugated goat anti-mouse IgM (1 : 100). Control cultures were treated with IgG isolated from rabbit prebleed serum, The whole procedure was performed a~ ~oom tern. perature.
Measurement of intracelhdar calcium Intracellular calcium concentration was measured using fluorescent Ca 2+ indicator dye fura-2. Dye loading was performed in monolayers of cells grown in plastic culture flasks. After 96 h of growth the monolayers were washed with buffer containing 130 mM NaCI, 5 mM KCI, 1 mM MgC! 2, 1 mM CaCI 2, 5 mM glucose and 20 mM Tris (pH 7.4). Cells were then incubated with the same buffer containing 2 p,M fura-2 acetoxy methyl ester (fura-2/AM). Fura2 / A M was thoroughly mixed with pluronic (2% final concentration) before addition to cells. Incubation was done in a shaker bath maintained at 37°C. After 15 min incubation, monolayers were washed 2 times with buffer and cells were released using 1% dispase. The cell suspension was washed again and resuspended in buffer without Ca 2+. In initial experiments we noticed that loaded dye was leaking out of the cells at a very fast rate. We therefore used sulfinpyrazone, an inhibitor of anion transport s, in the loading buffer to reduce the leakage of dye. We tried different concentrations of sulfinpyrazone and found that 500 p,M was the optimal concentration required to block the leakage of dye. Consequently 500 p,m sulfinpyrazone was present in the buffer during and after dye loading. Fluorescence measurements were performed with a SLM 8000C spectrofluorometer. Cells were suspended at a density of 0.5 x 106/mi and fluorescence intensities at 340 and 380 nm excitation wavelengths were measured keeping the emission wavelength fixed at 500 nm. We noticed that upon addition of 1 mM EGTA to cell suspension the fluorescence intensity, instantaneously reduced by 5-8%. This observation indicated that a small amount of fura-2 was bound to the plasma membrane of U-87 MG cells. [Ca 2 + ]i was estimated as described by Grynkiewicz et al? s. Calibration was performed by
Fig. 1. lmmunostaining of cultured human glioma U-87 MG cells with anti-GalC. Cells were fixed prior to incubation with polyclonal anti-GalC (1:40 dilution) and fluorescei1,.~ted secondary antibody.
110 measuring the fluorescence in cells permeabilized with 1% digitonin in the presence of 1 mM CaCI 2 or 1 mM EGTA. A macro progla,;, was used to calculate the 340/380 nm intensity ratio and [Ca `'+ ]i simultaneously during the fluorescence measurement and the K d value used was 225 riM. Fluorescence from control samples not loaded with fura-2 was subtracted before calculating 340/380 intensity ratio. While calculating [Ca 2 + ]i we have not applied any correction for the fluorescence intensity associated with membrane-bound fura-2. Changes in [Ca -'+ ]i induced by antibodies were determined by measuring the time course of fluorescence intensities at 340 and 380 nm. During the measurement cell suspension was continuously stirred in cuvette with a magnetic stirrer. Antibodies and other reagents were added through an injection port on the lid of the sample compartment of fluorometer. All fluorescence measurements were performed at room temperature (25°0.
7.0 G.0
815
555
•.-- 5-0 0 al)
357 £ ,,
