PtdIns3P and Rac direct the assembly of the NADPH oxidase on a novel, pre-phagosomal compartment during FcR-mediated phagocytosis in primary mouse neutrophils

Abstract

The generation of reactive oxygen species (ROS) by the nicotinamide adenine dinucleotide phosphate oxidase is an important mechanism by which neutrophils kill pathogens. The oxidase is composed of a membrane-bound cytochrome and 4 soluble proteins (p67(phox), p40(phox), p47(phox), and GTP-Rac). These components form an active complex at the correct time and subcellular location through a series of incompletely understood mutual interactions, regulated, in part, by GTP/GDP exchange on Rac, protein phosphorylation, and binding to lipid messengers. We have used a variety of assays to follow the spatiotemporal assembly of the oxidase in genetically engineered primary mouse neutrophils, during phagocytosis of both serum- and immunoglobulin G-opsonized targets. The oxidase assembles directly on serum-Staphylococcus aureus-containing phagosomes within seconds of phagosome formation; this process is only partially dependent (∼ 30%) on PtdIns3P binding to p40(phox), but totally dependent on Rac1/2 binding to p67(phox). In contrast, in response to immunoglobulin G-targets, the oxidase first assembles on a tubulovesicular compartment that develops at sites of granule fusion to the base of the emerging phagosome; oxidase assembly and activation is highly dependent on both PtdIns3P-p40(phox) and Rac2-p67(phox) interactions and delivery to the phagosome is regulated by Rab27a. These results define a novel pathway for oxidase assembly downstream of FcR-activation.

Figure 1

IgG-dependent extra-phagosomal ROS production. Primed WT mouse bone marrow–derived neutrophils (BMNs) were preincubated with NBT, left untreated (i), or incubated with serum-opsonized S aureus (ii); IgG-opsonized SRBCs (iii); immobilized immune complex IgG-BSA (iv); unopsonized (v) or IgG-opsonized (vi) latex beads; serum-opsonized (vii) or IgG-opsonized (viii) zymosan, as described in “Phagocytosis assays,” and supplemental Methods. Non-phagocytosed S aureus or SRBCs were lysed before processing. Samples were cytospun (i-ii) or allowed to adhere (iii-viii) onto glass coverslips, fixed in 4% paraformaldehyde, washed, and mounted as described in “Phagocytosis assays.” Dark formazan deposition was detected by DIC imaging on a Zeiss LSM 510 META point-scanning confocal microscope. Shown are representative DIC images, including an enlarged section of formazan deposition under each condition. The IgG-SRBC phagosome is indicated by a white star. Scale bars represent 5 μm.

Figure 2

Endogenous and heterologously expressed p67^phox or iPX are targeted to pre-phagosomal structures during phagocytosis of IgG-coated particles. (A) Primed WT neutrophils (5 × 10⁴) were incubated without or with 1 × 10⁶ serum-opsonized S aureus for 7 minutes at 37°C (i), or with 5 × 10⁴ unopsonized or IgG-opsonized beads for 20 minutes at 37°C (ii), as described in “Phagocytosis assays.” Samples were cytospun (i) or allowed to adhere (ii) onto glass slides, fixed, and stained for p67^phox as described in “Phagocytosis assays.” Mounted samples were visualized on a Zeiss LSM 510 META point-scanning microscope using fluorescence and DIC optics. Shown are representative fluorescence and DIC images for conditions tested. Phagosomal (i) and extra-phagosomal (ii; indicated by arrows) accumulation of p67^phox was quantified for at least 50 events under each condition using LSM 510 Image browser software and expressed as mean ± SEM as a percentage of cytosolic levels, *P < .0001, Student t test. (B-C) BMNs expressing p67^phox-GFP (Bi,Ci,Ciii) or GFP-iPX (Bii,Cii) in a WT genetic background were primed at a concentration of 2 × 10⁷/mL, placed on ice and gently mixed at regular intervals. Aliquots of 5 × 10⁵-1 × 10⁶ cells were settled onto glass coverslips in imaging chambers that were previously blocked with 100% FBS. Cells were incubated with serum-opsonized S aureus (1-5:1 S.aureus:neutrophil; B), or IgG-SRBCs (20:1 IgG-SRBCs:neutrophils; C). z-stacks of neutrophils undergoing phagocytosis were captured over time on a Nikon-Eclipse spinning disk confocal microscope equipped with an ultrasensitive EM-CCD camera, as described in “Live imaging.” A total of 20-22 slices were captured for each time frame, at a distance of 0.5 μm between each slice, encompassing the whole neutrophil. The exposure time was 100 milliseconds, and the time interval between each frame was 5.896 seconds (p67^phox-GFP), 10.52 seconds (GFP-iPX; Bii), or 11.046 seconds (GFP-iPX; Cii). 3D reconstruction was performed using Volocity software, and reconstructions at individual time frames representing different stages of phagocytosis are shown. The time of attachment of particle to neutrophil and/or the time of formation of the phagocytic cup was defined as t = 0. Stars indicated the position of particles during various stages of phagocytosis; white arrows indicate the position of bacterial phagosomes containing p67^phox-GFP or GFP-iPX (B) or pre-phagosomal accumulation of p67^phox-GFP or GFP-iPX (C). Imaging experiments were performed with neutrophils from individual mice, on at least 2 different days. (Ciii) Series of confocal images over time, showing fusion of a pre-phagosomal structure containing p67^phox-GFP with an IgG-SRBC phagosome.

Figure 3

Assembly and activation of the NADPH oxidase in response to serum-opsonized S aureus is partially dependent on binding of p40^phox to PtdIns3P, but fully dependent on Rac. (A) Primed neutrophils (1 × 10⁶) isolated from WT (black squares/bars) or p40^{phoxR58A/R58A} (R58A, dark gray diamonds/bars) genotypes (i,iii); or from WT (black squares/bars), Rac1 knockdown on a WT background (Rac1KD, gray diamonds/bars), Rac2^−/− (Rac2KO, light gray triangle/bar), and Rac1 knockdown on a Rac2 KO background (Rac1KD/Rac2KO, open circles/bar; ii-iii) were preincubated with luminol/HRP, before addition to serum-S aureus as described in “Measurement of ROS production.” Total ROS responses were measured by chemiluminesence and recorded on a 96-well plate using a Berthold Microlumat Plus luminometer as described in “Measurement of ROS production.” All incubations were performed in at least duplicate. Shown are data (mean ± range) from 1 experiment representative of at least 3, expressed as RLU/s (i-ii), as well as accumulated light emission (ROS response) over 20 minutes for a combination of ≥ 3 experiments (mean ± SEM), expressed as a percentage of the response in WT mouse neutrophils (iii). *P < .0001, paired Student's t test (WT/R58A), 1-way analysis of variance (ANOVA) (WT, Rac1KD, Rac2KO, Rac1KD/Rac2KO), compared with WT response. (B) Primed BMNs from the indicated genetic backgrounds, after phagocytosis of serum-S aureus were fixed and stained and imaged for p67^phox as described in Figure 2A. Phagosomal p67^phox fluorescence was quantified as described in panel A and is expressed (mean ± SEM) as a percentage of WT levels, *P < .0001, 1-way ANOVA. Arrows indicate the position of S aureus bacteria in the DIC images. (C) For quantification of cytosolic (1) and phagosomal (2) fluorescence, mouse BMNs expressing p67^phox WT-GFP on a WT or a p40^{phoxR58A/R58A} genetic background or expressing p67^phox H69E-GFP on a WT genetic background were primed and incubated with S aureus as described in Figure 2B. Images were recorded on an Olympus CellR epifluorescence microscope equipped with an Olympus 1 × 2-UCB camera and using a 60× oil objective, as further described in “Live imaging.” Images were recorded for 15 minutes in the GFP channel and DIC channel (not shown) at 4-second intervals and at an exposure time of 750 milliseconds. Representative images are shown. Fluorescence around the phagosomes was quantified using ImageJ software, as described in “Live imaging” and supplemental Figure 9. Data are presented as a ratio of MFI of phagosome:cytosol. The gray transparent area represents a ratio of phagosome:cytosol of 1.0 or lower, representing no phagosomal translocation. Data are mean ± SEM (n ≥ 19 phagosomes in at least 5 different neutrophils per construct/genetic background). *P < .0001 as determined by Student t test between WT and R58A, and between WT and H69E.

Figure 4

Assembly and activation of the NADPH oxidase in response to IgG-SRBCs is fully dependent on binding of p40^phox to PtdIns3P and on Rac2. (A) BMNs from WT, p40^{phoxR58A/R58A}, or Rac2KO genetic backgrounds were preincubated with NBT before incubation with IgG-SRBCs and processed as described in Figure 1, with the exception of Rac2KO samples, which were cytospun onto coverslips. Shown are representative images, with IgG-SRBC phagosomes indicated by stars. (B) Primed BMNs from WT (black squares/bar) or p40^{phoxR58A/R58A} genotypes (dark gray diamonds/bar; i,iii); or from WT (black squares/bar), Rac1KD (gray diamonds/bars), Rac2KO (light gray triangles/bars), and Rac1KD/Rac2KO (open circles/bars) genotypes (ii-iii), were subjected to luminol-dependent chemiluminescence assays for total ROS generation as described in Figure 3A. Shown are data (mean ± range) from 1 experiment representative of at least 3, expressed as RLU/s (i-ii), as well as accumulated light emission (ROS response) over 20 minutes for a combination of ≥ 3 experiments (mean ± SEM), expressed as a percentage of the response in WT mouse neutrophils (iii). *P < .0001, paired Student t test (WT/R58A), 1-way ANOVA (WT, Rac1KD, Rac2KO, Rac1KD/Rac2KO) compared with WT response. (C) For quantification of cytosolic fluorescence (1), fluorescence at sites of attachment of IgG-SRBCs to neutrophil (2), around the phagocytic cup (3), the phagosome (4), and after phagocytosis (5), widefield epifluorescence microscopy was used. Mouse BMNs expressing p67^phox WT-GFP on a WT or a p40^{phoxR58A/R58A} genetic background or expressing p67^phox H69E-GFP on a WT genetic background were primed and incubated with IgG-SRBCs as described in Figure 2C. Images were recorded and presented as described in Figure 3C. Quantification was performed using ImageJ software, as described in “Live imaging” and supplemental Figure 9. Data are presented as a ratio of MFI of phagosome:cytosol. The gray transparent area represents a ratio of phagosome:cytosol of 1.0 or lower, representing no phagosomal translocation. Data are mean ± SEM (n = 4-17 neutrophils). *P ≤ .01 as determined by Student t test between WT and R58A, and between WT and H69E.

Figure 5

Extra-phagosomal structures containing endogenous p67^phox are derived from azurophil and specific granules. Mouse neutrophils were incubated without (cells alone) or with IgG-opsonized 3-μm latex beads as described in “Phagocytosis assays” and were allowed to adhere on glass coverslips. Cells were fixed and permeabilized as described in “Phagocytosis assays,” stained with relevant antibodies, mounted, and visualized on a Zeiss LSM 510 META point-scanning confocal microscope. The position of the phagosome is indicated by a star. Arrows indicate the position of the p67^phox-positive extra-phagosomal structure and of granule markers colocalizing with this structure.

Figure 6

Phagosomal delivery of the oxidase in response to IgG-SRBCs is dependent on Rab27a. Primed BMNs from WT (black squares/bars) or ashen Rab27a^−/− (Rab27a KO, gray triangles/bars) animals were preincubated with luminol/SOD (intracellular) or isoluminol/HRP (extracellular; A) or NBT (B) as described in “Measurement of ROS production” and in the supplemental Methods. Cells (5 × 10⁵) were added to either serum-S aureus (i) or IgG-SRBCs (ii), prepared as described in the supplemental Methods. ROS responses (A) and formazan deposition (B) were measured as described in Figures 3A and 1, respectively. Shown are (A) data (mean ± range) from 1 experiment representative of at least 3, expressed as RLU/s, as well as accumulated light emission (ROS production) over 20 minutes for a combination of ≥ 3 experiments (mean ± SEM), expressed as a percentage of the response in WT mouse neutrophils. *P < .0001, paired Student t test compared with WT response and (B) representative images.

Figure 7

IgG dependency of extra-phagosomal superoxide production in human neutrophils. (A) Primed human peripheral blood neutrophils were preincubated with NBT and left untreated (cells alone) or incubated with S aureus opsonized with either 10% normal serum (complete) or serum depleted of IgG (IgG-depleted) before opsonization, as described in the supplemental Methods. Nonphagocytosed S aureus were lysed by lysostaphin, and samples cytospun onto glass coverslips. Samples were prepared and visualized as described in Figure 1. Shown are representative images. Arrows mark extra-phagosomal formazan deposition in response to phagocytosis of S aureus opsonized with complete serum. (Aii) Number of S aureus phagocytosed under each condition (S aureus/cell), and extra-phagosomal formazan deposits (extra-phagosomal formazan/cell) in at least 40 cells from each condition were counted and expressed as mean ± SEM, P < .0001. (B) Human neutrophils were preincubated with luminol/SOD (intracellular) or isoluminol/HRP (extracellular) as described in “Measurement of ROS production,” before addition to S aureus opsonized with either 10% normal serum (complete, comp) or serum depleted of IgG (IgG-depleted, IgG-dep) before opsonization. ROS responses were measured and data are presented as described in Figure 6A. *P ≤ .0001, paired Student t test compared with complete serum-opsonized S aureus–induced responses. (C) Neutrophils were incubated with NBT, left untreated (cells alone) or incubated with IgG-SRBCs, or IgG-opsonized or unopsonized 3-μm beads, as described in the supplemental Methods. Cells were allowed to adhere to coverslips for 3 minutes at 37°C, and samples were prepared and visualized as described in Figure 1. Shown are representative images. Position of the IgG-SRBC phagosome is indicated by a star.