Restricted accumulation of phosphatidylinositol 3-kinase products in a plasmalemmal subdomain during Fc gamma receptor-mediated phagocytosis

Abstract

Phagocytosis is a highly localized and rapid event, requiring the generation of spatially and temporally restricted signals. Because phosphatidylinositol 3-kinase (PI3K) plays an important role in the innate immune response, we studied the generation and distribution of 3' phosphoinositides (3'PIs) in macrophages during the course of phagocytosis. The presence of 3'PI was monitored noninvasively in cells transfected with chimeras of green fluorescent protein and the pleckstrin homology domain of either Akt, Btk, or Gab1. Although virtually undetectable in unstimulated cells, 3'PI rapidly accumulated at sites of phagocytosis. This accumulation was sharply restricted to the phagosomal cup, with little 3'PI detectable in the immediately adjacent areas of the plasmalemma. Measurements of fluorescence recovery after photobleaching were made to estimate the mobility of lipids in the cytosolic monolayer of the phagosomal membrane. Stimulation of phagocytic receptors induced a marked reduction of lipid mobility that likely contributes to the restricted distribution of 3'PI at the cup. 3'PI accumulation during phagocytosis was transient, terminating shortly after sealing of the phagosomal vacuole. Two factors contribute to the rapid disappearance of 3'PI: the dissociation of the type I PI3K from the phagosomal membrane and the persistent accumulation of phosphoinositide phosphatases.

Figure 1

Distribution of PH domain–GFP chimeras in macrophages. RAW 264.7 cells were transfected with chimeras of GFP and the indicated PH domain and then exposed to IgG-opsonized SRBCs to initiate phagocytosis. Confocal fluorescence (A, C, E, and F) and DIC images (B, D, and E, inset) were acquired. (A–B) Cells transfected with Akt-PH–GFP. The inset shows the distribution of Akt-PH–GFP before addition of SRBCs. (C–D) Cells transfected with Gab1-PH–GFP. (E) Cells transfected with Akt-PH–GFP were pretreated with 100 nM wortmannin for 15 min before addition of the SRBCs. (F) The main panel shows cells transfected with (R28C)Btk-PH–GFP, a mutant form of Btk-PH–GFP unable to bind 3′PI in vitro. The inset shows a cell transfected with wild-type Btk-PH–GFP. Arrows indicate sites of attachment of SRBCs. Images are representative of at least three experiments of each type. Bars, 10 μm.

Figure 2

Time course of accumulation of 3′PI in phagosomes. RAW 264.7 cells transfected with Akt-PH–GFP were exposed to IgG-opsonized SRBCs to initiate phagocytosis. Confocal fluorescence (B–E) and DIC images (A and F) were acquired at the indicated times. The time of closure of the two phagosomes noted by solid white arrows was arbitrarily chosen as time 0. A phagosome that had sealed before the acquisition of the first image is shown by an open arrow. Images are representative of six similar time course determinations.

Figure 3

Quantitation of the time course of 3′PI accumulation. RAW 264.7 cells transfected with PH domain–GFP chimeras or with soluble GFP were exposed to IgG-opsonized SRBCs to initiate phagocytosis, and confocal images were acquired. A representative image of a cell transfected with Akt-PH–GFP is shown in A. To quantify the accumulation of Akt-PH–GFP at the phagosomal membrane, lines were drawn through the digitized images, and the fluorescence intensity of individual pixels was determined. The line scan corresponding to A is shown in B; for reference, numbers have been assigned to the intersections with the phagosomal (1, 2) and contralateral plasma membrane (3). In C, the time course of accumulation of Akt-PH–GFP (▿) at the phagosomal membrane is compared with that of free GFP (○). In D, the course of association of Gab1-PH–GFP (□) and PLCδ/PH-GFP (•) are shown. The time of closure of the phagosomes was arbitrarily chosen as time 0 as in the legend to Fig. 2. The left ordinate, which applies to the open symbols, shows the fluorescence of the phagosomal membrane (p) minus the cytosolic fluorescence (c; subtracted to discount the free fluorophore present in the immediate vicinity of the phagosome) divided by c to normalize for expression level. The right ordinate, which applies to the solid circles (•), shows the fluorescence of the phagosomal membrane (p) divided by the fluorescence of a nonphagosomal region of the plasma membrane (m). Data are mean ± standard error of three to six experiments.

Figure 6

Measurement of PM–GFP FRAP. RAW 264.7 cells transfected with PM–GFP were detached by gentle scraping and then allowed to sediment on either IgG-coated or uncoated glass coverslips. (A) Confocal fluorescence images of a cell plated on IgG were acquired, and reconstruction of a vertical (x versus z) section is illustrated. (B) Confocal (x versus y) images of the adherent membrane were acquired before (Pre), immediately after (0 s), and at the indicated intervals following photobleaching of a spot near the edge of the membrane. In C, the rate of recovery after bleaching was quantified and compared for cells plated on glass (○) or IgG (□). Data are mean ± standard error of six individual experiments.

Figure 4

Quantitation of the time course of accumulation of Fc receptors, Syk, and PI3K on phagosomes. RAW 264.7 cells transfected with either FcγRIIa-GFP (A, ○), PM–GFP (A, ▿), Syk-GFP (B), or p85-GFP (C) were exposed to IgG-opsonized SRBCs to initiate phagocytosis, and confocal images were acquired. Representative images are shown in the insets. The time of closure of the phagosomes was arbitrarily defined as time 0. Quantitation in B and C was as in the legend to Fig. 2. In A, the ordinate refers to the fluorescence intensity of the phagosomal membrane (p) relative to the extraphagosomal (usually contralateral) plasmalemma (m). Data are mean ± standard error of at least three experiments.

Figure 5

Distribution of Akt-PH–GFP in cells adhering to IgG-coated versus uncoated glass. RAW 264.7 cells transfected with Akt-PH–GFP were detached by resuspension in HPMI with 2 mM EDTA and then allowed to sediment on either IgG-coated (A and B) or uncoated glass (C and D) for ∼10–20 min. DIC (A and C) and confocal fluorescence images were acquired, and vertical (x versus z) sections of Akt-PH–GFP are illustrated (B and D). Images are representative of five similar determinations.

Figure 8

Reversal of 3′PI accumulation by wortmannin. (A and B) RAW 264.7 cells were allowed to interact with IgG-opsonized polystyrene beads (8-μm diameter), and when phagocytosis commenced the cells were treated with either 100 nM wortmannin or with an equivalent volume of the vehicle (DMSO). After 60 s, the cells were fixed, and F-actin was stained with rhodamine-phalloidin. A and B show F-actin distribution, and insets show the corresponding DIC image. (A) DMSO-treated; (B) wortmannin-treated. In C, the cells were transfected with Akt-PH–GFP ∼24 h before exposure to the beads. The concentration of the chimera at the phagosomal cup was monitored by fluorescence microscopy, and when a sizable accumulation occurred the samples were treated with wortmannin or DMSO. The persistence of Akt-PH–GFP was monitored thereafter at 10 s intervals. Data were normalized to the fluorescence at the time of addition of the inhibitor to facilitate comparison. Data are mean ± standard error of three experiments.

Figure 7

Measurements of Akt-PH–GFP FRAP. RAW 264.7 cells transfected with Akt-PH–GFP were resuspended and allowed to sediment on IgG-coated coverslips as in the legend to Fig. 5. Confocal images of the adherent membrane were acquired before (A, Pre), immediately after (B, 0 s), and at the indicated intervals following photobleaching of a spot near the middle of the membrane (C–D). In E, the fluorescence intensity along line scans traversing the photobleached area was determined at the indicated times after photobleaching. The width of the initial bleached area is demarcated by the dotted lines. In F, the relative fluorescence intensity at the center of the area selected for analysis is plotted versus time after bleaching. Data are representative of five experiments.

Figure 9

Distribution of phosphoinositide phosphatases during phagocytosis. (A and B) Cells were exposed to human IgG-opsonized latex beads to initiate phagocytosis. After fixation and permeabilization, the distribution of endogenous PTEN was revealed by immunostaining using an affinity purified polyclonal antibody. Fluorescence (A) and DIC (B) images are shown. (C–F) RAW 264.7 cells were transfected with epitope-tagged SHIP1 and GFP and then exposed to opsonized SRBCs for either ∼6 (C–D) or ∼8 min (E–F). After fixation and permeabilization, the localization of SHIP1 was revealed by immunostaining using anti-HA monoclonal antibodies (C and E). The fluorescence of GFP is also shown (D and F). Arrows indicate the location of nascent (A–D) or sealed phagosomes (E–F). Results are representative of at least five experiments of each type.