Fig 1.
A pooled CRISPR screen identifies genes required for macrophage susceptibility to infection with Hc.
A. Diagram of screen approach. Cas9-expressing J774A.1 macrophage-like cells were transduced with a library of sgRNAs, challenged with Ura5-deficient Hc yeast, and subjected to 2–3 pulses of uracil treatment followed by recovery. sgRNAs amplified from Hc-infected and uninfected cells were deep-sequenced, and sequences were analyzed to identify guides that became enriched or depleted in the Hc-infected pool relative to the uninfected pool. B. Volcano plot showing the confidence score (casTLE score) versus the effect size (casTLE effect) for all genes. Genes that pass the 5% FDR cutoff are colored red, and genes individually validated in J774A.1 cells are labelled and colored in blue. C. Adjusted P-values for selected GO biological process annotations enriched in the screen hits. D. The 150 highest-scoring genes identified in the screen grouped based on their annotated function and localization in a cell, functional categories or complexes of genes are noted. Genes are colored according to their gene effect estimate, where yellow indicates enrichment in the Hc infected pool and blue indicates depletion.
Fig 2.
Identification of genes required for phagocytosis of yeast in J774A.1 cells and primary macrophages.
A. Diagram of approach used to individually validate the role of a gene in macrophage susceptibility to Hc infection. A mixture of WT (GFP-) and CRISPRKO (GFP+) J774A.1 cells were challenged with Hc yeast in the presence of uracil, and allowed to recover. Uninfected cells from the same mixture were passaged in parallel, and the percentage of mutant cells in the Hc infected pools was compared to that of the uninfected pools via flow cytometry (n = 3 biological replicates). B. Enrichment of gene-targeting guides in the Hc infected pool relative to the control pool, compared to that of non-targeting guides. C. Diagram of approach for determining the role of a gene in phagocytosis of Hc. A mixture of WT (GFP-) and CRISPRKO (GFP+) J774A.1 cells were infected with mCherry-expressing Hc yeast. Non-internalized yeasts were excluded using calcofluor white staining. Flow cytometry was used to determine the representation of mutant cells in the phagocytic compared to the non-phagocytic populations (n = 3). D. Identification of genes required for phagocytosis of yeast in J774A.1 cells using GFP expression to measure enrichment of sgRNA-expressing cells. E. Validation of gene involvement in BMDM phagocytosis of yeast using CRISPRKO BMDMs (Thy1.1+). A mixture of transduced (Thy1.1+) and untransduced (Thy1.1-) BMDMs were similarly infected with yeast and stained with calcofluor white and a Thy1.1 antibody to determine the representation of mutants in the phagocytic and non-phagocytic populations (n = 3 biological replicates).
Fig 3.
C3aR signaling plays a role in macrophage phagocytosis of fungi.
A. WT and C3ar-/- BMDMs were infected with live and PFA-killed mCherry-expressing Hc yeast (MOI2), and the phagocytosis rate was monitored over-time using flow-cytometry (n = 3 biological replicates). B. WT and C3ar-/- BMDMs were infected with FITC-labelled zymosan or mCherry-expressing Hc (MOI2) and the phagocytosis rate infected cells was monitored using flow cytometry (n = 3 biological replicates). C. BMDMs were infected with Candida albicans (Ca) (MOI3). Cells were imaged using confocal microscopy to quantify phagocytosis (n = 2 biological replicates, >350 cells/replicate). CFW staining was used to exclude extracellular Ca. D. BMDMs were infected with FITC-labelled Coccidioides posadasii (Cp) arthroconidia (MOI1), and extracellular conidia were labelled with calcofluor white. BMDM infection rates were determined using confocal microscopy (n = 3 biological replicates, 200–400 cells/rep). E. BMDMs were infected with FITC-labelled E. coli bioparticles (MOI4) and the E. coli-association with BMDMs was monitored via flow cytometry (n = 2 biological replicates). F. BMDMs were infected with 2 μm or 0.5 μm red fluorescent latex beads (MOI2), and the rate of BMDM association with the beads was measured using flow cytometry (n = 3 biological replicates). G. BMDMs were treated with a C3aR antagonist (1 μM SB290157) and infected with Hc yeast (MOI2). Phagocytosis was measured using flow cytometry (n = 3 biological replicates). H. BMDMs were pre-treated for 2 h with 1 μg/mL pertussis toxin (Ptx), which inhibits Gαi, and infected with Hc (MOI5, n = 3 biological replicates). I. BMDMs were pre-treated for 90 min with 10 μg/mL CD18 blocking antibody (GAME-46) and infected with Hc yeast (MOI5, n = 3 biological replicates) Phagocytosis was measured using flow cytometry. Emc1 is required for C3aR expression in BMDMs (J-L). J. Emc1 CRISPRKO BMDMs and control sgRNA transduced BMDMs, and C3aR levels were measured via flow cytometry following C3aR surface staining (n = 2 biological replicates). K. Histogram of C3aR levels in control and Emc1 CRISPRKO BMDMs. L. Frequency of C3aR+ cells in the indicated BMDMs. M. The mean fluorescence intensity (MFI) of the C3aR signal in the indicated BMDMs.
Fig 4.
Serum C3 promotes complement opsonization, C3a release, and macrophage phagocytosis of Hc yeast.
A. FBS stimulates macrophage phagocytosis of fungi in a C3aR-dependent manner. BMDMs were infected with mCherry-expressing Hc or FITC-labelled zymosan (30 min, MOI5) in the presence or absence of 20% heat-treated FBS (FBS). Phagocytosis was assessed via flow cytometry (n = 3 biological replicates). B. FBS does not promote macrophage phagocytosis of Hc via opsonization. Hc and zymosan particles were pre-incubated with 10% heat-treated FBS for 30 min at 37°C, washed, and used to infect BMDMs (2h, MOI2). Phagocytosis was measured using flow cytometry (n = 2 biological replicates). C-D. Prolonged or intense heat-treatment and zymosan treatment eliminates the phagocytosis-stimulating properties of serum. C. Macrophage phagocytosis of Hc (MOI5, 45 min, n = 3 biological replicates) was assessed in media supplemented with 10% FBS that had been subjected to heat treatment (C) at 56°C for up to 2h, at 65°C for 30 min, or that had been pre-treated with zymosan (D) (1X108 particles/mL, 60 min at 37°C). Phagocytosis was measured by flow cytometry. E. Normal mouse serum (NMS) stimulates macrophage phagocytosis of Hc in a C3-dependenent manner. BMDMs were infected with Hc yeast (MOI = 5, 60min) in serum-free media or media supplemented with 5% FBS, 5% NMS from WT mice, 5% NMS from C3-/- mice, or 5% heat-inactivated NMS (hiNMS) from WT mice and phagocytosis was measured by flow cytometry (n = 3 biological replicates). F. BMDMs in serum-free media were infected with Hc opsonized with 10% WT or C3-/- NMS. Phagocytosis was measured by flow cytometry (n = 3 biological replicates). G. C5-deficient serum promotes macrophage phagocytosis of Hc in a C3aR-dependant manner. BMDMs were infected with Hc yeast (MOI5) in media supplemented with 5% NMS from C57BL/6 mice or DBA2 (C5-deficient) mice. Phagocytosis was measured by flow cytometry (n = 2 biological replicates). H-J. Normal human serum (NHS) stimulates macrophage phagocytosis of Hc yeast. H. BMDMs were infected with Hc (MOI5, 60 min) in media supplemented with 5% untreated, heat-inactivated, or C3-depleted (C3d) NHS, and phagocytosis was monitored by flow cytometry (n = 3 biological replicates). I. Hc was opsonized with 10% untreated or C3d NHS, used to infect BMDMs in serum-free media (MOI5, 60 min), and phagocytosis was monitored by flow cytometry (n = 3 biological replicates) J. BMDMs were infected with Hc (MOI5) in media supplemented with 5% untreated or C5-depleted (C5d) NHS, and phagocytosis was monitored by flow cytometry (n = 3 biological replicates). K-L. Mouse serum promotes complement opsonization of yeast and release of C3a via multiple pathways. Hc was incubated in 10% serum from WT, C3-/-, or DBA2 mice for 30 min at 37°C. 10 mM EGTA or EDTA were added to the reactions to chelate Ca2+ or Mg2+, respectively. K. Supernatants were harvested following incubation, and mouse C3a levels were measured by ELISA (n = 3 biological replicates). L. Yeast were stained with a FITC conjugated anti-mouse C3, and imaged using confocal microscopy (representative slices are shown from 2 biological replicates).
Fig 5.
C3aR localizes to the early Hc-containing phagosome.
C3aR localizes to Hc-containing phagosomes (A) to a greater extent than latex bead-containing phagosomes (B). BMDMs were infected with the indicated particles (MOI = 5, n = 2 biological replicates per time point). Cells were then stained with a C3aR-specific antibody and imaged using optical sectioning with a confocal microscope. Representative images from a single slice are shown. C. Enlarged views of insets outlined in panels A and B by a white box. Scale bar = 20 μm. D. The mean fluorescence intensity of C3aR in the particle-containing phagosomes was quantified using ImageJ (N>91 phagosomes, **** p<0.0001, **p<0.01 by two-tailed Wilcoxon rank-sum test). The line represents the median phagosomal C3aR intensity.
Fig 6.
C3aR promotes the formation of actin-rich protrusions that facilitate capture of Hc yeast.
J774A.1 cells were engineered to express Lifeact-mEGFP to label F-actin, co-cultured with mCherry-expressing Hc yeast, and subjected to live-cell confocal microscopy in a temperature-and-CO2 controlled chamber in media supplemented with 10% FBS. Cells were treated with a C3aR antagonist (10 μM SB290157) or a vehicle control. A. Representative images from a confocal time series (S1 Movie) showing a macrophage extending an F-actin-rich protrusion towards an mCherry expressing Hc yeast, followed by phagocytosis and formation of an actin-rich phagosome. The corresponding DIC images are shown below. B. A similar time series (S3 Movie) of macrophages treated with SB290157 showing a failure to initiate formation of a membrane protrusion and much slower capture of Hc yeast. Scale bar = 20 μm. The movement of membrane structures that successfully caputured yeast were analyzed using MtrackJ to quantify the behaviors of these structures (C-E) (n = 2 biological replicates, >50 tracks per replicate), including the phagocytosis rate, quantified as the time required for the macrophage to successfully engulf the yeast divided by the distance of the yeast to the macrophage at the start of the series (C), the mean velocity of the membrane structure closest to the yeast (D), and the outreach ratio quantified as the max displacement of the track divided by the length of the track (E) (**** p<0.0001 by two-tailed Wilcoxon rank sum test). The line represents the median measurement.
Fig 7.
C3 and C3aR-deficiency do not dramatically alter mouse susceptibility to Hc infection.
A-B C3ar-/- mice (n≥10) and age-matched WT C57BL/6 mice (n≥10) were infected intranasally with varying doses of Hc yeast to initiate either a sub-lethal (A) or lethal (B) infection. D. C3-/- mice and age-matched WT mice were infected intranasally with a sub-lethal dose of Hc yeast. Susceptibility is illustrated by a Kaplan-Meier survival curve. ns = not significant, *p < 0.05 by logrank test. C,E. The indicated mice were infected with a sub-lethal dose of Hc. The fungal burden in lung and spleen homogenates was determined by enumeration of colony forming units (CFUs) at the indicated time points (n≥5). X-axis label for C is the same as that indicated for E.
Fig 8.
Model for the role of complement and C3aR in macrophage recognition of Hc yeast.
We propose the following model for the role of complement and C3aR in macrophage recognition of Hc: C3, derived from serum, reacts with the Hc cell-wall, leading to C3b/iC3b deposition on the cell-wall, and release of C3a, which diffuses away from the yeast surface leading to a concentration gradient emanating from the yeast cell-wall. C3a activates C3aR, which signals through Gαi and Gβ2 to promote the formation and directional movement of actin-rich membrane protrusions, and possibly to promote local activation or increased motility of the integrin receptor CR3. Active CR3 can then recognize C3b/iC3b or other features of the Hc cell-wall. C3aR and/or CR3 activation then coordinates actin polymerization and phagocytic cup formation by regulating the activity of actin polymerization regulators Arp2/3 and SCAR/WAVE. In the presence of C5-containing serum, the C5 convertase can similarly catalyze the cleavage of C5 at the fungal surface, leading to release of C5a and activation of C5aR, which may also drive local chemotaxis and activation of phagocytic integrins to promote phagocytosis.