| Glycobiology | Pages |
Anti-GM3 antibodies activate calcium inflow and inhibit platelet-derived growth factor beta receptors (PDGF[beta]r) in T51B rat liver epithelial cells
Introduction
Results
Discussion
Materials and methods
Acknowledgments
References
Anti-GM3 antibodies activate calcium inflow and inhibit platelet-derived growth factor beta receptors (PDGF[beta]r) in T51B rat liver epithelial cells
Introduction
Glycolipids, gangliosides in particular, modulate cellular functions including cell growth (Bremer et al., 1986; Dohi et al., 1988) (Bremer et al., 1986; Dohi et al., 1988) (Bremer et al., 1986; Dohi et al., 1988) and differentiation (Nagai and Tsuji, 1994). Anti-glycolipid antibodies are involved in autoimmune diseases (Gleeson, 1994) as well as anti-tumor immunity (Dyatlovitskaya and Bergelson, 1987) and have been suggested as vehicles for anti-tumor therapy (Reisfeld et al., 1994; Fish, 1996; Hakomori and Zhang, 1997). We are only beginning to understand the function of gangliosides at the cellular level. Exogenous gangliosides (Bremer et al., 1986) and ganglioside-specific antibodies (Dohi et al., 1988) exert similar inhibitory actions on cell growth. The mechanisms of ganglioside action in some cases involve growth factor receptors, including EGF (Bremer et al., 1986) and PDGF (Saqr et al., 1995) receptors. Inhibition of growth factor receptors by gangliosides results in an inhibition of downstream signaling cascades including growth factor-induced calcium fluxes (Guan et al., 1992). Exogenous gangliosides raise intracellular calcium (Isasi et al., 1995; Ortaldo et al., 1996; Yatomi et al., 1996) or increase stimulus-evoked calcium fluxes (Hilbush and Levine, 1992; Tanaka et al., 1997) in a variety of cell types. Ganglioside-binding proteins have similar effects (Dixon et al., 1987; Ortaldo et al., 1996). Gangliosides and ganglioside-binding proteins can act synergistically (Ortaldo et al., 1996) or have opposite effects (Fan et al., 1994). It is important to understand these processes because calcium fluxes control many cellular functions including muscle contraction, hormone release, learning and memory, and cell proliferation, to name a few. Activation of calcium inflow by ganglioside-binding proteins is best characterized in neurons, where the cholera toxin B subunit (CTB) activates dihydropyridine-sensitive channels through interaction with GM1 (Dixon et al., 1987; Carlson et al., 1994; Wu et al., 1996). CTB stimulates calcium inflow but does not trigger release from internal stores (Carlson et al., 1994). In other cases gangliosides stimulate production of inositol triphosphate (IP3) that will result in both inflow and release of calcium from internal stores (Ortaldo et al., 1996). The existing literature demonstrates ganglioside-specific signaling in the immune and nervous systems (Carlson et al., 1994; Wu and Ledeen, 1994; Isasi et al., 1995; Ortaldo et al., 1996). We need to understand how cell and tissue-specific the described signaling pathways are, and define model systems to study these signaling pathways at the molecular level.
The aim of the present study is to characterize the actions of anti-glycolipid antibodies in T51B liver epithelial cells, with regard to calcium fluxes and growth factor receptor function. Intracellular calcium increases trigger exocytosis in these cells (Dean and Boynton, 1995). Therefore, ganglioside-mediated calcium fluxes may modulate the secretory function of the liver.
Results
Characterization of cell surface glycoconjugates
Cell surface glycoconjugates expressed in T51B cells were identified by flow cytometry as shown in Figure
Figure 1. Estimation of antibody binding by flow cytometry. In all panels the left peak was obtained from cells treated with the secondary antibody only. The peak to the right was obtained with primary plus secondary antibody. Shifts observed with 1B2, TE13, DH2, Mab 126, and 3C9 demonstrate the presence of type 2 and type 1 chain lacto series, GM3, GD2, and T antigen, respectively. IE3 (anti-Tn) does not bind to T51B cells. Anti-glycoconjugate antibodies raise intracellular calcium.
Several antibodies with different glycoconjugate specificity were tested (Figure
Figure 2. Anti-glycolipid antibodies increase intracellular calcium. Left, TE13 fails to elicit a calcium increase (dilution 1:20 in HEPES buffer containing 2 mM Ca, hatched bar). Subsequent application of 1B2 (dilution 1:20 in the same buffer, solid bar) triggers an intracellular calcium increase. Data is the average of 10 cells in one field. Right, Calcium level expressed as percentage of baseline, measured 2.5 min after the start of antibody application. The 1B2 antibody rapidly increases calcium to nearly four times the baseline level, while DH2 nearly doubles this value. TE13, Mab 126, and 3C9 have little effect. Epidermal growth factor (EGF, 25 ng/ml) elicits calcium signals similar in magnitude to those of 1B2. Data is the average + SEM of 3-7 independent experiments, with each experimental value the average of 10 cells. Stimulation of calcium inflow
Application of 1B2 antibody to T51B cells results in a rapid intracellular calcium increase. To determine if this increase is due to inflow or to release from internal stores, the extracellular buffer was replaced with a low calcium (25 µM) buffer (Figure
Figure 3. 1B2 stimulates calcium inflow and does not trigger release from internal stores. (A) Intracellular calcium increases upon application of the 1B2 antibody (1:20 in HEPES buffer containing 2 mM Ca, solid bar). After calcium reaches a peak the extracellular calcium is reduced to 25 µM (open application bar), and intracellular calcium returns to baseline. Intracellular calcium increases again when the cells are returned to a buffer containing 2 mM Ca. The calcium signal is the average of 10 cells in one field. (B) In the presence of 25 µM Ca (open application bar) 1B2 (1:20, solid bar) fails to elevate intracellular calcium. Subsequent application of platelet-derived growth factor (PDGF, 10 ng/ml, cross hatch) induces a transient calcium increase due to release from internal stores. (C) In the same conditions as (B), DH2 (1:20, rising hatch) also fails to elevate intracellular calcium. Subsequent PDGF application (20 ng/ml, cross hatch) fails to elevate intracellular calcium. Each set of data is representative of 2-4 experiments, with each experiment shown the average of 10-25 cells.
Two anti-glycosphingolipid antibodies tested raise intracellular calcium. The 1B2 antibody produces the larger Ca responses. However, its actions will require further characterization because this antibody recognizes structures present on both glycolipids and glycoproteins. Carbohydrate-binding proteins can activate receptor tyrosine kinases (Zeng et al., 1995) as well as G-protein-coupled receptors (Golard, 1995; Matsuo et al., 1996). 1B2 does not activate the EGF receptor (Figure
Figure 4. None of the antibodies tested activate EGF receptors. Cells were treated with antibodies (1:20) or EGF (25 ng/ml) for 10 min. Whole cell lysates were obtained, separated by electrophoresis, transferred to PVDF membranes and assayed with an anti-phosphotyrosine antibody as described in Materials and methods. Data is from two separate experiments. Error! No index entries found.
Application of exogenous GM3 raises intracellular calcium in platelets, mostly through calcium release from internal stores (Yatomi et al., 1996). The anti-GM3 antibody DH2 does not trigger calcium release from internal stores. Rather, the DH2 and 1B2 act in a manner similar to cholera B subunit binding to GM1, in that they trigger calcium inflow but not calcium release from internal stores (Carlson et al., 1994). Exogenous gangliosides may act through interaction with endogenous glycolipids (Kojima and Hakomori, 1991) or gangliophilic proteins (Hattori et al., 1995). Exogenous ganglioside-binding proteins (antibodies, toxins, and lectins as well as myelin-associated glycoprotein (Yang et al., 1996)) will mimic the glycolipid-glycolipid interaction. The resulting reorganization of the membrane gangliosides may trigger signaling cascades in at least two ways. First, reorganization of ganglioside microdomains may activate membrane-associated enzymes such as phospholipase C (Yamamura et al., 1997). Second, the interaction between endogenous gangliosides and extracellular factors may disrupt specific existing ganglioside interaction with membrane proteins. Disrupting these interactions may interfere with the ability of the proteins to form dimers and other complexes. Association of PDGFr with a proteoglycan is required for optimal response to PDGF (Nishiyama et al., 1996). Gangliosides also specifically associate with membrane proteins (Cheresh et al., 1987), so disruption of these interactions may also affect the basal activity of the proteins involved. The ganglioside specificity of the signaling cascades may give cues to the mechanism. The ability of gangliosides to trigger calcium release from internal stores is not limited to a particular species of ganglioside (GM1, GM2, and GM3 all trigger calcium release in a single cell type; Yatomi et al., 1996). These effects may be due to reorganization of glycolipid microdomains. On the other hand, more specific actions (Bremer et al., 1986; Carlson et al., 1994; ) may be due to ganglioside association with the channel or growth factor receptor.
While many gangliosides interfere with growth factor receptor dimerization, exogenously applied GM3 inhibits PDGFr signaling downstream of receptor dimerization and phosphorylation (Bremer et al., 1984; Yates et al., 1993; Saqr et al., 1995; Sachinidis et al., 1996). Our data shows that anti-GM3 antibodies also inhibit PDGFr signaling. In the relatively short time following antibody application (2.5 min), the ganglioside-antibody complex is likely to remain in the plasma membrane. This suggests that GM3 inhibition of PDGFr signaling occurs at the level of the plasma membrane, or that antibody binding to GM3 microdomains (Yamamura et al., 1997) can rapidly turn on intracellular signaling cascades capable of blocking PDGFr signaling.
From the present results, it is apparent that GM3 can regulate calcium fluxes in epithelial cells in a manner similar to GM1 in neurons (Dixon et al., 1987; Carlson et al., 1994; Wu et al., 1996). Calcium inflow leads to exocytosis in many cell types (Kirillova et al., 1993; Tse et al., 1993; Lopez et al., 1994). It is required for exocytosis in T51B liver epithelial cells (Dean and Boynton, 1995). Therefore, gangliosides may modulate the secretory function of the liver. Understanding these processes in epithelial cells is especially important because these cells are exposed to a range of stimuli through interaction with soluble factors as well as circulating cells. We are now in the process of directly measuring exocytosis from these cells.
In addition, we have established a system with which the inhibition of PDGF[beta]r signaling by anti-GM3 antibodies can be studied with PDGF[beta]r clones. We have expressed several PDGFr mutants in T51B cells and can now assay which PDGFr signals are modulated by GM3.
Ganglioside control of calcium inflow and growth factor receptor function appear to be general phenomena present in neuronal, immune and epithelial cells alike. Specific events will depend on the proteins expressed in a given cell type. This may create a variety of tissue-specific actions from a limited set of gangliosides and binding partners. Cell culture
T51B cells were cloned from T51 cells established in Dr. E. Farber's laboratory from an adult rat liver superfusion culture. They are non-neoplastic epithelial cells and contain EGFr but not PDGF[alpha] or [beta] receptors. T51B cells were grown on glass coverslips in Basal Medium Eagle (BME) supplemented with 10% fetal calf serum. The plating density was 7000 cell cm-2, and the cells were used 6-8 days after plating. Under these conditions the cells are confluent and quiescent. One set of experiments was performed with T51B cells infected with a retrovirus carrying the PDGF[beta]r. Application of test solutions
Hybridoma cell culture supernatants were diluted 1:20 in a recording solution containing (in mM): NaCl 140, KCl 5, CaCl2 2, MgCl2 1, Na2HPO4 2.6, HEPES 10, pH 7.4 with NaOH. In experiments with low extracellular calcium the buffer was: NaCl 144, KCl 5, Na2HPO4 2.8, CaCl2 0.025, MgCl2 1, HEPES 10, pH 7.4 with NaOH. Antibodies were applied to the cells through a gravity-fed local superfusion system, allowing full solution exchange within 1-2 s. Calcium imaging
The cells were incubated 30 min at room temperature in the recording buffer containing 5 µM Fura 2 am (Molecular Probes). The dish was rinsed once, and the coverslip was placed in the recording chamber. Recordings were done on a Zeiss Axiovert 135 microscope equipped with a 63× oil immersion lens (n.a. 1.25) and an Attofluor enhanced CCD system. At the end of the recording, the ionophore 4-bromo-A23187 (1 µM) was applied, and fluorescence measured in normal buffer (2 mM Ca) and in a solution containing 0 ca and 0.5 mM EGTA. This gives values for R(Lo), R(Hi), Den(Lo), and Den(Hi), used to calibrate the calcium readings. Flow cytometry
Cells were collected from confluent cultures by incubation in PBS containing 0.2% EDTA. The cells were washed twice with phosphate-buffered saline (PBS) and aliquoted. The washed cells were incubated with the antibodies for 45 min at 4°C followed by three additional PBS washes. The cells were then stained with a FITC-labeled goat anti-mouse antibody (BioSource) at 1:40 dilution for 1 h at 4°C, followed by three PBS washes. The cells were fixed in PBS containing 1% paraformaldehyde for 30 min at room temperature, rinsed, and analyzed on a FACScan flow cytometer (Becton & Dickinson). Western blotting
Cell cultures were lysed for 30 min in cold buffer containing: HEPES buffered saline (HBS) containing 10% glycerol, 1% Triton X-100, 0.5% deoxycholate, 1 mM Na orthovanadate, 10 mm NaF, and 1 mM phenylmethylsulfonyl fluoride. The cells were centrifuged at 14,000 × g for 10 min and the pellets boiled with SDS-PAGE loading buffer containing 2% [beta]-mercaptoethanol. Lysates were resolved on SDS-PAGE and transferred to polyvinyl difluoride (PVDF) membranes. The membranes were then incubated for 1 to 2 h at room temperature with a monoclonal anti-phosphotyrosine antibody. This was followed by horse radish peroxidase-conjugated secondary antibodies, and processed with an enhanced chemiluminescence kit (Pierce).
Thanks are due to Dr. Alton Boynton for continuous support; Dr. Don Messner for performing one of the Western blots; Dr. Eric Holmes for helpful discussion and for providing the 1B2, DH2, Mab-126, and TE13 antibodies; Dr. Henrik Clausen for providing the 3C9 and IE3 antibodies; Drs. Anne Sherwood and Mark Stroud for reading the manuscript; and Peng Ao, Robert Barren, Thomas Greene, and Ajit Jagdale for technical assistance.
Discussion
Materials and methods
Acknowledgments
References
This page is run by Oxford University Press, Great Clarendon Street, Oxford OX2 6DP, as part of the OUP Journals
Comments and feedback: www-admin{at}oup.co.uk
Last modification: 19 Dec 1998
Copyright©Oxford University Press, 1998.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  E. A. Miljan and E. G. Bremer Regulation of Growth Factor Receptors by Gangliosides Sci. Signal., November 26, 2002; 2002(160): re15 - re15. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||