PMC

The inhibitory 7A6 and 12A10 mAbs recognize rat CD63. (A) RBL cell lysates were immunoprecipitated with 12A10, an anti-rat CD63 antibody (AD1), or isotype control (IgG1), and immunoblotted with either AD1 or 12A10 mAb. (B, top) Lysates from WT U937 cells and from U937 cells stably transfected with rat CD63 were immunoprecipitated with 12A10 or isotype control and immunoblotted with 12A10 mAb. (B, bottom) WT U937 cells and U937 cells stably transfected with rat CD63 were incubated with 7A6 or control isotype followed by FITC–anti-mouse and analyzed by flow cytometry.

Anti-CD63 suppresses degranulation in adherent cells, but not in cells in suspension. The capacity of anti-CD63 (12A10 mAb) to inhibit FcɛRI-induced degranulation was compared in RBL cells that were stimulated either in suspension or adherent (mean ± SEM of six experiments with duplicate samples). * P = 0.0051. Cells were loaded with [³H]serotonin and anti-DNP–IgE and triggered, or not, with 50 ng/ml DNP-HSA for 20 min.

Schematic representation of our model for the mechanism of anti-CD63 inhibition. The Lyn–Ca²⁺-dependent signaling pathway for degranulation is shown in gray. The Ca²⁺-independent PI3K–Gab2–PKCδ signaling pathway shared with the integrin–tetraspanin complex is shown in black and yellow. Mechanism 1 (steric hindrance of integrin binding to their ECM substrates) is represented in green. Mechanism 2 (general disruption of the membrane complexes containing CD63) is represented in blue. Mechanism 3 (sequestration of PI4K) is represented in red.

Anti-CD63 suppresses passive cutaneous anaphylaxis reactions in rats. Rats were sensitized intradermally with 10 or 25 ng anti-DNP IgE along with 50 μg 12A10 or control IgG1 in duplicate sites. 24 h later PCA was induced by i.v. injection of DNP-HSA along with Evan's Blue. The skin sites were excised, Evan's Blue was extracted, and quantified by absorbance at 610 nm. The results are expressed as mean ± SEM of μg Evan's Blue (four duplicate experiments).

mAb 12A10 suppresses FcɛRI-induced serotonin release, but does not inhibit LTC4 production. (A) Serotonin release from RBL-2H3 cells loaded with 0.2 μg/ml anti-DNP IgE and stimulated with DNP-HSA (DNP; 50 ng/ml) to induce FcɛRI cross-linking or with vehicle (nil) for 30 min. Results are expressed as percentage of total serotonin uptake (mean ± SEM of five experiments in duplicate). The two columns on the left are for cells without mAb preincubation. The four columns on the right are for cells preincubated with IgG1 control or 12A10 mAb (2 μg/ml) for 30 min before FcɛRI triggering. (B) Dose response of inhibition of FcɛRI-induced serotonin release by 12A10 or IgG1 control. Cells were triggered as in A. Results are expressed as mean ± SEM of four experiments with duplicate samples. ** P = 0.0117. (C) LTC4 production induced by FcɛRI aggregation (50 ng/ml DNP-HSA for 20 min) in RBL-2H3 cells preincubated with 10 μg/ml of IgG1 or 12A10 (mean± SEM of two experiments in triplicate; P = 0.7532).

Anti-CD63 leaves FcɛRI-induced global tyrosine phosphorylation and Ca^{2 +} mobilization intact. (A) Whole cell lysates of unstimulated (0 min) and FcɛRI-triggered (DNP-HSA) RBL-2H3 cells preincubated with 12A10 (+) or vehicle (−) were blotted with the anti- phosphotyrosine antibody 4G10. Cells were stimulated under adherent conditions. The data are representative of four experiments. (B) RBL-2H3 cell were preincubated with 10 μg/ml of 12A10 or control IgG1 and loaded with fura 2-AM. Ca²⁺ mobilization induced by FcɛRI triggering (50 ng/ml DNP-HSA) was measured in a spectrofluorometer and expressed as the ratio of emission with excitation at 340 and 380 nm. Shown are average traces from 11 for IgG1 (black lines) and 10 experiments for anti-CD63 (gray lines). Ratios (R) at each time point after addition of DNP-HSA were normalized to the baseline ratio during the first 60 s of recording (R₀).

Anti-CD63 potently suppresses both degranulation and adhesion on fibronectin and vitronectin, but not fibrinogen. (A) The capacity of anti-CD63 to inhibit adhesion to fibronectin, vitronectin, and fibrinogen was compared in RBL cells preincubated with control IgG1 or anti-CD63 (12A10) and plated for 30 min at 37°C on 96-well plates coated with fibronectin, vitronectin, or fibrinogen. Adhesion was quantified by crystal violet uptake measured by spectrophotometry at 570 nm in cell lysates and is expressed in arbitrary units after subtracting the value for adhesion to uncoated wells. Results are expressed as mean ± SEM of seven samples (fibronectin) or six samples (vitronectin and fibrinogen). *P < 0.05 of anti-CD63–treated cells versus IgG1-treated cells. Antibodies directed against the α5 and β1 chain of β1 integrins or against β3 integrins were used as positive controls. (B) The capacity of anti-CD63 (12A10 mAb at 10 μg/ml) to inhibit FcɛRI-induced serotonin release was compared in anti-DNP IgE-loaded RBL cells that were grown for 16 h in the absence of FCS on uncoated wells or surfaces coated with fibronectin, vitronectin, or fibrinogen. All wells were blocked with 2% BSA solution before adding cells. Shown are mean percentage of serotonin release ± SEM from 12 samples (fibronectin), 10 samples (vitronectin and fibrinogen), and 8 samples (uncoated). * P < 0.05.

Anti-CD63 inhibits Gab2-dependent signal transduction, which results in diminished activity of the Ca²⁺-independent PKCδ isoform in adherent cells. (A) Adherent RBL-2H3 cells were preincubated with 12A10 or left untreated. They were then either left unstimulated (nil) or stimulated by FcɛRI cross-linking (DNP). Cells were then lysed and Gab2 was immunoprecipitated. Immunoprecipitates were blotted with anti-phosphotyrosine mAb 4G10 (left) or with anti-Gab2 (right). These data are representative of three experiments. (B) Adherent RBL-2H3 cells were stimulated as described for A for 2 min. After lysis, plasma membrane fractions were separated from cytosolic fractions and subjected to immunoblotting with anti-Akt, anti-Gab2, and anti-Syk antibodies. These data are representative of three experiments. (C) Adherent RBL-2H3 cells were stimulated as described for A. Whole cell lysates were immunoblotted with anti-phospho–Akt (left) and anti-Akt (right) to confirm that equal amounts were loaded. These data are representative of three experiments. (D) RBL-2H3 cells, either adherent (top) or in suspension (bottom) were preincubated with 12A10, and either left unstimulated (nil) or stimulated by FcɛRI cross-linking for 3 min (DNP). Cells were then lysed and PKCδ was immunoprecipitated. Immunoprecipitates were subjected to in vitro kinase assay using MBP as substrate (left). The immunoprecipitates were blotted with anti-PKCδ to confirm that equal amounts of PKCδ were immunoprecipitated (middle). A densitometric analysis of the MBP band in in vitro kinase assays was performed (right) and the results are expressed as mean ± SEM of percent increase after FcɛRI aggregation compared with baseline activity in six (adherent) or three (suspension) experiments. *P = 0.0277 of anti-CD63–treated cells versus vehicle-treated cells.