Exogenous Administration of Gangliosides Inhibits FcεRI-Mediated Mast Cell Degranulation by Decreasing the Activity of Phospholipase Cγ

Cells, reagents, and Abs

RBL-2H3 cells, were cultured as described (10). The cells were harvested, resuspended at a concentration of 1.5 × 10⁶/ml in culture medium with 1% FCS (G-medium), supplemented with or without 50 μM gangliosides, and incubated for 20 h as monolayers. The gangliosides were isolated from human brains with chloroform-methanol-water extraction and partitioned into upper water-methanol phase. Mild alkaline catalyzed methanolysis for destruction of residual phospholipids, dialysis, lyophilization, and final silicic-acid chromatography were performed as described (11). The content of sialic acid in lyophilized samples of brain gangliosides after final chromatography was ∼29%, as determined by resorcinol-HCl reagent. Calculations of molarity of brain gangliosides-containing solutions were based on average molecular weights of major gangliosides present in the isolates (GT1b, GD1b, GD1a, and GM1). Isolated GT1b, GD1a, GM1, and asialo-GM1 were obtained from Sigma-Aldrich (St. Louis, MO). The origins of mAb specific for FcεRI α-subunit (5.14), Thy1.1 (MRCOX7), Syk kinase, LAT, FcεRI β subunit (JRK), as well as 2,4,6-trinitrophenyl (TNP)-specific IgE mAb (IGEL b4 1) and polyclonal Abs for IgE and Syk have been described (12). Polyclonal Abs to PLCγ1, PLCγ2, Akt1, phospho-Akt1 (specific for phosphorylated Ser⁴⁷³), ezrin, and HRP-conjugated donkey anti-goat IgG were obtained from Santa Cruz Biotechnology (Santa Cruz, CA). HRP-conjugated goat anti-mouse IgG, goat anti-rabbit IgG, and anti-phosphotyrosine (PY20) were purchased from Transduction Laboratories (Lexington, KY). Rabbit anti-PI3K p85 subunit (a mixture of equal amounts of antisera against the intact p85 subunit and the N-SH2 region of PI3K) was obtained from Upstate Biotechnology (Lake Placid, NY). Thy1.1 mAb was biotinylated using ImmunoPure NHS-LC-biotin (Pierce, Rockford, IL) according to the manufacturer’s instructions.

Cell activation, immunoprecipitation, immunobloting, and sucrose gradients

Gangliosides-treated and control cells were harvested, resuspended in G-medium at a concentration 10 × 10⁶ cells/ml, and sensitized in suspension with IgE (IGEL b4 1; ascites diluted 1:1000) or biotinylated OX7 mAb (3 μg/ml). After 30 min at 37°C in 5% CO₂ in air, the cells were washed in BSS (20 mM HEPES (pH 7.4), 135 mM NaCl, 5 mM KCl, 1.8 mM CaCl₂, 1 mM MgCl₂, and 5.6 mM glucose) supplemented with 0.1% BSA (BSS/BSA), and activated at 37°C by an exposure to TNP-BSA or streptavidin. In cells activated via FcεRI dimers, the sensitization step was omitted and the cells were directly activated by 5.14 mAb. Toward the end of the activation period the cells were cooled, briefly centrifuged, and β-glucuronidase released into supernatant was determined as described (12). The cell pellets were lysed in an ice-cold lysis buffer containing 50 mM Tris-HCl (pH 7.4), 150 mM NaCl, 2 mM EDTA, 10 mM β-glycerophosphate, 1 mM Na₃VO₄, 1 mM PMSF, 1 μg/ml aprotinin, 1 μg/ml leupeptin, and supplemented with 1% Nonidet-P40 (NP40) and 1% n-dodecyl β-d-maltoside (LAT, Syk, and PLCγ immunoprecipitation (IP)) or 0.2% Triton X-100 (FcεRI IP). Postnuclear supernatants were IP with corresponding Abs prebound to UltraLink-immobilized protein A (Pierce), size-fractionated by SDS-PAGE, and immunoblotted with the PY-20-HRP and other Abs as described (12). Quantitative analysis of protein tyrosine phosphorylation was always corrected for the amount of proteins immunoprecipitated as determined by densitometry of immunoblots after stripping and development with the corresponding Abs.

In experiments analyzing the subcellular distribution of PLCγ2, ezrin, and Akt, the cells were activated and then resuspended in ice-cold PBS supplemented with 0.1% saponin, 5 mM MgCl₂, and 1 mM Na₃VO₄ (permeabilization buffer). After 5 min incubation on ice, the cells were spun down and extracted for 15 min in a lysis buffer containing 1% NP40. Saponin/NP40-extracted material was IP on protein A beads coated with anti-PLCγ2 or protein G beads with anti-ezrin Ab, size fractionated by SDS-PAGE and analyzed by immunoblotting as above. Akt was analyzed by direct immunoblotting of SDS-PAGE fractionated saponin/NP40 extracts.

Isolation of GEMs by sucrose density gradient ultracentrifugation has been described (12).

PIP3 and IP3 determination

PIP3 was detected by a modified previously described procedure (13) adapted for the microcentrifuge tubes. Briefly, IgE-sensitized cells at a concentration of 10⁷/ml in BSS/BSA were incubated with 7.4 MBq/ml [³²P]orthophosphate (Amersham Pharmacia Biotech, Piscataway, NJ) for 1 h at 37°C before stimulation with TNP-BSA. The reaction was stopped by centrifuging the cells for 10 s at 4°C. The pellet was resuspended in ice-cold methanol/chloroform (2:1, v/v, 500 μl), followed by addition of ice-cold chloroform (350 μl) and 1.2 M HCl (350 μl). The tubes were vortexed, centrifuged, and the organic phase (400 μl) was extracted again with 400 μl methanol/0.6 M HCl (1:1, v/v). After centrifugation the organic phase (300 μl) was dried in a vacuum evaporator, resuspended in 25 μl chloroform/methanol (1:1, v/v), and 5 μl aliquots were applied to TLC Silica gel-60 plates (Merck, Darmstadt, Germany) precoated with potassium oxalate (1.2% in 40% methanol). Chromatography was performed for 3 h in a mixture of chloroform/acetone/methanol/acetic acid/water (80:30:26:24:14, all in vol). Radioactive phospholipids were detected by autoradiography and quantified by densitometry. Phosphoinositides of known composition, used as standards, were obtained from Sigma-Aldrich and Matreya (State College, PA) and detected by exposing the plates to iodine vapors.

IP3 was determined as previously described (14). Briefly, 6 × 10⁶ RBL-2H3 cells pretreated or not with gangliosides were sensitized with IgE and stimulated with TNP-BSA in 500 μl BSS/BSA. At various time intervals, the reactions were terminated by adding 100 μl of ice-cold TCA (100%). IP3 was extracted from the samples and quantified by using a commercially available assay kit (PerkinElmer, Boston, MA).

Calcium mobilization

Changes in the concentration of free intracellular Ca²⁺ were determined using 10 μM Fura Red-AM and 10 μM Fluoro-3-AM probes (Molecular Probes, Eugene, OR) as previously described (15). Calcium mobilization was determined using FACSCalibur (BD Biosciences, San Jose, CA) and expressed as the Fluoro-3/Fura Red fluorescence intensity ratios over time.

Filamentous actin (F-actin) assay

The total amount of polymeric actin was measured as previously described (16). Briefly, IgE-sensitized RBL-2H3 cells (10⁶ cells/200 μl BSS/BSA) were stimulated with TNP-BSA (1 μg/ml) for various time intervals. The reaction was terminated by adding 300 μl of PBS containing 50 μg lysophosphatidylcholine (Sigma-Aldrich), 6% formaldehyde, and 0.125 μg of FITC-labeled phalloidin (Sigma-Aldrich). After incubation for 10 min at 37°C, the cells were centrifuged (5 min at 400 × g), resuspended in PBS, and analyzed by flow cytofluorometry using FACSCalibur. The geometric mean fluorescence intensity was determined for each sample and the data points were plotted relative to the mean fluorescence intensity of nonactivated control cells (preincubated with G-medium alone).

Immune complex PI3K and PLCγ assays

PI3K activity was measured as previously described with minor modifications (17). FcεRI-activated or control RBL-2H3 cells (2 ×10⁶) were solubilized in 500 μl lysis buffer (20 mM Tris-HCl (pH 8.0), 137 mM NaCl, 2 mM EDTA, 1 mM Na₃VO₄, 1 mM PMSF, 10 μg/ml aprotinin and leupeptin) supplemented with 1% Triton X-100. PI3K in postnuclear supernatants was immunoprecipitated with anti-PI3K p85 subunit Ab and the immunocomplexes were collected on UltraLink immobilized protein A (10 μl). The beads were washed twice with lysis buffer supplemented with 0.1% Triton X-100, twice with kinase assay buffer (20 mM HEPES (pH 7.4), 20 mM MgCl₂ and 0.25 mM EGTA), and then suspended in 25 μl kinase buffer with or without 100 nM wortmannin. After 10 min at 25°C, PI3K assay was initiated by an addition of 25 μl of kinase buffer containing 10 μg of sonicated phosphatidylinositol (Sigma-Aldrich) and 37 kBq [γ-³²P]ATP (Amersham Pharmacia Biotech; final concentration 10 μM) and the reaction mixture was incubated for 30 min at 25°C; during this period, the formation of phosphatidylinositol phosphate was linear (data not shown). The reaction was terminated by adding methanol/chloroform (2:1, v/v, 250 μl), chloroform (175 μl), and 1.2 M HCl (175 μl). The organic phase was extracted again with 200 μl methanol/0.6 M HCl (1:1, v/v) and lyophilized. Lipids were resuspended in 15 μl of chloroform/methanol (1:1, v/v), spotted on TLC Silica gel-60 plate (Merck), precoated with potassium oxalate, and resolved in chloroform/methanol/4 M ammonium hydroxide (9:7:2, v/v/v) for 1 h. ³²P-labeled materials were visualized by autoradiography and quantified by Fuji Bio-Imaging Analyzer Bas 5000 (Tokyo, Japan).

PLCγ assay was performed essentially as described (18). FcεRI-activated or control RBL-2H3 cells were solubilized in a lysis buffer (20 mM HEPES (pH 7.5), 150 mM NaCl, 5 mM β-glycerophosphate, 0.2 mM Na₃VO₄, 1 mM EGTA, 1 mM PMSF, 1 μg/ml aprotinin, and 1 μg/ml leupeptin) supplemented with 1% Triton X-100. PLCγ1 or PLCγ2 in postnuclear supernatants were immunoprecipitated with the corresponding Abs and the immunocomplexes were collected on UltraLink immobilized protein A. The beads (15 μl) were washed once with lysis buffer supplemented with 0.1% Triton X-100, twice with reaction buffer (35 mM sodium phosphate (pH 6.8), 70 mM KCl, 0.8 mM EGTA, and 0.8 mM CaCl₂) and then suspended in 25 μl reaction buffer. After an addition of 10 μl substrate solution containing 25 mM sodium phosphate (pH 6.8), 50 mM KCl, 2.5% Triton X-100, 6 μg of PIP2, and 1.1 kBq of P[³H]IP2 (PerkinElmer), the beads were incubated for 30 min at 35°C. The reaction was stopped by adding 300 μl of ice-cold 0.5% BSA. The samples were centrifuged, and 300 μl of the supernatant was mixed with 100 μl of ice-cold 25% (w/v) TCA. Precipitates were removed by centrifugation (14,000 × g for 5 min) and the supernatants collected for quantitation of the released [³H]IP3 by liquid scintillation counting. Immunocomplexes bound to the pelleted beads were solubilized in SDS-PAGE sample buffer and analyzed by immunoblotting for the presence of PLCγ.

Statistical analysis

Statistical analysis was performed by two-tail unpaired Student’s t test; p < 0.05 was considered statistically significant.

Exogenously added gangliosides selectively inhibit the FcεRI-mediated secretory response

A preincubation of RBL-2H3 cells with highly purified human brain gangliosides inhibited the β-glucuronidase release induced by Ag-mediated aggregation of IgE-FcεRI complexes (Fig. 1⇓A). Pilot experiments indicated that the maximum inhibition was observed after incubation of the cells for 20 h with gangliosides at a concentration 50 μM in G-medium. Under these conditions, gangliosides did not exhibit any cytotoxic effect as determined by trypan blue dye exclusion test. A similar inhibitory effect was observed when the FcεRI was aggregated by anti-FcεRI mAb 5.14 (Fig. 1⇓B). The inhibitory effect was specific for FcεRI-mediated activation, because it was absent in Thy-1 glycoprotein activated cells, which responded even more than control cells (Fig. 1⇓C). The maximal inhibitory effect was seen at an Ag concentration that gave a maximum secretory response in control cells (1 μg/ml; Fig. 1⇓D). Gangliosides present in the human brain differ in their sialic acid and carbohydrate chain moieties. To determine whether the inhibitory effect is confined to specific gangliosides, we compared the effect of monosialoganglioside GM1, asialo-GM1, disialoganglioside GD1a, and trisialoganglioside GT1b on FcεRI-mediated secretory response. Preliminary experiments showed that isolated brain gangliosides, like the mixed brain gangliosides, exhibited maximum nontoxic inhibitory effects at a concentration 50 μM. Preincubation of the RBL-2H3 cells with GM1 significantly inhibited the secretory response although the inhibition was weaker compared with brain gangliosides (Fig. 1⇓E). Interestingly, asialo-GM1 inhibited the secretory response to a similar extent as GM1, suggesting that sialic acid had no significant role in the inhibitory effect of GM1. Pretreatment with GT1b and GD1a, the dominant brain gangliosides, resulted in a decrease of the secretory response, which was comparable to the inhibition observed after pretreatment with isolated brain gangliosides. Because of these results, isolated brain gangliosides were used in most experiments.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 1.

Exogenously administered gangliosides selectively inhibit FcεRI-mediated secretion. A–D, RBL-2H3 cells were incubated for 20 h with 50 μM isolated brain gangliosides in G-medium (○) or with G-medium alone (•). After incubation, the cells were sensitized for 30 min with TNP-specific IgE (A and D), biotinylated anti-Thy-1.1 mAb OX7 (C), or not sensitized (B). The cells were activated by aggregation of the IgE-FcεRI complexes by TNP-BSA (A; 1 μg/ml), dimerization of FcεRI by 5.14 mAb (B; 1 μg/ml) or aggregation of Thy-1-biotinylated OX7 complexes by streptavidin (Str; 10 μg/ml; C). The amount of β-glucuronidase released into supernatant was determined at various time intervals. In D, IgE-sensitized cells were stimulated with various concentrations of TNP-BSA, and β-glucuronidase was determined after 30 min. In E, the cells were pretreated for 20 h with G-medium supplemented with 50 μM GT1b (⋄), GD1a (⊡), GM1 (□), asialo-GM1 (▵) or in G medium alone (•), and activated as in A. Means ± SD of 4 (A–D) or 2 (E) independent experiments performed in duplicates or triplicates are presented.

Gangliosides-pretreated cells, compared with control cells, exhibited a decreased adhesion to the substratum, with a more rounded morphology and with less developed processes (Fig. 2⇓, A and B). Similar changes in cell morphology were observed in cells pretreated with GT1b, GD1a, GM1, and asialo-GM1 (not shown). These changes were associated with the decreased amount of F-actin observed in nonactivated gangliosides-pretreated cells (Fig. 2⇓C; inhibition 17 ± 7%; mean ± SD; n = 5; p < 0.01). After Ag-mediated activation, the amount of F-actin was significantly higher in both gangliosides-pretreated and control cells. In the former, however, the levels of F-actin remained lower at all time intervals analyzed (Fig. 2⇓C).

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 2.

Morphology and F-actin in ganglioside-pretreated cells. RBL-2H3 cells were incubated with isolated brain gangliosides in G-medium (A) or in G-medium alone (B). Phase contrast images were taken after 20 h. Bar, 10 μm. C, The amount of F-actin in ganglioside-pretreated (○) or control cells (•) was determined by flow cytofluorometry in IgE-sensitized cells activated for various time intervals by TNP-BSA (1 μg/ml). Means ± SD were calculated from five independent experiments.

Gangliosides do not prevent the association of aggregated FcεRI with GEM

When aggregated, the FcεRI rapidly associate with GEM; this was confirmed by sucrose density gradient ultracentrifugation of detergent-lysed cells (19). Because it has been shown that brain gangliosides interfere with the integrity of GEM (7), we further analyzed the solubility of FcεRI in cells exposed to isolated gangliosides. Data presented in Fig. 3⇓A indicate that the majority of nonaggregated FcεRI from both control and gangliosides-pretreated cells solubilized in 0.06% Triton X-100 was located in the high-density fractions of sucrose gradient (>35% sucrose). After aggregation, most of the IgE-FcεRI complexes from control cells were found in low-density fractions (15–35% sucrose). However, a significant amount of FcεRI was still found in the high-density fractions (>55% sucrose); it probably represents the cytoskeleton-bound aggregated FcεRI, which differs from the GEM-associated FcεRI (19). In cells pretreated with gangliosides, most of the aggregated FcεRI was again located in low-density fractions; yet the peak was broader and showed a slight shift to the higher density fractions. In accordance with a previous study (19), no FcεRI was associated with low density fractions in Ag activated cells solubilized with 1% Triton X-100, and gangliosides had no effect on that association (not shown). When surface Thy-1 was labeled with ¹²⁵I-labeled OX7 and the cells were solubilized with 1% Triton X-100, most of the Thy-1 was located in the low-density fractions. In gangliosides-pretreated cell, there was no shift to higher density fractions. In fact, a decreased density of Thy-1 complexes was seen (Fig. 3⇓B). Thus, the observed inhibitory effect of gangliosides on FcεRI-mediated secretion in RBL-2H3 cells does not seem to be a direct consequence of reduced association of aggregated FcεRI with GEM or displacement of GPI-anchored proteins from GEM.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 3.

Aggregation-dependent association of FcεRI with GEMs. RBL-2H3 cells were incubated with isolated brain gangliosides in G-medium (□) or in G-medium alone (○). The cells were sensitized in suspension with ¹²⁵I-labeled IgE (1 μg/ml), and IgE-FcεRI complexes were aggregated (□, ○) or not (▪, •) for 5 min at 37°C with rabbit anti-IgE Ab (10 μg/ml, A). Alternatively, the cells were labeled with ¹²⁵I-labeled OX7 (0.1 μg/ml, B). Unbound Abs were washed out and the cells were solubilized on ice in a lysis buffer containing 0.06% Triton X-100 (A) or 1% Triton X-100 (B). Total cell lysates were diluted 1:1 with 80% sucrose, loaded into sucrose step gradients and fractionated by ultracentrifugation for 4 h. Points show the percentage of total cpm present in individual fractions (left axis) and sucrose concentrations (▵, right axis).

Decreased tyrosine phosphorylation of PLCγ in gangliosides-pretreated cells

In an attempt to determine whether or not exogenously administered gangliosides inhibit the early FcεRI-mediated signaling events, we further assessed the tyrosine phosphorylation of FcεRI β and γ subunits, Syk, LAT, PLCγ1, and PLCγ2 in Ag activated cells. The data presented in Fig. 4⇓A show that pretreatment with gangliosides did not inhibit the tyrosine phosphorylation of FcεRI β and γ subunits, Syk PTK, and LAT. In fact, tyrosine phosphorylation of LAT, which is a typical component of GEM, was significantly increased. Interestingly, although gangliosides failed to inhibit the binding of LAT to PLCγ1 and PLCγ2 (Fig. 4⇓B), tyrosine phosphorylation of these enzymes was significantly reduced (Fig. 4⇓A). As indicated by immunoblotting with specific Abs after stripping, the observed inhibition was not attributable to a decreased amount of immunoprecipitated proteins.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 4.

Tyrosine phosphorylation and interactions of molecules involved in FcεRI-signaling. RBL-2H3 cells were incubated with isolated brain gangliosides in G-medium (+Ga) or in G-medium alone (−Ga), sensitized with IgE and activated (+) or not (−) with TNP-BSA (1 μg/ml). A, Tyrosine phosphorylation of proteins under study was detected after IP with the specific Abs followed by phosphotyrosine immunoblotting (IB) with PY20 mAb. The quantity of proteins in immunoprecipitates was assessed by immunoblotting with appropriate Abs on stripped membranes. The relative amount of tyrosine phosphorylated proteins of ganglioside-pretreated and stimulated cells are given in percent of stimulated controls. ∗, p ≤ 0.001. B, Association of LAT with PLCγ1 and PLCγ2 was determined after IP of LAT followed by immunoblotting with PLCγ1, PLCγ2, and LAT. Similar results were obtained in three independent experiments. Arrows indicate position of the corresponding proteins. Arrows and double-arrow in A, IP:FcεRI, indicate positions of, respectively, FcεRI β and γ subunits.

Decreased activity of PLCγ in gangliosides-pretreated cells

Full activation of PLCγ requires its membrane recruitment by an interaction with LAT and PI3K lipid products (20). The primary in vivo substrate of PI3K is PIP2, which is converted to PIP3. In additional experiments we measured the levels of PIP3 in the course of the activation of gangliosides-pretreated as well as control cells. Data presented in Fig. 5⇓, A and B indicate that the pretreatment with gangliosides decreased the baseline levels of PIP3 (0 min) and that these levels were also significantly lower in Ag-activated cells, compared with cells cultured in media without gangliosides. It should be noted that exogenously added gangliosides also reduced the baseline PIP2 levels (inhibition 21 ± 8%; mean ± SD; n = 4; p < 0.05). After Ag triggering the differences between control and gangliosides-pretreated cells were not significant.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 5.

PIP3 levels and activity of PI3K in gangliosides-pretreated and Ag-stimulated cells. A and B, RBL-2H3 cells were incubated in the presence (+Ga) or absence (−Ga) of isolated brain gangliosides and then sensitized with IgE. PIP3 levels were measured in [³²P]orthophosphate-labeled cells before (0 min) and following stimulation with TNP-BSA (1 μg/ml). The TLC separation pattern was determined using phosphoinositide standards. A representative TLC is shown (A). Quantitative measurement of PIP3 level, corrected for the total signal from phosphatidylinositol monophosphate (PIP1), PIP2, and PIP3, in controls (□) or ganglioside-pretreated cells (▩) was determined by densitometry of films (B). Data are presented as means ± SD, calculated from four independent experiments, and normalized to nonstimulated control cells. C, PI3K enzymatic activity was measured in cells preincubated with G-medium alone (−Ga) or G-medium supplemented with 50 μM brain gangliosides (+Ga), GD1a, or GT1b. The cells were stimulated with Ag for various time intervals and PI3K activity was determined as described in Materials and Methods. Samples labeled 5w and 1w were from cells stimulated with Ag for, respectively, 5 min and 1 min, and the immunocomplexes were exposed for 10 min to 100 nM wortmannin before PI3K assay. Migration of PIP is indicated by arrow. Representative results are shown on top panels. Quantitative measurements of PIP were normalized to unstimulated control cells and to the amount of PI3K detected by immunoblotting of whole cell lysates. Means of fold increase ± SD were calculated from three to five independent experiments.

The observed decrease in PIP3 levels could be the result of a decreased activity of PI3K, an increased activity of the corresponding phosphatases and/or a decreased availability of PI3K substrate. In additional experiments, we therefore evaluated the PI3K activity by immunocomplex kinase assay. PI3K was immunoprecipitated from control or gangliosides-pretreated cells activated for various time intervals with Ag. Data presented in Fig. 5⇑C indicate that control cells incubated for 20 h in the absence of gangliosides exhibited a relatively rapid increase in PI3K activity after exposure to Ag (2.5-fold increase in 15 s) which declined after 5 min. Pretreatment with gangliosides did not inhibit PI3K activity but resulted instead in its increase, especially at early time intervals after the exposure to Ag. A significant increase in PI3K activity was also observed after pretreatment of the cells with GD1a and GT1b gangliosides, the major components of brain gangliosides (Fig. 5⇑C). As expected, pretreatment of PI3K immunocomplexes with 100 nM wortmannin completely inhibited the formation of ³²P-labeled PIP. These data indicate that the observed decrease in PIP3 levels in gangliosides-pretreated cells are not due to a decreased activity of PI3K.

The major function of PLCγ is to hydrolyze PIP2 to IP3 and to diacylglycerol, which, respectively, induces the release of Ca²⁺ from intracellular stores, and activates protein kinase C. Decreased tyrosine phosphorylation of PLCγ1 and PLCγ2 in gangliosides-pretreated cells suggested their reduced enzymatic activity. To prove that the activity of PLCγ is indeed reduced, we measured the PLCγ activity in an immunocomplex assay. In control cells preincubated in G-medium alone, the activities of PLCγ1 and PLCγ2 were increased ∼6-fold and 8-fold, respectively, after 1 min exposure to Ag, and then declined (Fig. 6⇓, A and B). Pretreatment with gangliosides inhibited the activity of both enzymes at all time intervals analyzed. The observed inhibition was not due to decreased levels of the enzymes as could be inferred from the results of immunoblotting analyses of PLCγ1 and PLCγ2 immunoprecipitated from control and gangliosides-pretreated cells (not shown). Pretreatment of the cells with GD1a and GT1b inhibited the Ag-induced (1 min) PLCγ1 activity by, respectively, 54.5 ± 3.5% (mean ± SD, n = 4) and 51.8 ± 2.6% (mean ± SD, n = 3); this confirms that the two major components of brain gangliosides exhibit a similar inhibitory activity as do the isolated brain gangliosides.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 6.

Exogenously administered gangliosides inhibit the enzymatic activity of PLCγ, and Ca²⁺ mobilization in Ag activated cells. RBL-2H3 cells were incubated in the presence (○) or absence (•) of isolated brain gangliosides and then sensitized with IgE. A and B, Enzymatic activities of PLCγ1 and PLCγ2 were measured by immune complex PLC assay as described in Materials and Methods. Cells were collected before or at various time intervals after stimulation with TNP-BSA. Means ± SD were calculated from two to three experiments performed in duplicates. C, IP3 levels were measured before and after stimulation with TNP-BSA. Samples were collected at the indicated times. Means ± SD of three independent experiments performed in duplicates are presented. D–F, Calcium response in Fluo-3- and Fura-Red-loaded cells. The cells were pretreated (dash-dot line) or not (solid line) with gangliosides, sensitized with IgE (D and E) or not (F), and stimulated in the presence (D and F) or absence (E) of extracellular calcium with TNP-BSA (1 μg/ml; D and E, arrow, Ag), or calcium ionophore A23187 (1 μg/ml; F, arrow, A23). Fluorescence intensity was determined by flow cytofluorometry.

Our findings that gangliosides inhibit the activity of PLCγ were supported by measurements of IP3 levels in control and gangliosides-pretreated cells after stimulation with Ag (Fig. 6⇑C). In control cells, Ag stimulation induced a rapid augmentation of IP3 concentrations, with a peak at 15 s, followed by a slow return to baseline levels. In gangliosides-pretreated cells, the baseline level of IP3 was comparable to that in control cells at time 0 min, but was significantly reduced after stimulation with Ag (Fig. 6⇑C). The reduced IP3 levels in gangliosides-pretreated and activated cells were reflected in a decreased calcium response after activation in media supplemented with Ca²⁺ (Fig. 6⇑D), as well as in the absence of Ca²⁺ (Fig. 6⇑E). Exogenously added gangliosides had no effect on Ca²⁺ response induced by calcium ionophore A23187 (Fig. 6⇑F), suggesting that the signal transduction pathways involving PLCγ, rather than Ca²⁺ channels, are modulated by gangliosides.

Formation of signaling assemblies in gangliosides-pretreated cells

Previously, we described a simple method for determination of the subcellular translocations of signal transduction molecules, one which allows an analysis of posttranslational modifications during cell activation (21). The method is based on a gentle release of free cytoplasmic components after permeabilization of the cells with the cholesterol sequestering reagent saponin, followed by complete solubilization of all membrane components, including lipid rafts, by the detergent NP40. Using a modified version of this method we first compared the subcellular distribution of PLCγ2 and its phosphorylated form, in both control and gangliosides-pretreated cells. The data presented in Fig. 7⇓A indicate that FcεRI aggregation in saponin permeabilized cells is followed by a rapid and transient increase in the amount of tyrosine phosphorylated PLCγ2 (arrow) and associated proteins. The observed change was in part due to an increased amount of insoluble PLCγ2 remaining in permeabilized cells, as demonstrated by immunoblotting with anti-PLCγ2 Ab. Pretreatment with gangliosides of saponin permeabilized cells reduced the amount of tyrosine phosphorylated PLCγ2 and coimmunoprecipitated proteins. This effect was in part related to a decrease in the amount of insoluble PLCγ2 and was most profound 5 min after Ag triggering.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 7.

Decreased formation of PLCγ2-, Ezrin- and Akt-containing signaling assemblies in gangliosides-pretreated cells. RBL-2H3 cells were incubated with isolated brain gangliosides in G-medium (+Ga) or in G-medium alone (−Ga), sensitized with IgE, and activated with TNP-BSA (1 μg/ml). After the indicated time intervals, the cells were spun down and resuspended in permeabilization buffer containing 0.1% saponin, followed by centrifugation and extraction of the “empty” cells with NP40. The insoluble material (nuclei and cytoskeleton) was removed by centrifugation and the supernatant immunoprecipitated with anti-PLCγ2-coated protein A beads (A) or anti-ezrin-coated protein G beads (B) before SDS-PAGE and immunoblotting with anti-phosphotyrosine Ab (PY20). Alternatively, the supernatant was directly analyzed by SDS-PAGE and immunoblotting with anti-phospho-Akt (pAkt) Ab (C). After stripping, the membranes were tested for the presence of PLCγ2, ezrin, and Akt with the corresponding Ab. Arrows in A, B, and C indicate, respectively, positions of PLCγ2, ezrin, and Akt.

Because of the observed changes in cellular morphology and the decreased amount of F-actin in gangliosides-pretreated cells, we also analyzed the tyrosine phosphorylation of ezrin, which functions as an intermediate between the actin cytoskeleton and integral plasma membrane proteins (22). Although the amount of ezrin bound to permeabilized control cells (−Ga) did not show any significant change in the course of cell activation, its tyrosine phosphorylation was dramatically increased 30 s after activation, and then declined.

Pretreatment with gangliosides decreased the amount of ezrin in saponin permeabilized cells. Interestingly, the background level of tyrosine phosphorylated ezrin (0 min) was higher in gangliosides-pretreated cells than in control cells, and FcεRI aggregation resulted in a decreased tyrosine phosphorylation of ezrin in gangliosides-pretreated cells (Fig. 7⇑B).

Finally, we investigated the distribution of Akt, which is recruited to the plasma membrane in activated cells by virtue of its interaction with PIP3 and phosphatidylinositol 3,4-bisphosphate (23). In control cells (−Ga) the amount of tyrosine phosphorylated Akt in saponin permeabilized cells increased during FcεRI activation (Fig. 7⇑C). However, this increase was due not only to an augmentation of Akt tyrosine phosphorylation, but was mainly attributable to an increased association of Akt with insoluble material in saponin permeabilized cells, as indicated by anti-Akt immunoblotting. In gangliosides-pretreated cells, both Akt tyrosine phosphorylation and its association with insoluble material were reduced.