An RGD sequence in the P2Y(2) receptor interacts with alpha(V)beta(3) integrins and is required for G(o)-mediated signal transduction

Abstract

The P2Y(2) nucleotide receptor (P2Y(2)R) contains the integrin-binding domain arginine-glycine-aspartic acid (RGD) in its first extracellular loop, raising the possibility that this G protein-coupled receptor interacts directly with an integrin. Binding of a peptide corresponding to the first extracellular loop of the P2Y(2)R to K562 erythroleukemia cells was inhibited by antibodies against alpha(V)beta(3)/beta(5) integrins and the integrin-associated thrombospondin receptor, CD47. Immunofluorescence of cells transfected with epitope-tagged P2Y(2)Rs indicated that alpha(V) integrins colocalized 10-fold better with the wild-type P2Y(2)R than with a mutant P2Y(2)R in which the RGD sequence was replaced with RGE. Compared with the wild-type P2Y(2)R, the RGE mutant required 1,000-fold higher agonist concentrations to phosphorylate focal adhesion kinase, activate extracellular signal-regulated kinases, and initiate the PLC-dependent mobilization of intracellular Ca(2+). Furthermore, an anti-alpha(V) integrin antibody partially inhibited these signaling events mediated by the wild-type P2Y(2)R. Pertussis toxin, an inhibitor of G(i/o) proteins, partially inhibited Ca(2+) mobilization mediated by the wild-type P2Y(2)R, but not by the RGE mutant, suggesting that the RGD sequence is required for P2Y(2)R-mediated activation of G(o), but not G(q). Since CD47 has been shown to associate directly with G(i/o) family proteins, these results suggest that interactions between P2Y(2)Rs, integrins, and CD47 may be important for coupling the P2Y(2)R to G(o).

Figure 1

P2Y₂ peptide-coated bead binding to K562 cells is inhibited by a soluble RGD peptide and antibodies to α_Vβ₃/β₅ integrins and CD47. (A) Latex beads were coated with the indicated peptide or protein and incubated with K562 cells suspended in HBSS for 1 h at 37°C in the presence or absence of preincubation peptide. The amount of bead binding to K562 cells was determined and the AI values were calculated. The data are the mean ± SEM (n = 3–6). (B) K562 cells were preincubated for 15 min in HBSS alone or in the presence of antiintegrin, control anticlass II HLA, or anti-CD47 antibodies. P2Y₂ ^93–110 peptide-coated latex beads were added and the incubation was continued for 1 h at 37°C. The AI values were calculated and the percentage of bead binding to K562 cells compared with the control (HBSS alone) is shown. The data are the mean ± SEM (n = 3).

Figure 2

Colocalization of P2Y₂Rs with α_V integrins. Human 1321N1 cells expressing HA-tagged wild-type or RGE mutant P2Y₂Rs were incubated at 37°C with or without 1 mM UTP for 5 min. The HA-tagged receptors were detected by confocal microscopy and immunofluorescence with the dye Cy5 (shown in red) and α_V integrins in the same cells were detected by immunofluorescence with the dye Oregon green (shown in green). The combined fluorescence, indicative of colocalization, is shown in yellow. The data are both representative and the mean ± SEM (n = 6).

Figure 3

Calcium mobilization mediated by wild-type and RGE mutant P2Y₂Rs. (A) Human 1321N1 cells expressing either wild-type (open symbols) or RGE mutant (filled symbols) P2Y₂Rs were assayed for receptor activity by quantitating changes in the concentration of cytoplasmic free calcium ([Ca²⁺]_i) in response to the indicated concentration of ATP (squares) or UTP (circles). Data are expressed as a percentage of the maximum increase in [Ca²⁺]_i elicited by the most effective ligand for each receptor construct and represent the mean ± SEM (n = 3). (B) 1321N1 transfectants were treated overnight with or without 200 ng/ml Bordetella pertussis toxin (PTX). The maximum change in [Ca²⁺]_i in response to 1 μM UTP for the wild-type P2Y₂R transfectants or 100 μM UTP for the RGE mutant P2Y₂R transfectants is shown as the mean ± SEM (n = 3) from one experiment and is representative of three separate experiments. The resting [Ca²⁺]_i was ∼130 nM for each condition.

Figure 5

MAP kinase activation mediated by the wild-type and RGE mutant P2Y₂Rs. Human 1321N1 cells expressing either wild-type or RGE mutant P2Y₂Rs were stimulated with (A) the indicated concentration of UTP for 5 min at 37°C or with (B) 100 μM UTP after a 30 min preincubation with buffer or 10 μM BAPTA-AM. MAP kinase activation (ERK1 and ERK2 phosphorylation) was detected by Western blot analysis with an antiphospho ERK1/2 antibody as described in Materials and Methods. The data are representative of results from four to six experiments.

Figure 4

FAK phosphorylation mediated by the wild-type and RGE mutant P2Y₂Rs. Human 1321N1 cells expressing either wild-type or RGE mutant P2Y₂Rs were stimulated with (A) the indicated concentration of UTP for 5 min at 37°C or with (B) 1 mM UTP after a 30 min preincubation with buffer or 10 μM BAPTA-AM. The cells were lysed in RIPA buffer and FAK was immunoprecipitated from the lysate. Phosphorylation of FAK was detected by Western blot analysis with an antiphosphotyrosine antibody. The data are representative of results from three to six experiments.

Figure 6

Effect of antiintegrin antibodies on IP formation and FAK and ERK1/2 phosphorylation by the P2Y₂R. Human 1321N1 cells expressing wild-type P2Y₂Rs were incubated for 24 h with or without 100 μg/ml of the indicated antibody. The cells were then stimulated with 1 μM UTP for 10 min (A), 1 μM UTP for 5 min (B), or 10 nM UTP for 5 min (C). For A, generation of IP is expressed as the percentage of maximum stimulation with UTP after background subtraction (backgrounds for control, α₃-treated, and α_V-treated samples were 18,462 ± 1,090, 15,670 ± 1,000, and 30,450 ± 3,650 dpm, respectively). For B, FAK phosphorylation was assayed by Western blot analysis with antiphospho FAK antibody as described in Materials and Methods. For C, ERK1/2 phosphorylation was assayed as described in the legend to Fig. 5. The data for A are the mean ± SEM (n = 3). The data for B and C are both representative (bottom) and the mean ± SEM (n = 3–5) (top).

Figure 7

Effect of RGDS and RGES peptides on FAK and ERK1/2 phosphorylation by the P2Y₂R. Human 1321N1 cells expressing wild-type P2Y₂Rs were (A) incubated for 18 h with 100 μM of the indicated peptide or (B) treated with 100 μM UTP for 10 min to induce internalization of the P2Y₂Rs, rinsed, and incubated for 2 h with 100 μM of the indicated peptide during the time period when internalized P2Y₂Rs are being recycled to the plasma membrane (Garrad et al. 1998). The cells were then stimulated with 1 μM UTP (for FAK experiments) or 10 nM UTP (for ERK1/2 experiments) for 5 min at 37°C. Phosphorylation of FAK and ERK1/2 were assayed as described in the legends to Fig. 4 and Fig. 5, respectively. The data are representative of results from three to five experiments.