A new role for the P2X7 receptor: a scavenger receptor for bacteria and apoptotic cells in the absence of serum and extracellular ATP

Abstract

The P2X7 receptor is widely recognized to mediate the proinflammatory effects of extracellular ATP. However this receptor in the absence of ATP may have a function unrelated to inflammation. Our data show that P2X7 expressed on the surface of monocyte/macrophages or on epithelial HEK-293 cells greatly augments the engulfment of latex beads and live and heat-killed bacteria by effector phagocyte in the absence of ATP and serum. The expression of P2X7 on the effector also confers the ability to phagocytose apoptotic target cells and an accumulation of P2X7 can be seen at the attachment point to the target. Activation of the P2X7 receptor by ATP causes a slow dissociation (over 10-15 min) of nonmuscle myosin from the P2X7 membrane complex and abolishes further P2X7-mediated phagocytosis of these targets. The recent crystal structure of the homologous zebrafish P2X4 receptor shows an exposed "nose" of the ectodomain (residues 115-162) which contains three of the five disulfide bonds conserved in all P2X receptors. Three short biotin-labeled peptides mimicking sequence of this exposed region bound to apoptotic target cells but not to either viable cells or to other target particles. All three peptides contained one or two cysteine residues and their replacement by alanine abolished peptide binding. These data implicate thiol-disulfide exchange reactions in the initial tethering of apoptotic cells to macrophage and establish P2X7 as one of the scavenger receptors involved in the recognition and removal of apoptotic cells in the absence of extracellular ATP and serum.

Fig. 1

Typical time courses of uptake of 1.0 μm yellow-green (YG) beads by fresh human monocytes from the same subject. Cells were labeled with allophycocyanin conjugated anti-CD14 mAb and pre-treated with 1 mM ATP for 15 min or 20 μM CytD for 30 min before the addition of YG beads. The linear mean channel of fluorescence intensity for each gated CD14+ monocyte population over successive 10-s intervals was analyzed by WinMDI software and plotted against time

Fig. 2

Phagocytosis of apoptotic HPB cells by HEK-293 cells transfected with P2X7: upper panel bright field image (left) and confocal image (right) showing a HEK-293 cell transfected with DsRed tagged P2X7 (red) containing two engulfed CFSE-labeled HPB cells and another which is tethered or attached to the HEK cell. Images were acquired by an Olympus IX81 confocal microscope (×600). Lower panel: a typical flow cytometry histogram showing phagocytosis of CFSE-labeled HPB cells by P2X7-DsRed transfected HEK-293 cells. HEK-293 cells were labeled with BODIPY 630/650-SE first and pre-treated with 1 mM ATP or 20 μM CytD for 15 min prior to incubation with HPB cells. BODIPY⁺/DsRed⁺⁺ HEK-293 cells were gated for analysis. From Gu et al. [29]

Fig. 3

Pretreatment of apoptotic lymphocytes with purified Annexin V does not inhibit their phagocytosis by autologous monocyte-derived macrophages. Human lymphocytes were induced into apoptotic death by overnight incubation with 1 μM staurosporine then labeled with CFSE, and incubated with or without 20 μg/mL purified Annexin V for 30 min at room temperature, Macrophages were labeled with CMTMR prior to the addition of CFSE-labeled apoptotic autologous lymphocytes. Mixed cells were incubated for 3 h at 37°C before fixation and assay by flow cytometry on gated CMTMR⁺ macrophages

Fig. 4

Recombinant P2X7 extracellular domain (P2X7-ED) binds to the surface of apoptotic cells but not to viable cells. Human neuronal SH-SY5Y cells were induced into apoptosis with 0.2 μM staurosporine. Cells were resuspended in Na⁺ medium with 3 mM CaCl₂, and incubated with P2X7-ED at a concentration of 15 μg/mL for 30 min, followed by two washes and incubation with anti-5×His mAb, and subsequently with FITC-conjugated goat anti-mouse IgG, APC-conjugated Annexin V, and 7-AAD. All events were gated on the low 7-AAD population to exclude necrotic cells. From Gu et al. [29]

Fig. 5

Predicted extracellular and transmembrane domains of human P2X7 structure. The P2X7 model was based on the crystal structure of zebrafish P2X4 in the closed state (Ref. [36]) using the online Protein Homology/analogY Recognition Engine (PHYRE, V2.0). The six conserved cysteine residues in the exposed “dolphin nose” region (residues 115–168) form three disulfide bonds shown between the matching colors

Fig. 6

The binding of short peptides mimicking P2X₇ extracellular domain sequences to a range of phagocytic targets. 24 biotin-tagged peptides were incubated at a concentration of 10 μg/mL with either YG beads (3 μm, 5 μL), Alexa 488 conjugated Staphylococcus aureus or Escherichia coli (2 mg/mL, 5 μL), live S. aureus or E. coli (OD₆₉₀ = 0.9, 10 μL each) or apoptotic HPB cells (5 × 10⁶/mL, 200 μL) for 30 min, followed by washing and incubation with HRP-labeled streptavidin for 30 min. Particles were washed twice, and resuspended in 100 μL of HRP-labeled streptavidin (Jackson ImmunoResearch Lab, 1:2,000 diluted with PBS containing 1% BSA) for 30 min. After two washes, 500 μL of SuperSignal West Pico was added and the chemiluminescence was measured by a Glo 20/20 luminometer (Promega). From Gu et al. [29]

Fig. 7

Schematic illustration of the P2X7 membrane complex showing ATP causes dissociation of nonmuscle myosin from the complex. P2X7 in its inactivated state forms a tight molecular complex with NMMHC-IIA. Once P2X7 is activated by extracellular ATP, NMMHC-IIA slowly dissociates from the complex over 10–15 min. In contrast, Ro-52 remains associated with P2X7. The full-length P2X7 structure modeling is predicted by I-TASSER online server (http://zhanglab.ccmb.med.umich.edu/I-TASSER/). The other protein structures are obtained from RCSB PDB (http://www.pdb.org/pdb/home/home.do) and EMBL-EBI (http://www.ebi.ac.uk/Databases/structure.html)