Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2021 Jan-Jun:296:100256.
doi: 10.1016/j.jbc.2021.100256. Epub 2021 Jan 8.

Complement receptor 3 mediates both sinking phagocytosis and phagocytic cup formation via distinct mechanisms

Stefan Walbaum  1 , Benjamin Ambrosy  1 , Paula Schütz  1 , Anne C Bachg  1 , Markus Horsthemke  1 , Jeanette H W Leusen  2 , Attila Mócsai  3 , Peter J Hanley  4
Affiliations

Complement receptor 3 mediates both sinking phagocytosis and phagocytic cup formation via distinct mechanisms

Stefan Walbaum et al. J Biol Chem. 2021 Jan-Jun.

Abstract

A long-standing hypothesis is that complement receptors (CRs), especially CR3, mediate sinking phagocytosis, but evidence is lacking. Alternatively, CRs have been reported to induce membrane ruffles or phagocytic cups, akin to those induced by Fcγ receptors (FcγRs), but the details of these events are unclear. Here we used real-time 3D imaging and KO mouse models to clarify how particles (human red blood cells) are internalized by resident peritoneal F4/80+ cells (macrophages) via CRs and/or FcγRs. We first show that FcγRs mediate highly efficient, rapid (2-3 min) phagocytic cup formation, which is completely abolished by deletion or mutation of the FcR γ chain or conditional deletion of the signal transducer Syk. FcγR-mediated phagocytic cups robustly arise from any point of cell-particle contact, including filopodia. In the absence of CR3, FcγR-mediated phagocytic cups exhibit delayed closure and become aberrantly elongated. Independent of FcγRs, CR3 mediates sporadic ingestion of complement-opsonized particles by rapid phagocytic cup-like structures, typically emanating from membrane ruffles and largely prevented by deletion of the immunoreceptor tyrosine-based activation motif (ITAM) adaptors FcR γ chain and DAP12 or Syk. Deletion of ITAM adaptors or Syk clearly revealed that there is a slow (10-25 min) sinking mode of phagocytosis via a restricted orifice. In summary, we show that (1) CR3 indeed mediates a slow sinking mode of phagocytosis, which is accentuated by deletion of ITAM adaptors or Syk, (2) CR3 induces phagocytic cup-like structures, driven by ITAM adaptors and Syk, and (3) CR3 is involved in forming and closing FcγR-mediated phagocytic cups.

Keywords: Complement system; knockout mice; live-cell imaging; macrophages; phagocytosis.

PubMed Disclaimer

Conflict of interest statement

Conflict of interest The authors declare that they have no conflicts of interest with the contents of this article.

Figures

Figure 1
Figure 1
Phagocytic cup formation mediated by Fcγ receptors.A, 3D time-lapse images obtained by spinning disk confocal microscopy showing a WT macrophage (Mϕ) ingesting a human red blood cell (hRBC) opsonized with mouse IgG antibodies (mouse anti-human CD235a monoclonal IgG2b antibodies). The macrophage was labeled green fluorescent with Alexa Fluor 488–conjugated anti-F4/80 antibodies, and hRBCs were labeled red fluorescent with the plasma membrane stain CellMask Orange. Scale bars, 10 μm. B, ingestion of an IgG-opsonized hRBC by a morphologically spherical macrophage. Note that the tip of the ingested and squeezed hRBC is pinched off during cup closure (white arrow). This may serve as an immune defense strategy to lyse ingested cells. Scale bar, 10 μm (grid spacing [3D view] = 3.05 μm). C, lack of ingestion of IgG-opsonized hRBCs by No ITAM (NOTAM) (grid spacing = 6.46 μm) and Fcer1g−/− macrophages (grid spacing = 6.49 μm). NOTAM macrophages contain nonsignaling immunoreceptor tyrosine-based activation motifs (ITAMs) in the Fc receptor γ chain. D, cumulative plots of phagocytic (ingestion) events for WT, NOTAM, and Fcer1g−/− macrophages. The plots for NOTAM and Fcer1g−/− macrophages are overlaid. The x-axes show individual macrophages. E, summary box plots of phagocytic efficiency, defined as the percentage of IgG-opsonized hRBCs (in contact with a macrophage) which were fully ingested. Medians are represented by a horizontal line with a lilac-colored circle. Notably, following the introduction of IgG-opsonized hRBCs, the phagocytic efficiencies of NOTAM and Fcer1g−/− macrophages were negligible, whereas WT macrophages were highly efficient eaters. Data were analyzed using the Kruskal–Wallis one-way ANOVA (H = 89.8, degrees of freedom = 2, and p < 0.0001). Post hoc comparisons were performed using the Dunn test; n (number of macrophages) = 35 (from 3 WT mice), n = 26 (from 4 NOTAM mice), and n = 35 (from 4 Fcer1g−/− mice). ∗∗∗p < 0.0001. IgG, immunoglobulin G.
Figure 2
Figure 2
Ingestion of dual IgG- and complement C3b/iC3b-opsonized human red blood cells by NOTAM and Fcer1g−/−macrophages.A, human red blood cells (hRBCs) were dual-opsonized with IgG and complement C3b/iC3b by incubating IgG-opsonized hRBCs with complement C5-deficient (C5 null) mouse serum. B, macrophages isolated from No ITAM (NOTAM) mice contain nonsignaling immunoreceptor tyrosine-based activation motifs (ITAMs) in the Fc receptor γ chain, whereas the Fc receptor γ chain (encoded by Fcer1g) is deleted in macrophages from Fcer1g−/− mice, leading to loss of expression of the α-chains FcγRI, FcγRIII, and FcγRIV. The immunoreceptor tyrosine-based inhibition motif (ITIM)-containing FcγR, FcγRIIb, is functional in both NOTAM and Fcer1g−/− macrophages. C, ingestion of dual IgG- and C3b/iC3b-opsonized hRBCs by NOTAM macrophages via phagocytic cup formation (upper panel) or partial sinking phagocytosis (lower panel). The phagocytic cups in the upper panel arose from the extension of membrane protrusions, which rolled over the particle (hRBC). Particle number 3 was not ingested because of retraction of the membrane protrusion. 3D time-lapse imaging was performed for 16 min by spinning disk confocal microscopy. Scale bars, 10 μm. D, ingestion of a dual IgG–opsonized and C3b/iC3b-opsonized hRBC by a Fcer1g−/− macrophage via sinking phagocytosis. Scale bar, 10 μm. E, cumulative plots of phagocytic (ingestion) events for NOTAM and Fcer1g−/− macrophages. The x-axes show individual macrophages. F, summary box plots of macrophage phagocytic efficiency, defined as the percentage of IgG-opsonized hRBCs (in contact with a macrophage) which were fully ingested. Medians are represented by a horizontal line with a lilac-colored circle. Data were analyzed using the Kruskal–Wallis one-way ANOVA (H = 50.4, degrees of freedom = 2, and p < 0.0001). Post hoc comparisons were performed using the Dunn test; n (number of macrophages) = 35 (from 3 WT mice), n = 41 (from 4 NOTAM mice), and n = 31 (from 2 Fcer1g−/− mice). ∗∗∗p < 0.0001. IgG, immunoglobulin G.
Figure 3
Figure 3
Phagocytosis of dual IgM– and complement C3b/iC3b–opsonized human red blood cells by WT and Fcer1g−/−macrophages via sinking or phagocytic cup formation.A, ingestion of a dual IgM– and complement C3b/iC3b–opsonized human red blood cell (hRBC) by a WT macrophage. The macrophage was labeled green fluorescent with Alexa Fluor 488–conjugated anti-F4/80 antibodies, and the cytosol of the hRBC was loaded with the red fluorescent probe pHrodo Red. Time-lapse 3D imaging for 30 min was performed by spinning disk confocal microscopy. Particle engulfment was achieved by a tangental membrane protrusion (straight white arrow), which subsequently rolled over (curved white arrow) the hRBC. Scale bar, 10 μm. B, ingestion of several dual IgM– and complement C3b/iC3b–opsonized hRBCs by a Fcer1g−/− (Fc receptor γ-chain KO) macrophage. The white curved arrows indicate engulfment via membrane protrusions rolling over the hRBC. Scale bars, 10 μm. C, cumulative plots of phagocytic (ingestion) events for WT and Fcer1g−/− macrophages. The x-axes show individual macrophages. D, summary box plots of phagocytic efficiency, defined as the percentage of dual IgM– and complement C3b/iC3b–opsonized hRBCs (in contact with a macrophage), which were fully ingested. Medians are represented by a horizontal line with a lilac-colored circle. Data were analyzed using the Mann–Whitney U test (U = 1177.5, p = 0.47); n (number of macrophages) = 92 (from 3 WT mice) and n = 28 (from 2 Fcer1g−/− mice). n.s., not significant (p > 0.05). IgM, immunoglobulin M.
Figure 4
Figure 4
Conditional deletion of Syk in macrophages inhibits Fcγ receptor–mediated phagocytosis and phagocytic cup formation by complement receptors.A, ingestion of a pair of IgG-opsonized human red blood cells (hRBCs) by a WT macrophage. 3D time-lapse imaging for 16 min was performed by spinning disk confocal microscopy. Macrophages were labeled green fluorescent with Alexa Fluor 488–conjugated anti-F4/80 antibodies, and the plasma membrane was labeled red fluorescent with CellMask Orange. Scale bar, 10 μm. B, lack of ingestion of IgG-opsonized hRBCs by macrophages isolated from myeloid-restricted Syk conditional KO (cKO) mice. Scale bar, 10 μm. C, partial sinking of dual IgM– and complement C3b/iC3b–opsonized hRBCs into Syk cKO macrophages. The white arrows indicate sinking phagocytic events. Scale bars, 10 μm. D, summary box plots of projected cell area in WT and Syk cKO macrophages. The medians are represented by a horizontal line with a lilac-colored circle. Data were analyzed using the Mann–Whitney U test (U = 6896, p =0.0004); n (number of macrophages) = 122 (from 4 WT mice) and n = 88 (from 3 Syk cKO mice). ∗∗p < 0.001. E, cumulative plots of phagocytic (ingestion) events for Syk cKO macrophages presented with either IgG-opsonized or dual IgG– and complement C3b/iC3b–opsonized hRBCs. The x-axes show individual macrophages. IgG, immunoglobulin G.
Figure 5
Figure 5
Slow sinking phagocytosis mediated by complement receptors in Syk conditional KO (cKO) macrophages presented with dual IgM– and complement C3b/iC3b–opsonized targets.A and B, partial or complete ingestion of dual IgM– and complement C3b/iC3b–opsonized human red blood cells (hRBCs) by myeloid-restricted Syk conditional KO (cKO) macrophages. The inset in the lower panel (B) shows an XY optical section. 3D time-lapse imaging was performed by spinning disk confocal microscopy. Macrophages were labeled green fluorescent with Alexa Fluor 488–conjugated anti-F4/80 antibodies, and the cytosol of hRBCs was loaded with the red fluorescent probe pHrodo Red. Scale bars, 10 μm. C, cumulative plots of phagocytic (ingestion) events for WT and Syk cKO macrophages presented with dual IgM– and complement C3b/iC3b–opsonized hRBCs. The x-axes show individual macrophages. D, summary box plots of phagocytic efficiency, defined as the percentage of dual IgM– and complement C3b/iC3b–opsonized hRBCs (in contact with a macrophage), which were fully ingested. Medians are represented by a horizontal line with a lilac-colored circle. Data were analyzed using the Mann–Whitney U test (U = 1002, p =0.6); n (number of macrophages) = 31 (from 3 WT mice) and n = 61 (from 3 Syk cKO mice). n.s. = not significant (p > 0.05). IgM, immunoglobulin M.
Figure 6
Figure 6
Slow sinking phagocytosis in macrophages lacking the ITAM adaptor Fc receptor γ chain and DAP12.A, slow partial sinking of dual IgG– and complement C3b/iC3b–opsonized human red blood cells (hRBCs) into Fcer1g/Tyrobp double KO (dKO) macrophages (Mϕs), which lack the ITAM adaptor proteins Fc receptor γ chain and DAP12. Time-lapse 3D imaging for 16 min was performed by spinning disk confocal microscopy. Macrophages were labeled green fluorescent with Alexa Fluor 488–conjugated anti-F4/80 antibodies, and the plasma membrane was labeled red fluorescent with CellMask Orange. Scale bars, 10 μm. B, cumulative plots of phagocytic (ingestion) events for WT and Fcer1g/Tyrobp dKO macrophages presented with dual IgG– and complement C3b/iC3b–opsonized hRBCs. The x-axes show individual macrophages. C, summary box plots of phagocytic efficiency, defined as the percentage of dual IgG– and complement C3b/iC3b–opsonized hRBCs (in contact with a macrophage) which were fully ingested. Medians are represented by a horizontal line with a lilac-colored circle. Data were analyzed using the Mann–Whitney U test (U = 1465.5, p < 0.0001); n (number of macrophages) = 46 (from 2 WT mice) and n = 35 (from 2 Fcer1g/Tyrobp dKO mice). ∗∗∗p < 0.0001. IgG, immunoglobulin G; ITAM, immunoreceptor tyrosine-based activation motif.
Figure 7
Figure 7
Ingestion of dual IgM– and C3b/iC3b–opsonized human red blood cells by Fcer1g/Tyrobp double KO (dKO) macrophages. Ingestion of dual IgM– and complement C3b/iC3b–opsonized human red blood cells (hRBCs) by Fcer1g/Tyrobp dKO macrophages (Mϕs), which lack the ITAM-containing Fc receptor γ chain and the ITAM adaptor DAP12. Time-lapse 3D imaging for 30 min was performed by spinning disk confocal microscopy. Macrophages were labeled green fluorescent with Alexa Fluor 488–conjugated anti-F4/80 antibodies, and the cytosol of hRBCs was loaded with the red fluorescent probe pHrodo Red. Scale bars, 10 μm. IgM, immunoglobulin M; ITAM, immunoreceptor tyrosine-based activation motif.
Figure 8
Figure 8
Phagocytic cup formation is less impaired in macrophages lacking Tyrobp (DAP12) than in those lacking both Tyrobp and Fcer1g.A, time-lapse 3D images obtained by spinning disk confocal microscopy showing slow sinking phagocytosis events associated with various degrees of membrane protrusive activity. Tyrobp (DAP12) KO (Tyrobp−/−) macrophages (Mϕs) were presented with dual IgM– and C3b/iC3b–opsonized human red blood cells (hRBCs). Macrophages were labeled green fluorescent with Alexa Fluor 488–conjugated anti-F4/80 antibodies, and the cytosol of hRBCs was loaded with the red fluorescent probe pHrodo Red. The extended focus image at the top right shows resident peritoneal Tyrobp−/− cells labeled with Alexa Fluor 594–conjugated anti-CD11b (CD11b is a component [α-subunit] of complement receptor 3) and Alexa Fluor 488–conjugated anti-F4/80 antibodies (inset). Nuclei (blue channel) were stained with Hoechst 33342, a fluorescent nucleic acid stain. Scale bars, 10 μm. B, cumulative plots of phagocytic (ingestion) events for WT, Fcer1g/Tyrobp double KO (dKO), and Tyrobp−/− macrophages presented with dual IgM– and complement C3b/iC3b–opsonized hRBCs. The x-axes show individual macrophages. IgM, immunoglobulin M.
Figure 9
Figure 9
Phagocytosis of dual IgM– and C3b/iC3b–opsonized human red blood cells is abrogated in Itgb2−/−(complement receptor 3 [CR3] KO) macrophages.A and B, 3D time-lapse images obtained by spinning disk confocal microscopy showing lack of phagocytic activity in Itgb2−/− (CR3 KO) macrophages, irrespective of whether cells were spread out (A) or rounded up (B). Itgb2−/− (CR3 KO) macrophages (Mϕs) were presented with dual IgM– and complement C3b/iC3b–opsonized human red blood cells (hRBCs). Macrophages were labeled green fluorescent with Alexa Fluor 488–conjugated anti-F4/80 antibodies, and the cytosol of hRBCs was loaded with the red fluorescent probe pHrodo Red. Scale bars (A and B), 10 μm. C, cumulative plots of phagocytic (ingestion) events for Itgb2−/− macrophages presented with either dual IgM– and complement C3b/iC3b–opsonized hRBCs or IgG-opsonized hRBCs. D, summary box plots of phagocytic efficiency, defined as the percentage of dual IgM– and complement C3b/iC3b–opsonized hRBCs or IgG-opsonized hRBCs (in contact with a macrophage), which were fully ingested. Medians are represented by a horizontal line with a lilac-colored circle. Data were analyzed using the Kruskal–Wallis one-way ANOVA (H = 121.8, degrees of freedom = 3, and p < 0.0001). Post hoc comparisons were performed using the Dunn test; n (number of macrophages) = 46 (from 3 WT mice), n = 44 (from 3 CR3 KO [Itgb2−/−] mice), n = 52 (from 3 WT mice), and n = 48 (from 3 CR3 KO mice), respectively, from left to right. ∗∗∗p < 0.0001. IgM, immunoglobulin M; IgG, immunoglobulin G.
Figure 10
Figure 10
Aberrant elongated and cylindrical Fcγ receptor–mediated phagocytic cups in Itgb2−/−(complement receptor 3 [CR3] KO) macrophages.A, ingestion of an IgG-opsonized human red blood cell (hRBC) by a WT macrophage. 3D time-lapse imaging was performed by spinning disk confocal microscopy. Macrophages were labeled green fluorescent with Alexa Fluor 488–conjugated anti-F4/80 antibodies and the cytosol of hRBCs was loaded with the red fluorescent probe pHrodo Red. Scale bars, 10 μm. B and C, Itgb2−/− (CR3 KO) macrophages ingest IgG-opsonized hRBCs via markedly elongated, cylindrical phagocytic cups. Scale bars, 10 μm. D, elongated, cylindrical phagocytic cup produced by a morphologically spherical Itgb2−/− macrophage. Scale bar, 10 μm. E, summary box plots of projected cell area in WT and Itgb2−/− macrophages. Medians are represented by a horizontal line with a lilac-colored circle. Data were analyzed using the Mann–Whitney U test (U = 3015, p =0.69); n (number of macrophages) = 55 (from 3 WT mice) and n = 114 (from 3 Itgb2−/− mice). n.s., not significant (p > 0.05). F, summary box plots of the maximal phagocytic cup length in WT and Itgb2−/− macrophages. Data were analyzed using the Mann–Whitney U test (U = 290, p < 0.0001); n (number of macrophages) = 62 (from 3 WT mice) and n = 56 (from 3 Itgb2−/− mice). ∗∗∗p < 0.0001. IgG, immunoglobulin G.
Figure 11
Figure 11
Kinetics of phagocytic cup formation and sinking phagocytosis.A, ingestion of an IgG-opsonized human red blood cell (hRBC) by a WT macrophage. 3D time-lapse imaging was performed by spinning disk confocal microscopy. Macrophages were labeled green fluorescent with Alexa Fluor 488–conjugated anti-F4/80 antibodies, and the plasma membrane was labeled red fluorescent with CellMask Orange. Note that the contact point between the IgG-opsonized hRBC and the macrophage elicits local membrane protrusions, which mark the initiation of phagocytic cup formation. Scale bars, 10 μm. B, ingestion of dual IgM– and complement C3b/iC3b–opsonized hRBCs by a Syk cKO macrophage via slow sinking phagocytosis. Local membrane protrusions (indicated by white arrows) accompany sinking phagocytosis to varying extents. Scale bar, 10 μm. C, formation of a phagocytic cup upon a filopodium, extending from an Itgb2−/− (complement receptor 3 deficient) macrophage. The white arrow shows thickening of the membrane at the tip of a filopodium, where it makes contact with an IgG-opsonized hRBC. The bulge of the macrophage cell body, marked by the white numerical digit 1, indicates where an IgG-opsonized hRBC has been fully internalized. D, summary box plots of ingestion times for phagocytic cup formation and sinking phagocytosis. Medians are represented by a horizontal line with a lilac-colored circle. hRBCs were singly opsonized with IgG or dual-opsonized with IgM and complement C3b/iC3b, as indicated. Data were analyzed using the Kruskal–Wallis one-way ANOVA (H = 81.3, degrees of freedom = 3, and p < 0.0001). Post hoc comparisons were performed using the Dunn test; n (number of macrophages) = 59 (from 6 WT mice), n = 37 (from 3 WT mice), n = 52 (from 3 Itgb2−/− mice), and n = 23 (from 3 Syk cKO mice), respectively, from left to right. ∗∗∗p < 0.0001. IgM, immunoglobulin M; IgG, immunoglobulin G.
Figure 12
Figure 12
Schematic summary.A, WT macrophages ingest IgG-opsonized human red blood cells (IgG-hRBCs) with high efficiency via rapid (2–3 min) phagocytic cup formation, which includes extension of ruffles, especially peripheral ruffles. Phagocytic cup formation after application of IgG–RBCs is abrogated in both Fcer1g−/− macrophages, which lack the Fc receptor γ chain, and NOTAM macrophages, which express a mutant Fc receptor γ chain with a nonsignaling ITAM. Phagocytic cups are also lacking in Syk conditional KO (cKO) macrophages. In contrast, phagocytic cups become elongated and cylindrical in Itgb2−/− (complement receptor 3 [CR3] KO) macrophages presented with IgG–hRBCs. A model of the signal transduction underlying Fcγ receptor (FcγR)–mediated phagocytosis is shown on the right. Note that phagocytic cups can form upon a filopodium and cup closure can pinch off the tips of cells, as shown in the lower panel. B, whereas FcγRI, FcγRIII, and FcγRIV associate with the FcR γ chain, which contains an ITAM, the inhibitory FcγR (FcγRIIb) contains an immunoreceptor tyrosine-based inhibition motif (ITIM) in its cytosolic tail. Phagocytic cup formation was less efficient in NOTAM macrophages (which have surface expression of all four FcγRs, but lack a functional ITAM) than WT macrophages, but largely blocked in Fcer1g−/− (FcR γ chain KO) macrophages, which do not express surface FcγRI, FcγRIII, and FcγRIV. Thus, in Fcer1g−/− macrophages, FcγRIIb has less competition for binding to opsonic IgG, which would promote ITIM signaling, as shown in the model on the right. C, WT macrophages ingest dual IgM– and complement C3b/iC3b–opsonized hRBCs via phagocytic cup formation, extended ruffles, and varying degrees of sinking. Dual IgM– and C3b/iC3b–opsonized hRBCs are not ingested by Itgb2−/− (CR3 KO) macrophages, implying that CR3, rather than CRIg, induces phagocytic cup formation and sinking phagocytosis. Notably, IgM, in contrast to IgG, is not recognized by Fcγ receptors. CR3-mediated phagocytic cup formation, but not sinking phagocytosis, is abrogated in macrophages lacking Syk (Syk conditional KO [cKO] macrophages) or the immunoreceptor tyrosine-based activation motif (ITAM) adaptors FcR γ chain and DAP12 (Fcer1g/Tyrobp double KO [dKO] macrophages). The endocytic pathway during sinking phagocytosis is restricted in diameter and hRBCs, which are highly deformable, are squeezed into the cell interior, producing a characteristic intermediary hourglass morphology. A model of the signal transduction underlying CR3-mediated phagocytosis is shown on the right. IgG, immunoglobulin G; IgM, immunoglobulin M; ITAM, immunoreceptor tyrosine-based activation motif; NOTAM, No ITAM.
See this image and copyright information in PMC

Similar articles

See all similar articles

Cited by

See all "Cited by" articles

References

    1. Metschnikoff E. Lecture on phagocytosis and immunity. Br. Med. J. 1891;1:213–217. - PMC - PubMed
    1. Tauber A.I. Metchnikoff and the phagocytosis theory. Nat. Rev. Mol. Cell Biol. 2003;4:897–901. - PubMed
    1. Flannagan R.S., Jaumouille V., Grinstein S. The cell biology of phagocytosis. Annu. Rev. Pathol. 2012;7:61–98. - PubMed
    1. Munthe-Kaas A.C., Kaplan G., Seljelid R. On the mechanism of internalization of opsonized particles by rat Kupffer cells in vitro. Exp. Cell Res. 1976;103:201–212. - PubMed
    1. Kaplan G. Differences in the mode of phagocytosis with Fc and C3 receptors in macrophages. Scand. J. Immunol. 1977;6:797–807. - PubMed

Publication types