Differential regulation of the platelet GPIb-IX complex by anti-GPIbβ antibodies

Abstract

Background: Platelets' initial recognition of endothelial damage proceeds through the interaction between collagen, plasma von Willebrand factor (VWF), and the platelet glycoprotein (GP)Ib-IX complex (CD42). The GPIb-IX complex consists of one GPIbα, one GPIX, and two GPIbβ subunits. Once platelets are immobilized to the subendothelial matrix, shear generated by blood flow unfolds a membrane-proximal mechanosensory domain (MSD) in GPIbα, exposing a conserved trigger sequence and activating the receptor. Currently, GPIbα appears to solely facilitate ligand-induced activation because it contains both the MSD and the binding sites for all known ligands to GPIb-IX. Despite being positioned directly adjacent to the MSD, the roles of GPIbβ and GPIX in signal transduction remain murky.

Objectives: To characterize a novel rat monoclonal antibody 3G6 that binds GPIbβ.

Methods: Effects of 3G6 on activation of GPIb-IX are characterized in platelets and Chinese hamster ovary cells expressing GPIb-IX (CHO-Ib-IX) and compared with those of an inhibitory anti-GPIbβ antibody, RAM.1.

Results: Both RAM.1 and 3G6 bind to purified GPIbβ and GPIb-IX with high affinity. 3G6 potentiates GPIb-IX-associated filopodia formation in platelets or CHO-Ib-IX when they adhere VWF or antibodies against the ligand-binding domain (LBD) of GPIbα. Pretreatment with 3G6 also increased anti-LBD antibody-induced GPIb-IX activation. Conversely, RAM.1 inhibits nearly all GPIb-IX-related signaling in platelets and CHO-Ib-IX cells.

Conclusions: These data represent the first report of a positive modulator of GPIb-IX activation. The divergent modulatory effects of 3G6 and RAM.1, both targeting GPIbβ, strongly suggest that changes in the conformation of GPIbβ underlie outside-in activation via GPIb-IX.

Keywords: biomechanics; filopodia; glycoprotein Ib-IX complex; microscopy; platelet activation.

Figure 1.. RAM.1 and 3G6 bind to GPIbβ with high affinity.

(A) Western blot analysis detecting purified GPIbβ extracellular domain (IbβE) and GPIb-IX complex with 3G6 (top) or RAM.1 (bottom) under reducing (R) or nonreducing (NR) conditions. In each gel, molecular weight ladders are marked on the left, and the protein band are marked on the right. (B) Quantitative ELISA measurements of 3G6 (top) or RAM.1 (bottom) binding to the immobilized extracellular domain of GPIbβ (IbβE, red) or the full GPIb-IX complex (blue). Apparent K_d for each interaction is listed on the respective graph.

Figure 2.. 6B4 and 6B4 Fab, but not AK2, activate CHO-Ib-IX cells.

(A) Representative confocal microscopy images of CHO-Ib-IX cells adhering to immobilized AK2, 6B4, or 6B4 Fab. Cells are stained with TRITC-Phalloidin. Each image is a max intensity projection of several z-stacks spanning the height of the cells. Scale bar = 10 μm. (B) Quantitation of mean length of filopodia, # of filopodia per cell, and cell area. Filopodia were defined as protrusions ≥ 1.5 μm in length. Each point represents the mean value of ≥30 cells from an independent experiment. All values obtained via analysis by a uniform ImageJ macro. Error bars represent mean ± SD. Significance determined by one-way ANOVA with post-hoc Tukey. p≤0.0001; ****.

Figure 3.. 3G6 and RAM.1 differentially modulate CHO-Ib-IX activation by anti-LBD MAbs and VWF.

(A) Mean length of filopodia extended by CHO-Ib-IX cells on AK2, 6B4, 6B4 Fab, or VWF + botrocetin (Bc), ±3G6/RAM.1 in solution. (B) Mean # of filopodia (≥ 1.5 μm) per cell for each condition in (A). (C) Quantitation of cell area for each condition in (A). For all graphs, each point represents the mean value of ≥30 cells from an experimental replicate. Error bars are mean ± SEM. Significance determined by one-way ANOVA with post-hoc Tukey. p≤0.05; *, p≤0.01; **, p≤0.001; ***, p≤0.0001; ****.

Figure 4.. 3G6 and RAM.1 differentially modulate platelet activation by anti-LBD MAbs or VWF.

(A-B) Representative images of human platelets adhering to 6B4/AK2 (A) or VWF (B) with or without RAM.1 or 3G6 in solution. For all images, cells were stained with TRITC-Phalloidin and each image represents a max intensity projection of several z-stacks spanning the height of the cells. Scale bars = 10 μm. (C-D) Stacked bar graphs showing the ratio of platelets adopting a spread, filopodia-extending, or resting morphology under the conditions in (A) and (B), respectively. Morphological classification was performed by individuals blinded to experimental conditions. Error bars represent mean ± SD, n=4. (E-F) Individual graphs of each morphology in (C) and (D), respectively. Filopodia were defined as protrusions ≥0.5 μm in length. Significance determined by one-way ANOVA with post-hoc Tukey. p≤0.05; *, p≤0.01; **, p≤0.0001; ****.

Figure 5.. 3G6 and RAM.1 modulate platelet activation downstream of GPIb-IX.

Graphs of percentage of platelets positive for P-selectin (top) or both P-selectin and β-galactose (bottom) in platelets treated with AK2 (left) or 6B4 (right) over the indicated shear range. Platelets treated with AK2/6B4 alone (black), AK2/6B4 and 3G6 (red), AK2/6B4 and RAM.1 (blue), RAM.1 alone (green), or 3G6 alone (purple). Error bars represent mean ± SD, n=3. Significance determined by one-way ANOVA with post-hoc Tukey. p≤0.01; **, p≤0.001; ***, p≤0.0001; ****. ECL (Erythrina cristagalli lectin), PSel (anti-P-Selectin MAb).