Characterisation of innate fungal recognition in the lung

Abstract

The innate recognition of fungi by leukocytes is mediated by pattern recognition receptors (PRR), such as Dectin-1, and is thought to occur at the cell surface triggering intracellular signalling cascades which lead to the induction of protective host responses. In the lung, this recognition is aided by surfactant which also serves to maintain the balance between inflammation and pulmonary function, although the underlying mechanisms are unknown. Here we have explored pulmonary innate recognition of a variety of fungal particles, including zymosan, Candida albicans and Aspergillus fumigatus, and demonstrate that opsonisation with surfactant components can limit inflammation by reducing host-cell fungal interactions. However, we found that this opsonisation does not contribute directly to innate fungal recognition and that this process is mediated through non-opsonic PRRs, including Dectin-1. Moreover, we found that pulmonary inflammatory responses to resting Aspergillus conidia were initiated by these PRRs in acidified phagolysosomes, following the uptake of fungal particles by leukocytes. Our data therefore provides crucial new insights into the mechanisms by which surfactant can maintain pulmonary function in the face of microbial challenge, and defines the phagolysosome as a novel intracellular compartment involved in the innate sensing of extracellular pathogens in the lung.

Figure 1. Surfactant-mediated aggregation modulates the number of particle-cell contacts but does not influence non-opsonic recognition.

(A) Binding of unopsonised or surfactant-opsonised (BAL) zymosan to alveolar macrophages isolated from wild-type (black bars) or Dectin-1-deficient (white bars) mice, in the presence or absence of soluble β-glucan. Total fluorescence is represented as a percentage of fluorescent units (RFU), relative to the surfactant-opsonised zymosan sample. (B) Binding of surfactant-opsonised zymosan or unopsonised particles to peritoneal macrophages isolated from wild-type mice in the presence or absence of soluble β-glucans, as indicated. Recognition of serum opsonised zymosan is included for comparison. Total fluorescence is represented as a percentage of relative fluorescent units (RFU), normalized to the serum-opsonised sample. (C) Quantitation of binding of unopsonised or surfactant-opsonised zymosan to alveolar macrophages by microscopy. (D) FSC and SSC flow cytometric analysis of zymosan, BAL-opsonised zymosan, and BAL-opsonised zymosan in the presence of 100 mM EDTA. (E) Quantitation of the effects of surfactant opsonisation on the binding of A. fumigatus resting or swollen conidia, as indicated, to alveolar macrophages isolated from wild-type mice (black bars) or Dectin-1-deficient (white bars) mice by microscopy. (F) TNF production from alveolar macrophages isolated from wild-type (black bar) or Dectin-1-deficient (white bars) mice following the addition of unopsonised or surfactant-opsonised zymosan or resting A. fumigatus conidia, as indicated. (A)–(D) The data shown are the mean+SD and are representative of at least two independent experiments. (E)–(F) The data shown are mean+SEM of data pooled from two independent experiments. *p<0.05.

Figure 2. Inhibition of surfactant function exacerbates pulmonary inflammation in response to zymosan particles.

(A) Pulmonary histology following intra-tracheal administration of zymosan, zymosan plus EDTA, PBS or EDTA, stained with haematoxylin and eosin (H&E) or for the presence of zymosan (α-zymosan), as indicated. (B) Quantitation of neutrophil (CD11b⁺GR-1⁺) cells in the lungs of infected mice. (C) Production of selected cytokines and chemokines in the lungs of individual mice following intra-tracheal challenge with zymosan or zymosan plus EDTA. The data shown are representative of at least two independent experiments, except for (B) which is a single experiment. *p<0.05.

Figure 3. Inhibition of surfactant function has no effect on pulmonary inflammation in response to A. fumigatus resting conidia.

(A) Haematoxylin and eosin staining of pulmonary sections following intra-tracheal administration of A. fumigatus resting conidia in the presence or absence of EDTA. (B) Quantitation of neutrophil (CD11b⁺GR-1⁺) cells in the lungs of infected mice. (C) Production of selected cytokines and chemokines in the lungs of individual mice after intra-tracheal challenge with resting conidia or resting conidia plus EDTA. The data shown are representative of at least two independent experiments. *p<0.05.

Figure 4. Dectin-1 is only required for inflammatory responses to Aspergillus resting conidia in vivo.

(A) The Dectin-1 dependency of selected pulmonary cytokines and chemokines expressed as the fold change of wild type responses versus those obtained in Dectin-1 deficient mice, following the intra-tracheal administration of resting A. fumigatus conidia in the presence of EDTA. (B) The Dectin-1 dependency of selected pulmonary cytokines and chemokines expressed as the fold change of wild type responses versus those obtained in Dectin-1 deficient mice, following the intra-tracheal administration of zymosan in the presence of EDTA. (C) The Dectin-1 dependency of selected pulmonary cytokines and chemokines expressed as the fold change of wild type responses versus those obtained in Dectin-1 deficient mice, following the intra-tracheal administration of A. fumigatus swollen conidia. The data shown are the mean+SD of relative protein levels, determined by cytokine ELISA, and are representative of at least two independent experiments, except for (C) which is data from two pooled independent experiments. *p<0.05.

Figure 5. The induction of inflammatory responses to Aspergillus resting conidia is delayed and occurs after particle uptake.

(A) The production of TNF over time following the infection of thioglycollate-elicited or alveolar macrophages with zymosan particles (black circles) or resting Aspergillus conidia (black squares), as indicated. The data shown are the mean ± SD, and are representative of at least two independent experiments. *p<0.05. (B) Orthogonal confocal projections of CFSE-labelled thioglycollate-elicited macrophages (green) 3 and 6 hr after the uptake of resting Aspergillus conidia (labelled with calcofluor, red).

Figure 6. Phagosomes containing Aspergillus resting conidia mature normally into phagolysosomes.

(A) Immunofluorescent confocal analysis of Aspergillus conidia phagosomes in thioglycollate-elicited macrophages at 3 and 6 hr after infection, and examined for acquisition of LAMP-1, preloaded Dextran-red and fluorescence of lysotracker, indicating acidification. Arrowheads point to the location of conidia. Scale bar indicates 10 µm. (B) Kinetic analysis, determined by immunofluorescence microscopy, of the acquisition of Lamp-1 to conidial phagosomes in thioglycollate-elicited macrophages from wild type (black circles) and Dectin-1^−/− (white circles) mice. Data shown are mean ± SD. The data shown are representative of at least two independent experiments.

Figure 7. Phagolysosomes containing Aspergillus conidia acquire innate signalling components.

(A) Immunofluorescent confocal analysis of phagosomes containing Aspergillus conidia at 3 and 6 hr after infection of thioglycollate-elicited macrophages, showing the recruitment of Dectin-1, phosphoSyk, CARD9 or MyD88. Scale bar indicates 10 µm. Arrowheads point to the location of conidia; bold arrow indicates cytoplasmic straining. (B) Kinetic analysis of the acquisition of Dectin-1 to the conidial phagosomes in wild type and Dectin-1^−/− thioglycollate-elicited macrophages. (C) Flow cytometric analysis of the influence of pH on the ability of soluble Fc-Dectin-1 to bind to zymosan. (D) Kinetic analysis of the acquisition of phosphoSyk to conidial phagosome in wild type and Dectin-1^−/− thioglycollate-elicited macrophages. (E) Immunofluorescent confocal analysis of BAL-cell phagosomes 3 and 8 hr after i.t. infection of mice with resting Aspergillus conidia, showing the recruitment of Dectin-1 and phosphoSyk (pSyk) to the conidial phagosome in vivo. Scale bar indicates 10 µm. Arrowheads point to the location of conidia. The data shown are representative of at least two independent experiments.