Iron restriction inside macrophages regulates pulmonary host defense against Rhizopus species

Abstract

Mucormycosis is a life-threatening respiratory fungal infection predominantly caused by Rhizopus species. Mucormycosis has incompletely understood pathogenesis, particularly how abnormalities in iron metabolism compromise immune responses. Here we show how, as opposed to other filamentous fungi, Rhizopus spp. establish intracellular persistence inside alveolar macrophages (AMs). Mechanistically, lack of intracellular swelling of Rhizopus conidia results in surface retention of melanin, which induces phagosome maturation arrest through inhibition of LC3-associated phagocytosis. Intracellular inhibition of Rhizopus is an important effector mechanism, as infection of immunocompetent mice with swollen conidia, which evade phagocytosis, results in acute lethality. Concordantly, AM depletion markedly increases susceptibility to mucormycosis. Host and pathogen transcriptomics, iron supplementation studies, and genetic manipulation of iron assimilation of fungal pathways demonstrate that iron restriction inside macrophages regulates immunity against Rhizopus. Our findings shed light on the pathogenetic mechanisms of mucormycosis and reveal the role of macrophage-mediated nutritional immunity against filamentous fungi.

Fig. 1

Persistence of Rhizopus conidia inside alveolar macrophages (AMs). a Fungal loads in lungs of immunocompetent C57BL/6 (B6) mice (n = 3 per group) infected via intratracheal administration of a standardized inoculum (5 × 10⁶ conidia per mice) of A. fumigatus, R. oryzae, or R. delemar. ***P < 0.0001, Mann–Whitney test. b Survival of immunocompetent C57BL/6 (B6) mice (n = 8 per group) infected as in a, with either A. fumigatus, R. oryzae, or R. delemar. c Representative photomicrographs of the lungs from mice infected as in b with either A. fumigatus or R. oryzae and sacrificed on day 5. Histopathological sections were stained with Grocott methenamine silver (GMS; left panels) or hematoxylin and eosin (H&E; right panels). The presence of R. oryzae conidia (black color) in the lungs is shown by GME stain. Original magnification ×400. d Representative photomicrographs of the lungs from mice infected as in b, sacrificed on day 5. Lungs were stained by IHC for CD68 or CD11 and counterstained with hematoxylin and PAS. There is evidence of extracellular R. oryzae conidia surrounded by neutrophils (top panel), and intracellular R. oryzae conidia inside AMs. e FACS analysis of total number of professional phagocytes in the lungs of immunocompetent mice (n = 3 per group) infected with R. oryzae as in a, assessed on days 1 and 5. The gating strategy for identification of neutrophils, monocytes, AMs, interstitial macrophages (IMs), and dendritic cells (DCs) is shown in Supplementary Fig. 1. f FACS analysis of association of labeled (Fluorescent Brightener 28; CW) conidia of R. oryzae with professional phagocytes of mice infected as in a, assessed on days 1 and 5. g In vivo phagocytosis rates of R. oryzae conidia on day 5 of infection of mice infected as in a. ***P < 0.0001 Mann–Whitney test. h Representative confocal image of sorted AM from mice infected as in a with fluorescent-labeled, live conidia of R. oryzae (day 1), fixed and stained with Cathepsin D. Cross-section analysis was performed to discriminate intracellular conidia from conidia associated/bound to the cell surface of AM

Fig. 2

Rhizopus conidia display resistance to ex vivo killing by macrophages. a–c Comparative studies on phagocytosis of A. fumigatus and R. oryzae conidia by BMDMs and neutrophils (PMNs) assessed by confocal imaging. Data on quantification of phagocytosis are presented as mean ± SEM of five independent experiments ***P < 0.0001, **P < 0.001, Mann–Whitney test. d, e Intracellular killing of A. fumigatus and R. oryzae (Mucorales) conidia by BMDMs. Symbols connected with a line represent time points of the same independent experiment (n = 9 per group). ***P < 0.0001, Mann–Whitney test. f, g Assessment of in vitro susceptibility of A. fumigatus and R. oryzae conidia to f oxidative damage induced by increasing concentrations of H₂O₂ or g to damage induced by enzymatic activity of increasing concentrations of lysosomal extracts of BMDMs, assessed by measurement of fungal metabolic activity using the XTT assay at 24 h. h In vitro fungicidal activity of increasing concentrations of lysosomal extracts against conidia of A. fumigatus or R. oryzae assessed by CFU plating. i Induction of apoptosis in unstimulated BMDMs or BMDMs infected with A. fumigatus or R. oryzae at an MOI of 3:1 (effector:fungal cells) for 6 h. Apoptotic BMDMs were assessed by FACS analysis following Annexin V/PI staining. Data are representative of one out of three independent experiments. NS not significant

Fig. 3

Rhizopus conidia induce phagosome maturation arrest in BMDMs. a, b BMDMs from GFP-LC3 mice were infected at different time points with live conidia of A. fumigatus or R. oryzae at an MOI 3:1 (effector:fungal cells). At the indicated time point, cells were fixed and analyzed by confocal microscopy. Representative fluorescence images are presented in a. Bar, 5 µm. Data on quantification of LC3⁺ phagosomes are presented as mean ± SEM of five independent experiments in b. c BMDMs from C57BL/6 (B6) mice were infected as in a. Cells were fixed at the indicated time point and stained for Rab5B. Data on quantification of Rab5⁺ phagosomes are presented as mean ± SEM of three independent experiments. d, e BMDMs were pre-loaded with FITC-Dextran, infected as in a, and phagolysosomal fusion was assessed at the indicated time point based on acquisition of FITC-Dextran in the phagosome. Representative fluorescence images are shown in d. Bar, 5 µm. Data on quantification of FITC-Dextran⁺ phagosomes are presented in e as mean ± SEM of three independent experiments. f, g BMDMs were stimulated as in a, cells were fixed and stained for the lysosomal protein marker Cathepsin D. Data on quantification of Cathepsin D⁺ phagosomes are presented as mean ± SEM of three independent experiments in f, while representative fluorescence images are shown in g. Bar, 5 µm. h Representative electron microscopy of acid phosphatase, a lysosomal enzyme marker of phagolysosomal fusion (shown as dark color on the phagosome membrane) in 15 min and 4 h phagosomes containing A. fumigatus or R. oryzae conidia. ***P < 0.0001, **P < 0.01, Mann–Whitney test. Bar, left upper and lower panels, 1 µm; right upper panel, 0.5 µm; right lower panel, 2 µm

Fig. 4

Rhizopus inhibits phagosome biogenesis in AMs during in vivo infection. GFP-LC3 (a, b) or C57BL/6 (B6) (c, d) mice were infected intratracheally with 5 × 10⁶ conidia of either R. oryzae or A. fumigatus. AMs were obtained by bronchoalveolar lavage at the indicated time point, fixed, stained with anti-GFP (a) or Cathepsin D (b) antibodies, and assessed by confocal imaging. Representative fluorescence images are shown (a, c). Data on quantification of GFP-LC3⁺ (b) and Cathepsin D⁺ (d) phagosomes are presented as mean ± SEM of three independent experiments. ***P < 0.0001, **P < 0.001 Mann–Whitney test. Scale bar, 5 μm

Fig. 5

Rhizopus melanin blocks early event in phagosome biogenesis. a GFP-LC3 BMDMs were infected at 1 or 24 h with live conidia of A. fumigatus or R. oryzae at an MOI 1:1. Cells were fixed and the conidial diameters of intracellular conidia was measured by confocal microscopy. Data on quantification of conidial diameter are presented from one out of three independent experiments. Each symbol represents the value of maximum diameter of individual fungal cell and horizontal bars represent the mean diameter. ***P < 0.0001, Mann–Whitney test. b Representative DIC images from a are shown. c X-band room temperature EPR spectra of purified melanin obtained from A. fumigatus, R. oryzae, or synthetic melanin are shown. d BMDMs pre-loaded with FITC-Dextran were stimulated with conidia of A. fumigatus or R. oryzae or melanin ghosts (purified melanin particles) obtained from conidia of the indicated A. fumigatus or R. oryzae strains at an MOI of 3:1 (effector:fungal conidia). Cells were removed at 4 h and assessed by confocal imaging. Data on quantification of FITC-Dextran⁺ phagosomes are presented as mean ± SEM of three independent experiments. e, h BMDMs from GFP-LC3 (e, f) or C57BL/6 (B6) (g, h) mice were infected with R. oryzae conidia (WT conidia) or R. oryzae conidia following chemical degradation of melanin with H₂O₂ bleaching (albino conidia) at an MOI of 3:1 (effector:fungal cells). Cells were removed at 1 h of infection, fixed, stained, and analyzed by confocal imaging. Data on quantification of GFP-LC3⁺ (f) or Cathepsin D⁺ (h) phagosomes are presented as mean ± SEM of three independent experiments. e, g Representative fluorescent images from experiments on f and h are shown. i, j Fungal loads from lungs of immunocompetent mice (n = 9 per experimental group) infected with 10⁶ conidia of R. delemar grown in regular media (WT Rhizopus) or under conditions of copper starvation to inhibit melanization (albino Rhizopus) (i). Mice (n = 3 per condition) were sacrificed at the indicated time points, lungs were homogenized, and fungal loads were assessed by CFU plating (j). ***P < 0.0001, **P < 0.001 Mann–Whitney test. Scale bar, 10 μm

Fig. 6

AMs have a non-redundant role in immunity against Rhizopus. a GFP-LC3 BMDMs were infected with dormant or swollen conidia of R. oryzae at an MOI of 3:1 (effector:fungal cells) and phagocytosis was assessed at 4 h. Data on quantification of phagocytosis of Rhizopus conidia are presented as mean ± SEM of three independent experiments. ***P < 0.0001, Mann–Whitney test. b Survival of immunocompetent C57BL/6 (B6) mice (n = 8 per group) infected via intratracheal administration of a standardized inoculum (5 × 10⁶ conidia per mice) of dormant or swollen conidia of R. oryzae. c Representative photomicrographs of the lungs from mice infected with 5 × 10⁶ swollen conidia of R. oryzae at 24 h of infection. Lungs were stained by H&E or IHC for CD11 and counterstaining with hematoxylin and PAS. d Representative histopathology from lungs of C57BL/6 (B6) mice following intratracheal administration of 100 μl of clodronate liposomes or control liposomes and 48 h later infection with 10⁷ R. oryzae conidia. Invasive hyphal growth is present in the lungs of clodronate liposome group of mice. c, d Original magnification ×400. e Survival of C57BL/6 (B6) mice treated with clodronate liposomes (n = 10) or control liposomes (n = 15) and infected as in d. f Fungal loads in the lungs of C57BL/6 (B6) mice treated with clodronate liposomes or control liposomes and infected as in d. Lungs were removed on day 2 of infection and fungal loads were assessed by CFU plating. **P < 0.001, Mann–Whitney test. g CD11c-DTR mice were intranasally inoculated with 40 µl PBS or with 20 ng/g of body weight of diphtheria toxin in 40 µl PBS and analyzed for the presence of CD11c⁺ cells (AMs and DCs) at 24 h. Results represent two to three experiments with two to three mice per group per experiment. h Survival of CD11c-DTR mice (n = 5 per group) treated with DT (20 ng/kg of mice) or PBS (control) and infected 48 h later with 5 × 10⁶ conidia of R. delemar. The nonparametric log-rank test was used to determine differences in survival times

Fig. 7

Global analysis of differential gene expression during infection. RNA-seq-based expression analysis of iron-related a R. delemar genes and b mouse genes following in vitro infection of BMDMs. a Each column represents an individual sample (biological triplicates of four different conditions; n = 12). Log-transformed absolute expression normalized across all samples. Red indicates high gene expression. Blue indicates low gene expression. For the 0 h column, R. delemar was incubated in tissue culture media without BMDMs for 1 min. b Values are presented as in a and represent the average of three biological replicates for each condition. For the 0 h columns, BMDMs were incubated in culture media in the absence of R. delemar spores for 1 hr

Fig. 8

Macrophages inhibit intracellular growth of Rhizopus via iron starvation. a, b BMDMs from GFP-LC3 mice were infected with WT Rhizopus delemar strain at an MOI of 1:2 (effector:fungal cells) in regular culture media or media containing DFO (100 μM), iron (FeCl₃; 100 μM), or DFO plus iron. At 1 h of infection, cells were extensively washed to remove non-phagocytosed conidia, media were replaced, and intracellular germination was assessed at 12 h by confocal imaging. BMDMs from GFP-LC3 mice were also infected with R. delemar-attenuated mutants for FTR1 (c, d) or FOB1/2 (e, f), which are defective in the high affinity iron permease expression regulating iron assimilation under conditions of limited iron availability and DFO receptor FOB1/2 mediating fungal iron uptake from DFO, respectively, as in a. The attenuated mutants were derived from the wild-type strain used in a, b by RNAi^, . Representative immunofluorescence images are shown in a, c, and f. Data on quantification of germination of intracellular conidia of the indicated Rhizopus strain are presented as mean ± SEM of three independent experiments. ***P < 0.0001, *P < 0.01, Mann–Whitney test. Bar, 5 µm

Fig. 9

Invasive pulmonary mucormycosis in a patient with intracellular persistence (a). A 60-year-old white male with relapsed acute myelogenous leukemia received salvage chemotherapy with fludarabine and high-hose cytarabine (FLAG) in August 2017. On September 2017, the patient developed febrile neutropenia, left-sided chest pain, and evidence of necrotizing pneumonia with chest wall myositis in computer tomography of the chest. He received treatment with broad-spectrum antibiotics and anidulafungin. Despite neutrophil recovery, the infection progressed with infiltration of the abdominal wall and the spleen. In tissue biopsy mucormycosis was diagnosed based on characteristic histopathological findings (b, c). The patient was started on liposomal amphotericin B and radical surgery with splenectomy was performed in November 2017. In histopathological analysis of muscle biopsy, there is evidence of extensive tissue necrosis and growth of Mucorales hyphae inside blood vessels (b, c). In spleen histopathology there are areas of granulomatous lesions containing macrophages and multinucleated giant cells (d). e In higher magnification (d, inset), there is evidence of a Mucorales conidium (spore) inside a giant cell (arrow). Original magnifications ×200, ×800. Eosin and hematoxylin stain; white arrows: spores; black arrows: hyphae

Fig. 10

Proposed model of nutritional immunity inside macrophages against Mucorales. Following inhalation, Mucorales conidia are predominantly phagocytosed by alveolar macrophages (AMs). Intracellular conidia of Mucorales remain dormant and establish prolonged persistence inside the nascent phagosome of AMs. Surface retention of cell wall melanin in intracellular conidia of Mucorales blocks phagosome biogenesis and LAP, impedes killing, and induces anti-apoptotic signaling in macrophages to establish persistence. Inhibition of intracellular germination of Mucorales conidia via iron restriction is a central host defense mechanism against mucormycosis. In addition, incompletely understood nutritional immunity mechanisms, including transferrin-mediated restriction of free iron availability in serum, inhibit germination of extracellular conidia. In parallel, rapid recruitment of neutrophils results in clearance of extracellular conidia of Mucorales. Quantitative and qualitative defects in innate immunity associated with prolonged chemotherapy-induced neutropenia, and/or corticosteroid-induced immunosuppression compromise the ability of phagocytes to inhibit germination of intracellular or extracellular conidia and result in invasive fungal growth. On the other site, failure of nutritional immunity mechanisms in patients with abnormalities on iron metabolism (e.g., diabetic acidosis) allow germination of intracellular or extracellular conidia and promote invasive tissue growth. Mucorales responses inside AMs are highlighted in blue. Nutritional immunity responses are highlighted in red