Glutamine Metabolism Supports the Functional Activity of Immune Cells against Aspergillus fumigatus

Our first aim was to investigate the modulation of glutamine metabolism in macrophages in response to A. fumigatus infection. Using available transcriptomic data from human macrophages infected with A. fumigatus (11), we started by evaluating whether the expression of genes involved in glutamine metabolism was modulated in response to infection. Principal-component analysis (PCA) of glutamine metabolism genes revealed that the expression profile of infected macrophages is different from noninfected controls (Fig. 1A). Likewise, unsupervised hierarchical clustering reveals that infected macrophages cluster separately from noninfected cells (Fig. 1B). Glutaminase (GLS), the amino acid transporter SLC7A5 (solute carrier family 7 member 5), and glutamine-fructose-6-phosphate transaminase 2 (GFPT2) were among the genes that were upregulated in the infected macrophages. The transcriptional induction of glutaminolysis was confirmed by the rapid glutamine consumption observed early after infection of macrophages relatively to noninfected cells (Fig. 1C). Given the contribution of glutamine metabolism to the replenishment of the TCA cycle, we next assessed the intracellular levels of several TCA intermediates using liquid chromatography tandem-mass spectrometry (LC-MS/MS) after infection of macrophages. Our results reveal a significant increase in TCA intermediates derived from glutamine metabolism, namely, α-ketoglutarate, succinate, and fumarate (Fig. 1D). Although they did not reach statistical significance, glutamate concentrations also tended to be higher in infected macrophages relative to noninfected controls.

The production of proinflammatory cytokines, phagocytosis, and fungal clearance are among the most relevant effector functions required for protective immune responses to A. fumigatus (22). We tested, therefore, how the depletion of glutamine could affect these processes during fungal infection. We confirmed that glutamine depletion resulted in a notable impairment in the production of proinflammatory cytokines such as IL-1β, IL-6, and TNF by macrophages (Fig. 2A), as well as the production of IFN-γ and IL-22 by PBMCs (Fig. 2B). To investigate whether the decreased cytokine production upon glutamine deprivation was associated with altered catabolism to lactate, we measured the concentrations of secreted lactate in culture supernatants. Our results show that lactate secretion following glutamine deprivation is impaired after infection compared to noninfected controls (Fig. 2C), suggesting that glutamine metabolism could also be fueling lactate production upon fungal infection.

Under glutamine-deprived conditions, macrophages also displayed an impaired phagocytic ability (Fig. 2D), a finding supported by data showing that phagocytosis is inhibited upon oxidative phosphorylation inhibition (13). In contrast, the conidiacidal activity was not compromised by depletion of glutamine (Fig. 2E), and, in line with this, no major alterations in ROS production were observed (Fig. 2F). Of note, glutamine supplementation failed to improve the overall antifungal effector capacity of macrophages (Fig. 2A to F).

To identify the steps in glutamine metabolism that were required for the effector mechanisms of macrophages, we resorted to specific pharmacologic inhibitors (Fig. 3). Treatment of macrophages with GPNA, a competitive inhibitor of glutamine uptake by the SLC1A5 transporter, decreased the production of IL-1β and TNF, but not IL-6, which was instead increased, in infected macrophages (Fig. 4A). The production of both IL-22 and IFN-γ was also compromised after inhibition of glutamine uptake in PBMCs (Fig. 4B).

Next, we assessed the effect of the inhibition of glutaminase, an enzyme that catalyzes the first step of glutaminolysis and converts glutamine into glutamate, using two different compounds: DON, a broad-spectrum glutaminase inhibitor, and BPTES, a specific inhibitor of mitochondrial glutaminase. These treatments promoted a similar defect in cytokine production in infected macrophages as described above, characterized by lower production of IL-1β and TNF, but not IL-6 secretion, whose levels were instead increased, similar to the GPNA treatment (Fig. 4A). The amounts of IL-22 and IFN-γ produced by PBMCs were also lower after the pharmacological inhibition of glutaminase activity (Fig. 4B).

Given that our results demonstrated a decreased lactate production in glutamine-depleted condition (Fig. 2C), we next sought to validate this finding using the inhibitors described above. Under these conditions, the production of lactate was compromised after the inhibition of either the transporter or glutaminase activity (Fig. 4C). The ability of macrophages to phagocytose the fungus was also impaired in all tested conditions (Fig. 4D), suggesting the need for both glutamine import and catabolism for the proper internalization of the fungus. Similar to what was observed in glutamine-deprived conditions, no differences were observed regarding the conidiacidal activity of macrophages treated with all inhibitors (Fig. 4E). Despite these observations, transporter inhibition in infected macrophages resulted in a decreased production of cellular ROS. In contrast, treatment with BPTES resulted in defective production of mitochondrial ROS (Fig. 4F), a finding in line with differential requirements for glutamine metabolism to ROS production.

In light of the data demonstrating a pivotal role for macrophage glutamine metabolism for antifungal host defense, we next sought to validate this requirement in an in vivo model of pulmonary aspergillosis. Immunocompetent C57BL/6 mice were treated with BPTES or PBS prior to intranasal A. fumigatus infection. In support of the activation of glutamine metabolism in response to infection, we detected a significant induction of genes involved in glutamine import and utilization, such as Gls and Slc1a5, in the lungs of mice after A. fumigatus infection (Fig. 5A). Inhibition of glutamine metabolism rendered mice more susceptible to infection, as revealed by the increased fungal burden (Fig. 5B) and inflammatory pathology in the lungs (Fig. 5C). In accordance with the in vitro data, production of both proinflammatory and T-cell-mediated cytokines in lung homogenates of infected mice subjected to glutaminolysis inhibition were also lower than in noninfected controls under the same conditions (Fig. 5D). In line with these findings, the concentrations of lactate in the lungs of infected mice were also lower relatively to noninfected animals (Fig. 5E). On the other hand, glutamine supplementation failed to improve the overall ability of mice to deal with A. fumigatus infection (Fig. 5B to E): this argues that if cellular capacity to use glutamine is normal, additional supplementation does not further improve the outcome.

We next examined whether genetic variation in the genes involved in the glutamine pathway could affect the host response to A. fumigatus. To do so, we tested the association of common single nucleotide polymorphism (SNPs) (minor allele frequency, ≥5%) in relevant candidate genes with differential cytokine production after fungal stimulation of PBMCs from healthy individuals of the 500FG cohort in the Human Functional Genomics Project (HFGP) (23). We identified a total of 12 cytokine quantitative trait loci (cQTLs) in genes involved in glutamine metabolism (Table 1). Several suggestive cQTLs with a nominal P value < 0.05 were identified, particularly in the gene encoding the glutamine transporter SLC7A5, with the rs11117336 SNP influencing the production of both IFN-γ and IL-22 by human PBMCs in response to A. fumigatus stimulation.

We next assessed the impact of the identified cQTLs on gene expression, using a publicly available meta-analysis of expression QTLs (eQTLs) in nontransformed peripheral blood samples available from the eQTLGen Consortium (24). Importantly, all of the identified cQTLs resulted being also known eQTLs (Table 1), therefore confirming that these variants can regulate both gene expression and cytokine production. Collectively, our data demonstrated that glutamine metabolism is an essential process required for the host defense against A. fumigatus infection.

The functional experiments involving cells isolated from the peripheral blood of healthy volunteers at the Hospital of Braga, Portugal, were approved by the Ethics Subcommittee for Life and Health Sciences (SECVS) of the University of Minho, Portugal (no. 014/015). The Human Functional Genomics Project (HFGP) study was approved by the Ethical Committee of Radboud University Nijmegen, the Netherlands (no. 42561.091.12). Experiments were conducted according to the principles expressed in the Declaration of Helsinki, and participants provided written informed consent.

C57BL/6 mice purchased from Charles River Laboratories were bred under specific pathogen-free conditions and kept at the animal facilities at the Life and Health Sciences Research Institute (ICVS). Mice were fed ad libitum and maintained under light/dark cycles of 12 h, temperature of 18–25°C, and humidity 40–60%. Animal experimentation was performed on eight to twelve-week-old gender-matched mice, following biosafety level 2 (BSL-2) protocols approved by the Institutional Animal Care and Use Committee (IACUC) of the University of Minho, and both ethical and regulatory approvals were consented by CEICVS (no.074/016). All procedures followed the EU-adopted regulations (Directive 2010/63/EU) and were conducted according to the guidelines of Direção-Geral de Alimentação e Veterinária (DGAV).

A. fumigatus was grown on 2% malt extract agar for 7 days at 28°C. The conidia were harvested from agar plates using phosphate buffer saline (PBS) (Gibco) with 0.05% Tween 20 (Sigma-Aldrich), followed by gentle agitation and subsequent filtration through a 40-μm-pore size cell strainer (Falcon). The concentration of conidia/mL was determined by counting in a Neubauer chamber.

Peripheral blood mononuclear cells (PBMCs) were enriched from buffy coats from healthy donors by density gradient using Histopaque-1077 (Sigma-Aldrich), washed twice with PBS (Gibco), and resuspended in RPMI 1640 culture medium with 2 mM glutamine (Thermo Fisher) supplemented with 10% human serum (Sigma-Aldrich), 10 U/mL penicillin/streptomycin and 10 mM HEPES (Thermo Fisher). Monocytes were isolated from PBMCs by resorting to a positive magnetic bead separation with anti-CD14⁺-coated beads (Miltenyi) according to the manufacturer’s instructions. To evaluate the purity of the isolated monocytes, 5 × 10⁵ cells were stained with BV 650 anti-human CD14 (clone M5E2) antibody for 30 min at 4°C. Cell viability was assessed by staining with Zombie Green fluorescent dye (BioLegend) for 30 min at 4°C. Pellets were washed and resuspended in FACS buffer (PBS containing 2% FBS and 2 mM EDTA) prior to analysis. Data were obtained on a BD FACS LSRII instrument (Becton, Dickinson) and processed using FlowJo (Tree Star Inc.). The obtained CD14⁺ population displayed a purity higher than 94%, and with more than 97% viable cells. Isolated monocytes were then resuspended in a complete RPMI medium and seeded at a concentration of 1 × 10⁶ cells/mL in 24-well or 96-well plates (Corning) and left for 7 days with 20 ng/mL of recombinant human granulocyte-macrophage colony-stimulating factor (GM-CSF, Miltenyi) at 37°C and 5% CO₂. The culture medium was replaced every 3 days, and acquisition of macrophage morphology was confirmed by visualization in a BX61 microscope (Olympus).

Unless otherwise indicated, cells (5× 10⁵/well in 24-well plates) were left noninfected or infected with live conidia of A. fumigatus at a 1:10 effector-to-target ratio for 24 h at 37°C and 5% CO₂. The experiments involving glutamine depletion or supplementation used RPMI 1640 medium without glutamine and glutamine (Thermo Fisher). To interfere with glutamine metabolism, macrophages were pretreated for 1 h with 600 μM L-γ-Glutamyl-p-nitroanilide (GPNA), 10 μM 6-Diazo-5-oxo-l-norleucine (DON), and 50 μM Bis-2-(5-phenylacetamido-1,3,4-thiadiazol-2-yl)ethyl sulfide (BPTES) (all from Sigma-Aldrich). Viability and expression of surface markers (HLA-DR, CD86, CD163, and CD206) were evaluated in all experimental conditions to test for population homogeneity. Macrophages were viable under all conditions tested. Data were assessed in triplicates or quadruplicates and are shown as the mean value for each individual.

RNA sequencing was performed as previously described (11). Briefly, macrophages (5 × 10⁵ cells/mL in 24-well plate) were left noninfected or infected with live conidia of A. fumigatus at a 1:2 effector-to-target ratio for 2 h. Samples were processed, sequencing, and analyzed at IMGM Laboratories GmbH (Germany). Data have been deposited in Gene Expression Omnibus (GEO) with the accession code GSE128661. To better visualize the most significantly represented genes of a specific functional class, heatmaps were created using ClustVis: a web tool for visualizing the clustering of multivariate data (http://biit.cs.ut.ee/clustvis/). PCA for this data set was conducted using R software v4.2.0.

For the quantification of glutamine, macrophages (5 × 10⁵/well in 24-well plates) were infected for 15 min, 30 min, 1h, 2h, 6h, and 24h at 37°C for 5% CO₂ with live A. fumigatus conidia at a 1:10 effector-to-target ratio. After infection, supernatants were collected and centrifuged, and glutamine levels were quantified using the Glutamine and Glutamate Determination kit (Sigma-Aldrich), following the manufacturer’s instructions.

Macrophages (5 × 10⁵ cells/mL in 24-well plate) were infected with A. fumigatus conidia at a 1:2 effector-to-target ratio for 6 h at 37°C in 5% CO₂. Cells were detached with accutase (GRiSP), and the resulting cell suspensions were pooled from three different wells. The resulting cell suspensions were centrifuged at 4°C, the supernatant discarded, and the resulting pellet was immediately frozen in liquid nitrogen to quickly quench the cell metabolism. The analysis of the metabolites was performed with liquid chromatography-mass spectrometry (LC-MS/MS). The analysis was performed in an Agilent 1290 Infinity using a 1290 Infinity Binary Pump (1200 bar) and a 1260 Infinity Quaternary Pump (400 bar). The LC system was coupled to an Agilent 6470 triple-quadrupole mass spectrometer using an electrospray ionization (ESI) interface working in multiple reaction monitoring (MRM) mode. The chromatographic method used HILIC interaction chromatography (Agilent InfinityLab Poroshell). MRM transitions were chosen according to the Metabolomics dMRM library and method from Agilent Technologies (p/n G6412AA).

For the in vivo studies, from day 0 (day of infection) and until the end of the experimental protocol, 500 mg/kg of glutamine and 150 μg/Kg BPTES were administered to mice daily via oral gavage or via the intraperitoneal route (i.p.), respectively. Control groups consisted of mice to which an equal volume of vehicle (sterile PBS) was administered. On day 0, mice were challenged with 1 × 10⁸ live conidia of A. fumigatus (A1163 strain) by a noninvasive intranasal (i.n.) infection procedure upon anesthesia with 5 mg/Kg of ketamine (Ketamidor, Ritcher Pharma) and 1 mg/Kg of medetomidine (Domtor, Ecuphar), and weight and well-being were monitored during the infection period. At 3-days postinfection, mice were sacrificed, and the lungs were PBS-perfused and excised. To assess the fungal burden, the right lobes were removed, weighted, and homogenized in 1 mL of sterile PBS with a tissue homogenizer (Glas-Col). The resulting homogenate was serially diluted and plated on solid growth media. The resulting CFU were counted, and the fungal burden was calculated log₁₀ CFU/g of lung.

For the histological analysis, at 3-days postinfection, lungs were perfused with PBS, excised, and fixed in buffered formalin solution 10% (Sigma-Aldrich) for 48 h under agitation at room temperature. Lungs were then processed, paraffin-embedded, and lung sections of 5 μm thickness were stained with hematoxylin and eosin (H&E) to assess inflammation. The images were acquired on a BX61 widefield upright microscope using a DP70 high-resolution camera (Olympus), and the area of inflammation was determined using ImageJ software v1.50i (Fiji).

To evaluate phagocytosis, macrophages (5 × 10⁵/well in 24-well plates) were infected with fluorescein isothiocyanate (FITC)-labeled conidia of A. fumigatus at a 1:5 effector-to-target ratio. The incubation period was synchronized for 30 min at 4°C, and phagocytosis was initiated by shifting the coincubation to 37°C and 5% CO₂ for 1 h. Phagocytosis was stopped by washing wells with ice-cold PBS, and extracellular conidia were stained with 0.25 mg/mL Calcofluor White (Sigma-Aldrich) for 15 min at 4°C to avoid further ingestion. The cells were detached with accutase and resuspended in PBS. The percentage of macrophages with ingested green conidia was collected on a BD FACS LSRII instrument (Becton, Dickinson) and processed using FlowJo v10.5.3 (Tree Star Inc.). Unstained samples were used to set voltages.

Macrophages (1 × 10⁵/well in 96-well plates) were infected with A. fumigatus conidia at a 10:1 effector-to-target ratio. To allow the internalization of conidia, cells were incubated for 1 h at 37°C and 5% CO₂. Medium containing the noningested conidia was removed, and cells were washed twice with prewarmed PBS. To measure the conidiacidal activity, macrophages were allowed to eliminate the internalized conidia for 2 h at 37°C in 5% CO₂. After incubation, culture plates were snap-frozen at −80°C and thawed at 37°C to cause the cell lysis and release of internalized conidia. Serial dilutions of cell lysates were plated on solid growth media and incubated at 37°C for 24 h. The number of CFU was counted, and the percentage of eliminated conidia was calculated.

Macrophages (5 × 10⁵/well in 24-well plates) were infected for 4 h at 37°C in 5% CO₂ with live A. fumigatus conidia at a 1:5 effector-to-target ratio. The supernatant was removed, and cells were detached with accutase. The cell suspensions were collected, centrifuged for 5 min at 2,000 rpm, and then 10 μM dihydroethidium (DHE; Thermo Fisher) or 5 μM Mitosox (Thermo Fisher) were added, before incubation for 30 min at 37°C, protected from light. Cytosolic and mitochondrial reactive oxygen species (ROS) were measured on a BD FACS LSRII instrument (Becton, Dickinson) and processed using FlowJo v10.5.3 (Tree Star Inc.).

Macrophages (5 × 10⁵/well in 24-well plates) were infected with A. fumigatus live conidia at a 1:10 effector-to-target for 24 h, while PBMCs (2.5 × 10⁶/well in 96-well plates) were infected with A. fumigatus heat-killed conidia at a 1:4 effector-to-target ratio for 7 days at 37°C and 5% CO_2. Then, the supernatants were collected, and cytokine levels were quantified using ELISA MAX Deluxe set Kits (BioLegend), according to the manufacturer’s instructions. Cytokine quantification was also performed on the supernatants of lung single-cell suspensions 3-day postinfection using ELISA MAX Deluxe set Kits (BioLegend).

Macrophages (5 × 10⁵/well in 24-well plates) infected for 24 h at 37°C in 5% CO₂ with live A. fumigatus conidia at a 1:10 effector-to-target ratio. Then the supernatants were removed and centrifuged, and lactate levels were determined using the enzymatic colorimetric lactate assay (SpinReact), following the manufacturer’s instructions. For the quantification of lactate in vivo, lungs were removed and homogenized in lactate assay buffer, and lactate levels were determined using the l-Lactate assay kit (Abcam), following the manufacturer’s instructions.

The total RNA from the lungs was extracted using the GRS Total RNA kit-Tissue (GRiSP), on day 3 postinfection, according to the manufacturer’s instructions. The concentration and quality of total RNA in each sample were determined by spectrophotometry using the ND-100 UV-visible light spectrophotometer (NanoDrop). One microgram of the total RNA was retro-transcribed using the first-strand cDNA Synthesis kit (Nzytech). PowerUp SYBR green Master Mix (Applied Biosystems, Thermo Fisher Scientific) was used to perform quantitative PCR in an Applied Biosystems 7500 fast qPCR system (Applied Biosystems). Data were analyzed using 7500 Software v2.0.6 software (Applied Biosystems). Amplification efficiencies were validated, and the expression levels of the transcripts were normalized using the ubiquitin (Ubb) gene.

PBMC collection, stimulation experiments with A. fumigatus, and cQTL mapping was previously described (23). Briefly, genotype, cytokine, and cell count data measured by FACS for total lymphocytes data were available for 442 individuals from a population-based cohort, 500FG cohort, which is part of the HFGP (www.humanfunctionalgenomics.org). Log-transformed cytokine levels mapped to genotype data using a linear model with age, sex and cell counts as covariates. Given that the total number of independent tests among all phenotypes was much less than the number of phenotypes measured, correction for multiple testing was based on the number of SNPs tested for each trait, and a P value of < 5 × 10⁻⁸ was defined as the threshold for genome-wide significant cytokine QTLs. A nominal P value of < 0.05 was defined as the threshold for suggestive cQTLs.

The data were expressed as means ± SEM. Statistical significance of differences was determined by two-tailed Student's t test or two-way ANOVA with post hoc tests for multiple comparisons (P < 0.05 was considered statistically significant). Analyses were performed in GraphPad Prism software v9.3.1.