2020 Article 16120

ARTICLE
https://doi.org/10.1038/s41467-020-16120-z OPEN
Phagosomal removal of fungal melanin reprograms

macrophage metabolism to promote antifungal
immunity
Samuel M. Gonçalves1,2, Cláudio Duarte-Oliveira1,2, Cláudia F. Campos1,2, Vishukumar Aimanianda3,
Rob ter Horst 4, Luis Leite5, Toine Mercier 6,7, Paulo Pereira 8, Miguel Fernández-García9,
Daniela Antunes 1,2, Cláudia S. Rodrigues1,2, Catarina Barbosa-Matos 1,2, Joana Gaifem1,2, Inês Mesquita1,2,
1234567890():,;
António Marques10, Nuno S. Osório 1,2, Egídio Torrado 1,2, Fernando Rodrigues1,2, Sandra Costa1,2,
Leo AB. Joosten 4, Katrien Lagrou 7,11, Johan Maertens6,7, João F. Lacerda8, António Campos Jr.5,
Gordon D. Brown12, Axel A. Brakhage13,14, Coral Barbas 9, Ricardo Silvestre 1,2, Frank L. van de Veerdonk4,
Georgios Chamilos15,16, Mihai G. Netea 4,17, Jean-Paul Latgé 3, Cristina Cunha 1,2,18 &
Agostinho Carvalho 1,2,18 ✉
In response to infection, macrophages adapt their metabolism rapidly to enhance glycolysis

and fuel specialized antimicrobial effector functions. Here we show that fungal melanin is an
essential molecule required for the metabolic rewiring of macrophages during infection with
the fungal pathogen Aspergillus fumigatus. Using pharmacological and genetic tools, we reveal
a molecular link between calcium sequestration by melanin inside the phagosome and
induction of glycolysis required for efficient innate immune responses. By remodeling the
intracellular calcium machinery and impairing signaling via calmodulin, melanin drives an
immunometabolic signaling axis towards glycolysis with activation of hypoxia-inducible factor
1 subunit alpha (HIF-1α) and phagosomal recruitment of mammalian target of rapamycin
(mTOR). These data demonstrate a pivotal mechanism in the immunometabolic regulation of
macrophages during fungal infection and highlight the metabolic repurposing of immune cells
as a potential therapeutic strategy.
1 Life and Health Sciences Research Institute (ICVS), School of Medicine, University of Minho, 4710-057 Braga, Portugal. 2 ICVS/3B’s - PT Government
Associate Laboratory, Guimarães/Braga, Portugal. 3 Unité des Aspergillus, Institut Pasteur, 75015 Paris, France. 4 Department of Internal Medicine and
Radboud Center for Infectious Diseases, Radboud University Medical Centre, 6500HB Nijmegen, Netherlands. 5 STMO, Instituto Português de Oncologia,
4200-072 Porto, Portugal. 6 Department of Hematology, UZ Leuven, 3000 Leuven, Belgium. 7 Department of Microbiology and Immunology, KU Leuven,
3000 Leuven, Belgium. 8 Instituto de Medicina Molecular, Faculdade de Medicina de Lisboa, 1649-028 Lisboa, Portugal. 9 Center for Metabolomics and
Bioanalysis, Faculty of Pharmacy, San Pablo CEU University, 28668 Madrid, Spain. 10 Serviço de Imuno-Hemoterapia, Hospital de Braga, 4710-243
Braga, Portugal. 11 Department of Laboratory Medicine, UZ Leuven, 3000 Leuven, Belgium. 12 MRC Centre for Medical Mycology, University of Aberdeen,
Aberdeen AB25 2ZD, UK. 13 Department of Molecular and Applied Microbiology, Leibniz-Institute for Natural Product Research and Infection Biology, 07745
Jena, Germany. 14 Institute of Microbiology, Friedrich Schiller University, 07743 Jena, Germany. 15 School of Medicine, University of Crete, 70013
Heraklion, Greece. 16 Institute of Molecular Biology and Biotechnology, FORTH, 70013 Heraklion, Greece. 17 Department of Genomics & Immunoregulation,
Life and Medical Sciences Institute, University of Bonn, 53115 Bonn, Germany. 18These authors contributed equally: Cristina Cunha, Agostinho Carvalho.
✉email: [email protected]
NATURE COMMUNICATIONS | (2020)11:2282 | https://doi.org/10.1038/s41467-020-16120-z | www.nature.com/naturecommunications 1

ARTICLE NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-020-16120-z
T
he reprogramming of cellular metabolism is a fundamental and dendritic cells28. Moreover, there are several examples of
mechanism through which innate immune cells meet the genetic variants that predispose humans to aspergillosis by
energetic and anabolic needs during host defense against affecting the ability of myeloid cells to produce cytokines or exert
invading pathogens1. Sensing of microbial ligands drives the their killing activity29–32. Although the fine-tuned regulation of
upregulation of glycolysis, which delivers a rapid source of energy cellular metabolism is required for the functional activity of
to support antimicrobial functions and the production of cyto- macrophages, how these processes are orchestrated in response to
kines2. The enhanced glycolytic activity also directly supports A. fumigatus remains undefined.
cytokine expression through mechanisms that involve moon- Here, we sought to understand the mechanisms through which
lighting activities of the enzymes themselves3,4. The metabolic infection with A. fumigatus rewires macrophage metabolism
pattern of myeloid cells activated by canonical stimuli (e.g., toward efficient innate immune responses. We show that fungal
lipopolysaccharide, LPS) generally implies the downregulation of melanin is an essential PAMP required for the Warburg shift and
mitochondrial carbon metabolism and oxidative phosphoryla- the ensuing immunometabolic responses in macrophages. Our
tion5–8. The disruption in the tricarboxylic acid cycle promotes results define how the host is able to counter the immune inhi-
the accumulation of mitochondrial metabolites that in turn reg- bitory mechanisms deployed by fungal melanin in order to pro-
ulate the expression of glycolytic enzymes to support inflamma- mote efficient antifungal immune responses required to control
tory responses and antimicrobial functions9. This metabolic infection.
reprogramming generates lactate from glucose without further
mitochondrial oxidation, despite normoxic conditions, a pheno-
type observed in cancer cells and known as the Warburg effect10. Results
It is now clear that, under conditions of microbial challenge, Macrophage metabolism is modulated by A. fumigatus infec-
glucose metabolism is critically required; for example, com- tion. Our first aim was to investigate host cellular metabolism
pounds that block the metabolic shift to glycolysis (e.g., inhibitors during the interaction of human macrophages with A. fumigatus.
of the mammalian target of rapamycin (mTOR) pathway or We chose two time points after infection corresponding to the
metformin) restrain cytokine production and dampen immune early transcriptional events following phagocytosis (2 h) and the
responses, hampering pathogen clearance9,11–13. Importantly initial phases of fungal escape by germination (6 h). RNA-seq
however, other studies have shown that stimuli other than LPS analysis revealed substantial changes in the macrophage tran-
induce upregulation of both glycolysis and oxidative phosphor- scriptome after 2 h of infection, encompassing 1187 differentially
ylation in immune cells14. expressed genes relative to uninfected controls (Fig. 1a). The
The recognition of pathogen-associated molecular patterns number of differentially expressed genes decreased markedly at
(PAMPs) drives substantial changes in cellular metabolism and 6 h to a total of 83, indicating that the bulk of transcriptional
effector functions of immune cells10. Owing to its dynamic changes occurred early after infection. Enrichment analysis
composition and structural plasticity, the cell wall is considered demonstrated that pathways involved in cytokine production and
the most relevant repository for fungal PAMPs15,16. Exposure to signaling, response to inflammatory stimuli, and immune cell
β-1,3-glucan has been shown to promote the metabolic repro- differentiation were overrepresented in the upregulated genes
gramming of monocytes leading to a trained immunity pheno- after 2 h of infection (Fig. 1b). In accordance with the low number
type characterized by enhanced cytokine production in response of differentially expressed genes detected 6 h after infection, genes
to heterologous secondary stimulation11,17–19. The requirement involved in the negative regulation of transcription and DNA
for the metabolic rewiring of myeloid cells was also demonstrated binding were enriched in the downregulated pathways.
in vivo, since the pharmacological impairment of glycolysis20 or A targeted analysis of the differentially expressed genes after 2 h
the blockade of metabolic pathways with metformin21 increased of infection using hallmark gene sets from the Molecular
susceptibility of mice to systemic candidiasis. In turn, fungal Signatures Database33 revealed a substantial upregulation of genes
pathogens have evolved intricate virulence strategies to withstand involved in glycolysis, but not oxidative phosphorylation (Fig. 1c).
the host immune response, by exploiting nutritional weaknesses The commitment of macrophages toward glycolysis was reflected
of immune cells leading to their death22. by the upregulation of the glucose transporters SLC2A1 (GLUT1)
Although β-1,3-glucan is a major fungal cell wall component, and SLC2A6 (GLUT6), the key glycolytic regulator 6-phospho-
not much is known about the metabolic regulation of immunity to fructose-2-kinase/fructose-2,6-bisphosphatase 3 (PFKFB3), and
fungi other than C. albicans and whether other cell wall poly- hypoxia-inducible factor 1 subunit alpha (HIF1A). In contrast,
saccharides participate in these signaling events. Here, we investi- thioredoxin-interacting protein (TXNIP), which encodes a redox
gated the immunometabolic response of macrophages to the sensor that suppresses glucose uptake and metabolism34, was
opportunistic fungal pathogen Aspergillus fumigatus. This fungus instead downregulated. These results were consistent with a
can cause a wide spectrum of diseases with distinct clinical man- skewing of infected macrophages toward an M1 inflammatory
ifestations23. Epidemiological data have revealed that A. fumigatus phenotype characterized by the upregulation of several proin-
causes >200,000 invasive infections each year in hematological flammatory cytokines, including IL1B, TNF, and IL-6.
patients under aggressive chemotherapy or undergoing solid organ Gene expression analysis using a more detailed time course of
or allogeneic stem-cell transplantation24. Because there are no infection demonstrated that induction of glycolytic genes such as
licensed vaccines and the currently available diagnostic tests lack GLUT1, hexokinase 2 (HK2), and PFKFB3 initiated early after the
accuracy, mortality rates after infection are estimated above 30%25. challenge, and was sustained throughout infection (Fig. 1d). Glut1
Macrophages are considered critical in preventing fungal ger- and several glycolytic enzymes were also induced in the lungs of
mination and tissue invasion early after infection, particularly mice as early as day 1 and until day 3 after fungal infection
before the influx of neutrophils26. One relevant mechanism is (Fig. 1e). The transcriptional induction of glycolysis was confirmed
represented by the instruction of programmed necrosis in mac- by the increased secretion of lactate and glucose consumption by
rophages by calcineurin, which by stimulating the lateral transfer human macrophages throughout the infection, an effect that was
of conidia between macrophages, enables the control of fungal dependent on the multiplicity of infection (Fig. 1f), but was not
germination27. Other studies have highlighted the importance of influenced by the differentiating stimulus (Supplementary Fig. 1a).
inflammatory monocytes in experimental aspergillosis through Fungal infection also elicited an increased ADP/ATP balance
their ability to orchestrate the conidiacidal activity of neutrophils (Fig. 1g). To gain further insight into the metabolites produced
2 NATURE COMMUNICATIONS | (2020)11:2282 | https://doi.org/10.1038/s41467-020-16120-z | www.nature.com/naturecommunications

NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-020-16120-z ARTICLE
a 2 h post-infection 6 h post-infection b GO Biological Processes

140 12
cytokine-mediated signaling pathway 121
120 372 815 10 44 39 cellular response to lipopplysaccharide 70
Adj P value (–log10)
Adj P value (–log10)

100 cytokine production 102
8
cellular response to lipid 88
80
6 leukocyte differentiation 73
60 negative regulation of transcription 42
4 detection of biotic stimulus 5
40
small GTPase mediated signal transduction 21
20 2
negative regulation of DNA binding 6
0 0 dendritic cell migration 4
–5 0 5 10 –4 –2 0 2 4 6 0 10 20 30 40 50
Expression (log2 fold) Expression (log2 fold) Adj P value (–log10)
c d f
Ctrl1
Ctrl2
Ctrl3
Ctrl1
Ctrl2
Ctrl3
Af1
Af2
Af3
Af1
Af2
Af3
Human Mø
GLUT1 HK2 PFKFB3 15
mRNA (relative expression)

GPC1 IL1R1 Ctrl
30 4 10 P = 0.0001
GPC3 KYNU P < 0.0001 1:2
CD44 IL12B P = 0.0009 8 1:10
Lactate (mM)
P = 0.0017 P = 0.049
AGRN 3 P = 0.0029
SOCS3 10
20
P < 0.0001
P = 0.0019 P = 0.0055
PKP2 IRF1 6
B4GALT1 P = 0.0026
CCL4 2
M1 markers
HIF1A IL1B P = 0.0002 4

10 5
IER3 TNF P = 0.027 1
Glycolysis
PFKFB3 CD83 2
SOD2 IL23A
SLC2A6 PTX3 0 0 0 0
TPBG IL6 1 2 4 8 1 2 4 8 1 2 4 8 0 8 16 24
SLC2A1 CD40
GLRX
Time (h) Time (h) Time (h) Time (h)
GBP1
GFPT2 CCR7
e 15
AKR1B15 Mouse lung
IDO1
P = 0.037 P = 0.0069
ADORA2B Glut1 Hk2 Pfkb3
Glucose (mM)
80 12 30
Ctrl1
Ctrl2
Ctrl3
P < 0.0001
TXNIP P < 0.0001 P = 0.0003 10
Af1
Af2
Af3
GCLC P = 0.0005
ANKZF1 IL1RN 60
M2 markers
IL10 8 P < 0.0001 20

Ctrl1
Ctrl2
Ctrl3
SOCS1 5 Ctrl
Af1
Af2
Af3
40 1:2
IRF4
CYB5R3 ID3 P = 0.0023 4 10 P = 0.0002 1:10
PDP1 MAF 20 0
OXPHOS
CASP7
0 8 16 24
SLC25A37 Normalized 0 0 0
ATP2B1 relative expression Time (h)
PDK4 1 3 1 3 1 3
SDHAF1 Row min Row max Time (days) Time (days) Time (days)
MRPS24
PBS A. fumigatus
g Ctrl h Glucose-6P Fructose-6P DHAP Glyceraldehyde-3P 2/3PG PEP Pyruvate

A. fumigatus
0.025 0.08 0.6 P = 0.049 0.08 0.20 0.10 0.005 P = 0.0095
0.15 P = 0.031 P = 0.016
Metabolite (mg/L)
0.020 0.06 0.06 0.15 0.08 0.004

P = 0.046 0.4
ADP/ATP ratio
0.10 0.015 0.06 0.003

0.04 0.04 0.10
0.010 0.04 0.002
0.2
0.05 0.02 0.02 0.05
0.005 0.02 0.001
0.000 0.00 0.0 0.00 0.00 0.00 0.000

0.00 Glycolysis
Ctrl A. fumigatus
Fig. 1 A. fumigatus induces glycolysis in macrophages. a Transcriptome analysis of human macrophages infected with A. fumigatus for 2 or 6 h. Numbers
indicate genes with differential expression, up- (red) or downregulated (blue) in infected relative to uninfected cells. b Pathway analysis of up- (red) or
downregulated (blue) genes 2 h after infection. Genes were categorized into the most represented pathways, in which the gene products are involved.
c Transcriptional profiles of macrophages left untreated (Ctrl) or infected with A. fumigatus (Af) for 2 h (n = 3). Expression of genes is presented as
centered, and scaled log2 fluorescence intensity (blue and red keys) grouped by product function. d mRNA expression of GLUT1, HK2, and PFKFB3 in
macrophages infected for 1, 2, 4, or 8 h relative to uninfected cells (n = 6). e mRNA expression of Glut1, Hk2, and Pfkfb3 in mouse lungs sampled 1 or 3 days
after infection (n = 8, representative of three independent experiments). f Lactate secretion and glucose consumption by macrophages left untreated or
infected at 1:2 or 1:10 for 24 h (n = 6). g ADP/ATP ratio (n = 5) and h targeted metabolomics (n = 6) in macrophages left untreated or infected for 6 h.
Data are expressed as mean values ± SEM. P-values were calculated using Student’s two-tailed t test or two-way ANOVA with Tukey’s multiple
comparisons test.
along the glycolytic pathway, we performed a targeted analysis of infected with A. fumigatus did not display a significant loss of
metabolic pathways after 6 h of infection using liquid chromato- viability, even after 24 h of infection (Supplementary Fig. 1b).
graphy tandem-mass spectrometry (LC–MS/MS). In these condi- Likewise, the total number of macrophages in the lungs of
tions, increased levels of most glycolytic intermediates were infected mice also remained intact after infection (Supplementary
detected, namely, fructose-6-phosphate, dihydroxyacetone phos- Fig. 1c). Further supporting the lack of macrophage death due to
phate, glyceraldehyde-3-phosphate, and pyruvate (Fig. 1h). competition for glucose, expression of fungal glycolytic enzymes,
The metabolic rewiring of macrophages during infection including ATP-dependent 6-phosphofructokinase (pfkA) and
commits them to glucose metabolism for survival, and this hexokinase-1 (hxkA), was not significantly modulated during
phenotype is exploited by C. albicans to trigger cell death by infection (Supplementary Fig. 1d), a finding in line with the
depleting glucose levels22. In contrast, human macrophages negligible amount of lactate measured during fungal culture

(Supplementary Fig. 1e). These results reveal the glycolytic Induction of host glycolysis depends on fungal melanin. To
reprogramming of macrophages in response to A. fumigatus, identify the mechanism(s) underlying the activation of glycolysis
while highlighting different metabolic strategies across fungal during infection, we compared lactate secretion by macrophages
genera to survive and thrive during infection. challenged with different fungal morphotypes for 24 h. Secretion
of lactate was higher when stimulation involved dormant conidia
compared with swollen conidia or germ tubes (Fig. 3a). We
Glycolysis is required for immune responses to A. fumigatus. hypothesized that the fungal molecule(s) inducing glycolysis were
Production of proinflammatory cytokines, phagocytosis and present on dormant conidia, but were mostly lost after germi-
killing are major effector functions of macrophages during nation. Rodlet and melanin, made up of hydrophobic protein
infection with A. fumigatus16. We tested therefore how the RodA and polymerized 1,8-dihydroxynaphthalene (DHN),
inhibition of glycolysis using 2-deoxyglucose (2-DG), a compe- respectively, form the outermost layer of dormant conidia and are
titive inhibitor of hexokinase, affected these processes. We con- both removed during germination35. To test their contribution,
firmed that the treatment with 2-DG decreased lactate secretion we assessed lactate secretion after infection with dormant conidia
by infected macrophages in a dose-dependent manner (Fig. 2a), of the ΔrodA mutant devoid of the surface rodlet layer36 or the
without affecting cellular viability (Supplementary Fig. 2a). ΔpksP mutant lacking the polyketide synthase responsible for the
Likewise, macrophages cultured in glucose-deprived media or initial step in DHN–melanin biosynthesis37. Infection with ΔrodA
media containing galactose, impairing the glycolytic flux, were conidia induced lactate secretion (Fig. 3b) and glucose con-
also unable to secrete lactate during infection (Supplementary sumption (Supplementary Fig. 3a) to an extent comparable with
Fig. 2b). Upon treatment with 2-DG (Fig. 2b) or using media that measured for the parental strain Δku80, whereas ΔpksP or
without glucose or with galactose (Supplementary Fig. 2c), mac- ΔrodA/pksP conidia failed to efficiently trigger this metabolic
rophages displayed an impaired conidiacidal activity, which was reprogramming. Macrophages infected with ΔpksP conidia also
in line with the decreased production of reactive oxygen species displayed decreased mRNA levels of glycolysis-related genes
(Fig. 2c). The phagocytic ability instead remained intact (Fig. 2d), (Fig. 3c). Collectively, these findings are consistent with a critical
a finding supported by data showing that phagocytosis relies on role for fungal melanin in reprogramming macrophage metabo-
oxidative phosphorylation14. In addition, treatment of macro- lism during infection.
phages with 2-DG (Fig. 2e) or the use of glucose-deprived media To determine at which stage the putative ligand(s) inducing
(Supplementary Fig. 2d) impaired the production of proin- glycolysis were synthesized in the pathway, we screened deletion
flammatory cytokines, including IL-1β, TNF, and IL-6, after 24 h mutants in the enzymes required to catalyze each step for their
of infection. The production of adaptive cytokines by PBMCs, ability to activate glycolysis. We confirmed that lactate secretion
such as IFNγ and IL-17A, detected after 7 days of infection, was was severely affected throughout the entire DHN–melanin
also compromised following the blockade of glycolysis (Supple- biosynthetic pathway (Supplementary Fig. 3b), suggesting that
mentary Fig. 2e). fully mature melanin particles, and not the heptaketide naphtho-
Since the blockade of hexokinase by 2-DG might also affect pyrone synthesized by PksP alone, is required for the metabolic
the pentose phosphate pathway (PPP), we evaluated lactate reprogramming. The requisite role for melanin in the activation of
secretion by macrophages treated with 3-(3-pyridinyl)-1-(4- glycolysis was confirmed by similar defective levels of lactate
pyridinyl)-2-propen-1-one (3PO), a selective inhibitor of secreted by macrophages infected with the albino strains of A.
PFKFB3, and 6-aminonicotinamide (6-AN), an inhibitor of the fumigatus CBS 110.46 and CBS 386.7538 (Fig. 3d). To elucidate
6-phosphogluconate dehydrogenase enzyme from the PPP. In whether melanin pigments that differ from DHN–melanin could
these conditions, we confirmed an impaired lactate secretion also activate glycolysis, we infected macrophages with A. nidulans
(Fig. 2f) and cytokine production (Supplementary Fig. 2f) using whose melanin differs from that of A. fumigatus39. In these
3PO, but not 6-AN. To gain further insight into lung conditions, lactate secretion occurred to a degree similar to that
microenvironment-associated effects on glycolysis, we measured induced by the Δku80 strain of A. fumigatus (Fig. 3e).
lactate secretion by murine alveolar macrophages after infection Because the recently identified melanin receptor MelLec could
(Fig. 2g; Supplementary Fig. 2g). The results showed an be involved in melanin-induced activation of glycolysis, we
induction of lactate secretion, an effect that was abrogated stimulated macrophages with 1,8-DHN, polymerization of which
following treatment with 2-DG. Collectively, these data highlight leads to melanin biosynthesis, and that contains the conserved
the critical requirement for glycolysis to host defense mechan- naphthalene-diol unit recognized by MelLec32. We observed that
isms in response to A. fumigatus. the levels of lactate secreted by macrophages stimulated with 1,8-
Because our results pointed to a pivotal role of glucose DHN were significantly lower than those after infection with the
metabolism in antifungal host defense, we next validated this Δku80 strain (Fig. 3f), suggesting that recognition of 1,8-DHN
requirement in a mouse model of pulmonary aspergillosis, in melanin and signaling via MelLec is, to a large extent, dispensable
which immunocompetent animals were treated with 2-DG or PBS for the activation of glycolysis. In support of this, infection of
prior to infection (Fig. 2h). In line with the activation of glucose bone marrow-derived macrophages (BMDMs), which lack
metabolism, we detected systemic hypoglycemia in mice after the expression of MelLec32, with ΔpksP conidia also induced lower
challenge (Supplementary Fig. 2h). Blocking glycolysis rendered concentrations of secreted lactate (Supplementary Fig. 3c).
mice more susceptible to infection, as revealed by the increased Moreover, human macrophages carrying the loss-of-function
fungal burden in the lungs 1 day after infection (Fig. 2h), a rs2306894 variant in MelLec displayed comparable levels of
finding that did not involve loss of viability of alveolar lactate secreted after infection to that of wild-type cells
macrophages (Supplementary Fig. 2i). In addition, the levels of (Supplementary Fig. 3d).
cytokines in lung homogenates from 2-DG-treated mice were To directly link the presence of melanin on the conidial surface
significantly lower than those from mock-treated animals (Fig. 2i). and activation of glycolysis, we stimulated macrophages with
Consistently, the production of cytokines after restimulation of melanin ghosts alone or added together with ΔpksP conidia.
splenocytes from 2-DG-treated mice (Fig. 2j) and their Strikingly, none of these conditions significantly induced lactate
conidiacidal activity (Fig. 2k) was also impaired. Taken together, secretion (Fig. 3g), a finding suggesting the need for melanin to be
these data confirm that glucose metabolism plays a central role in actively shed or removed during conidial germination to activate
the induction of host responses to A. fumigatus in vivo. glycolysis. Indeed, only live, but not heat-killed or UV-

a P = 0.0003 b c Ctrl d
P = 0.0004 A. fumigatus
10
Conidiacidal activity (%)

80 A. fumigatus + 2-DG 80
P = 0.0015 P = 0.020
30
Phagocytosis (%)
P = 0.0004 P < 0.0001
ROS production (RFU)
8
Lactate (mM)
60 P < 0.0001 60
P < 0.0001
6 P < 0.0001
40 20 40
4
2 20 10 20
0 0 0
Ctrl 2-DG 2-DG 0 2-DG
0 4 8 12 16
A. fumigatus A. fumigatus A. fumigatus
Time (h)
e Human Mø f g Mouse alveolar Mø

IL-1β TNF IL-6 15 P = 0.0057 2.5 P = 0.0029
2.5 15 25 P = 0.0004 P < 0.0001
P < 0.0001 P < 0.0001
Lactate (mM)
Lactate (mM)
Cytokine (ng/mL)
2.0 20 10 2.0
10
1.5 15
1.0 10 5 1.5
5
0.5 5
0 1.0
0.0 0 0
–
G
O
G
AN
trl
trl
D
3P
D
– 2-DG – 2-DG – 2-DG
C
2-
2-
6-
A. fumigatus A. fumigatus
h 7
i Mouse lung
P = 0.0001
A. fumigatus (1 × 10 8 IL-1β TNF IL-6
CFU (log10).g–1
500 P = 0.0014 80 300 P = 0.0035

conidia/20 μl) or PBS i.n. 6
P = 0.0069
Cytokine (pg/mL)
400 60
5 200
300
day –3 day 0 day +1 40
4
200
2-DG (100 mg/Kg) or 100
3 100 20
PBS i.p. daily
G
S
0 0 0
PB
D
2-
G
S
S
PB
PB
PB
D
D
2-
2-
2-
j Mouse splenocytes k
IL-1β TNF IL-6
Conidiacidal activity (%)
100 500 2000 P < 0.0001 100

P = 0.039 P = 0.0034 P = 0.0069
Cytokine (pg/mL)
80 400 1500 80
60 300 60
1000
40 200 40
20 100 500 20
0 0 0 0
G
G
S
S
PB
PB
PB
PB
D
D
2-
2-
2-
2-
Fig. 2 Glycolysis is required for antifungal immune responses. a Lactate secretion (n = 4) and b conidiacidal activity (n = 8) of macrophages left untreated
(Ctrl) or infected with A. fumigatus for 24 or 3 h, respectively, without or with 5, 10, or 20 mM 2-DG (n = 4). c ROS production by macrophages left
untreated or infected for 24 h without or with 10 mM 2-DG (n = 8). d Phagocytosis of macrophages infected for 1 h without or with 5, 10, or 20 mM 2-DG
(n = 6). e Production of IL-1β, TNF, and IL-6 by macrophages infected for 24 h without or with 10 mM 2-DG (n = 10). f Lactate secretion by macrophages left
untreated (Ctrl) or infected with A. fumigatus for 24 h without or with 10 mM 2-DG, 30 µM 3PO or 500 nM 6-AN (n = 7). g Lactate secretion by mouse
alveolar macrophages left untreated (Ctrl) or infected for 24 h without or with 10 mM 2-DG (n = 3). h Experimental setup for glycolysis inhibition in vivo
during infection. Fungal burden (log10) per gram of lung tissue was determined in PBS- or 2-DG-treated mice after 1 day of infection (n = 8, representative
of three independent experiments). i Levels of IL-1β, TNF, and IL-6 in lung homogenates of PBS- or 2-DG-treated mice after 1 day of infection (n = 5).
j Production of IL-1β, TNF, and IL-6 and k conidiacidal activity of mouse splenocytes isolated from PBS- or 2-DG-treated mice and restimulated for 24
(n = 10) or 2 h (n = 5), respectively. Data are expressed as mean values ± SEM. P-values were calculated using Student’s two-tailed t test or two-way
ANOVA with Tukey’s multiple comparisons test.
inactivated, conidia triggered lactate secretion by macrophages of melanin isolated from A. fumigatus conidia. The ability of the
(Fig. 3h), a finding highly suggestive of a cell wall remodeling ΔrodA/pksP strain to induce lactate secretion was rescued by
associated with melanin release. To further test this, we the melanin coating in a dose-dependent manner (Fig. 3i), a
performed functional complementation experiments using live finding supported by the increase in the mRNA expression of
dormant ΔrodA/pksP conidia (allowing enhanced adhesion due to glycolytic genes (Fig. 3j). In line with the link between glycolysis
removal of the hydrophobic layer) coated with purified fragments and cytokine production, infection with melanin-coated live

a P = 0.0022 b P < 0.0001

c d 15 P = 0.0003
15 P = 0.0005
15 P < 0.0001 GLUT1 HK2 PFKFB3 P < 0.0001

8 5 10
Lactate (mM)
P = 0.0023 P < 0.0001 10
Lactate (mM)
Lactate (mM)
8 P = 0.0004
10 10 6 4
3 6
4 5
5 5 2 4
2 2
1 0
0 0
38 6
75
C Δku l
80
tr
0 0 0
BS .4
C
n
6.
S nt
C 100
or l
Δr Δ dA
Δr 0
Δk trl
A sP
P
be
D Ctr
er lle
Δku80
u8
a
ks
C
G wo
o
m
tu
od pk
/p
BS
ΔpksP
m
P = 0.022
e f P < 0.0001 g P = 0.011 h i P = 0.0056
20 15 20 20 P = 0.0080 15 P = 0.0003
P = 0.0037
P = 0.0084 P < 0.0001
15
Lactate (mM)
Lactate (mM)
15
Lactate (mM)
15
Lactate (mM)
Lactate (mM)
10 10
10 10 10
5 5
5 5 5
0 0 0 0 0
ns
ve
+ el n
ks in
Δk l
Δk l
du 0
N
0
Δk trl
trl
Δp 0
Δk trl
tr
0
M ksP
tr
M ni
8
u8
Melanin
Δp an
H
u8
u8
C
U
la
Li
P
C
C
A. u
a
D
el
8-
ΔrodA/pksP
ni
1,
j GLUT1 HK2 PFKFB3

k IL-1β TNF IL-6
80 40 10 4 20 75
P < 0.0001 P < 0.0001 P < 0.0001
P < 0.0001 P = 0.0036
P < 0.0001 3 15 50
Cytokine (ng/mL)
8 P < 0.0001
60 30 P < 0.0001 P < 0.0001 P = 0.0006
P < 0.0001 2 10 25
6 1 5
40 20 0.03 0.06 0.12
4
0.02 0.08
20 10 0.03
2 0.01 0.04
0 0 0 0.00 0.00 0.00
Δku80 Δku80
ΔrodA/pksP ΔrodA/pksP
Melanin-coated ΔrodA/pksP Melanin-coated ΔrodA/pksP
Fig. 3 Intracellular removal of fungal melanin induces glucose metabolism. a Lactate secretion by macrophages left untreated (Ctrl) or infected with
dormant or swollen conidia, or germ tubes of A. fumigatus for 24 h (n = 6). b Lactate secretion by macrophages left untreated (Ctrl) or infected with
dormant conidia from the parental Δku80 strain or the cell wall mutants ΔrodA, ΔpksP, or ΔrodA/pksP for 24 h (n = 6). c mRNA expression of GLUT1, HK2,
and PFKFB3 in macrophages infected with Δku80 or ΔpksP strains for 2 h relative to uninfected cells (n = 5). d Lactate secretion by macrophages left
untreated (Ctrl) or infected with the albino CBS 100.46 and 386.75 strains of A. fumigatus (n = 6), e the WG355 strain of A. nidulans (n = 6), f 1,8-DHN
(n = 6), or g the ΔpksP strain, DHN–melanin, or DHN–melanin added together with ΔpksP conidia (n = 4) for 24 h. Infection with Δku80 conidia was used
as control. h Lactate secretion by macrophages left untreated (Ctrl) or infected with live, heat-killed (HK), or UV-inactivated (UV) conidia for 24 h (n = 5).
i Lactate secretion by macrophages infected with the Δku80, ΔrodA/pksP, or the ΔrodA/pksP strain coated with 100, 300, and 600 µg/mL of DHN–melanin
for 24 h (n = 5). j mRNA expression of GLUT1, HK2, and PFKFB3 (n = 8) and k production of IL-1β, TNF, and IL-6 (n = 5) by macrophages infected with the
Δku80, ΔrodA/pksP, or the melanin-coated ΔrodA/pksP strain for 2 or 24 h, respectively, relative to uninfected cells (n = 8). Data are expressed as mean
values ± SEM. P-values were calculated using Student’s two-tailed t test or one-way ANOVA with Tukey’s multiple comparisons test.
ΔrodA/pksP conidia also reverted the defective production of was confirmed by the enhanced phosphorylation of the p70S6
proinflammatory cytokines (Fig. 3k). The same experimental kinase, a downstream target of the mTORC1 complex (Fig. 4b;
approach using inactivated conidia was instead unable to restore Supplementary Fig. 4a). Remarkably, ΔpksP conidia failed to
lactate secretion to normal levels (Supplementary Fig. 3e). These elicit a significant increase in p-p70S6K, indicating that melanin
results demonstrate that the active intracellular removal of is required for the induction of mTOR signaling. Because
melanin is required for the reprogramming of glucose metabolism mTOR activation is often mediated by the intermediary acti-
in macrophages. vation of the phosphatidylinositol-4,5-bisphosphate 3-kinase
(PI3K)/Akt pathway41, we assessed phosphorylation of Akt in
Fungal melanin rewires host metabolism via mTOR and HIF- the same conditions. Similar to mTOR, the levels of p-Akt were
1α. The metabolic reprogramming of myeloid cells toward increased during infection with Δku80 conidia, but less when
glycolysis is driven by the activation of mTOR, an intracellular the ΔpksP strain was used (Fig. 4b; Supplementary Fig. 4b). As
sensor that functions as a master regulator of glucose meta- these results suggest a coupled activation of Akt/mTOR and
bolism40. Consistent with this, the transcriptome of human glycolysis, we next investigated the causality between these two
macrophages after 2 h of infection with A. fumigatus was processes during infection. Inhibiting this pathway with rapa-
markedly enriched in genes involved in mTOR signaling mycin (mTOR) (Fig. 4c) or wortmannin (Akt) (Supplementary
(Fig. 4a). The activation of the mTOR pathway during infection Fig. 4c) during infection with Δku80 conidia impaired the

a b Δku80 ΔpksP Δku80 ΔpksP c 20
Ctrl1
Ctrl2
Ctrl3
P = 0.0006
Af1
Af2
Af3
0
0
0
0
24
0
24
Time (min):
60
60
0
30
30
24
24
30
60
12
12
30
60
12
12
0
0
15
Lactate (mM)
DPYSL2
p-p70S6K 70 kD
TRIB3
SLC7A11 10
p70S6K 70 kD
EIF2AK3
DDIT4 p-Akt 62 kD 5
RHOF
PRKAB2 Akt 62 kD
0
PPP1R15A
NFIL3 β-actin 42 kD Ctrl
RHOG Δku80
Donor A Donor B Δku80 + rapamycin
PIK3CD
PSMA6
NAMPT
MTHFD2 d IL-1β TNF IL-6
e
PPP3CC Ctrl
2.5 25 P = 0.0014 60
mTOR signaling
XBP1 P = 0.025 P = 0.021

Cytokine (ng/mL)
PITPNB 2.0 20 50
EIF1B
40
PPP4R2 1.5 15
FNBP1 30
RHOV 1.0 10
20
PPP1R2
BTG2 0.5 5 10
NFKBIB
EDEM1 0.0 0 0 Δku80
SLC7A5 Δku80
RHOB Δku80 + rapamycin
PPP1R15B
PPP1R18
RHOH
f 4 P < 0.0001
g Δku80
CXCR4 P < 0.0001 ΔpksP
DDIT3 3 0.4 P < 0.0001
Lactate (mM)
PPP1R37 P = 0.0019
PIKFYVE 2 0.3 ΔpksP
IL-1β (ng/mL)
PIK3CG
PPP1R9B P = 0.024
1 0.2
Normalized
relative expression
0 0.1
Row min Row max
Δp 0
Δk r l
Δp 0
Δk r l
P
P
u8
t
u8
t
ks
C
ks
C
0.0
C57BL/6 HIF
C
C57BL/6 HIFC
ΔrodA/pksP
h Time (min):
0
8
60
60
60
60
Normalized pixel density

12
12
12
12
0
(p-p70S6K/p70S6K)
p-p70S6K 70 kD
6
p70S6K 70 kD
4
β-actin 42 kD
2
Melanin-coated
pk ed
sP
P
80
ks
A/ at
0
sP
u
ΔrodA/pksP
Δp
/p
Δk
od co
dA
Δr nin-
60
60
0
0
pk ed 60
60
0
trl
o
12
12
12
12
Δr
C
a
el
M
80
P
P
s
ks
A/ at
u
sP
pk
Δk
Δp
od co
A/
Δr in-
od
an
Δr
el
M
HIF-1α DAPI A. fumigatus
Fig. 4 mTOR and HIF-1α reprogram metabolism in response to A. fumigatus. a Transcriptional profiles of macrophages left untreated (Ctrl) or infected
with A. fumigatus (Af) for 2 h (n = 3). Expression of genes is presented as centered and scaled log2 fluorescence intensity (blue and red keys). b The total
and p-p70S6K and total and p-Akt in macrophages infected with the Δku80 or ΔpksP strains for 4 h (representative of three independent experiments, with
β-actin used as loading control). c Lactate secretion and d production of IL-1β, TNF, and IL-6 by macrophages left untreated (Ctrl) or infected with the
Δku80 strain for 24 h without or with 10 mM rapamycin (n = 6). e Expression of HIF-1α in macrophages left untreated (Ctrl) or infected with the Δku80,
ΔrodA/pksP or the melanin-coated ΔrodA/pksP strain for 2 h (representative of three independent experiments). The white arrows indicate accumulation in
the nuclei of macrophages. Scale bars, 100 µm. f Lactate secretion and g production of IL-1β by BMDMs from C57BL/6 and HIF-1c mice left untreated (Ctrl)
or infected with the Δku80 or ΔpksP strains for 24 h (n = 4). h Levels of total and p-p70S6K in macrophages left untreated (Ctrl) or infected with the
Δku80, ΔpksP, ΔrodA/pksP, or melanin-coated ΔrodA/pksP strains for 3 h (representative of three independent experiments, with β-actin used as loading
control). The pixel density of the p-p70S6K/p70S6K ratio was normalized to β-actin. Data are expressed as mean values ± SEM. P-values were calculated
using Student’s two-tailed t test or one-way ANOVA with Tukey’s multiple comparisons test.
ability of macrophages to reprogram their metabolism, leading Based on evidence that mTOR-mediated induction of glyco-
to lower levels of secreted lactate. Along the same line, blocking lysis requires activation of HIF-1α and stimulation of glycolytic
mTOR (Fig. 4d) or Akt (Supplementary Fig. 4d) impaired the enzymes42, we next assessed the expression of HIF-1α in infected
production of cytokines after infection. macrophages. We found higher amounts of HIF-1α accumulated

in the nucleus of cells infected with the Δku80 than the ΔpksP conditions, cells infected with the ΔpksP conidia still failed to
strain (Fig. 4e; Supplementary Fig. 4e), a finding indicating a upregulate lactate levels (Fig. 5e). In addition, we silenced
melanin-dependent induction of HIF-1α-mediated gene expres- stromal-interacting molecule 1 (STIM1), a calcium sensor
sion. Likewise, HIF-1α was more abundant in lysates from essential for calcium store depletion-triggered calcium influx45
macrophages infected with Δku80 conidia (Supplementary (Supplementary Fig. 5c). The inhibition of STIM1 also did not
Fig. 4f), and a similar profile was determined for HIF1A mRNA alter the profile of lactate secretion by macrophages infected with
(Supplementary Fig. 4g). To further investigate the link between the Δku80 or ΔpksP strains (Fig. 5f). Together, these observations
activation of HIF-1α and the metabolic reprogramming of indicate that ER calcium stores, independently of calcium entry,
macrophages, we assessed the impact of HIF-1α deficiency in regulate intracellular signaling pathways, which are required for
response to infection using BMDMs from wild-type (C57BL/6) the activation of glucose metabolism by fungal melanin.
and myeloid-restricted HIF-1α-deficient (HIF-1c) mice. In To gain further insight into the calcium-induced signaling
contrast to wild-type BMDMs, the ability of melanin to induce pathways mediating activation of glycolysis by melanin, we next
lactate secretion was lost in HIF-1c cells (Fig. 4f), a finding tested whether calcium/calmodulin (CaM) signaling was
recapitulated in human macrophages upon treatment with involved. We treated macrophages with W7, a specific CaM
ascorbate, a co-factor for the hydrolases that negatively regulate antagonist, and assessed the levels of lactate secretion after
HIF-1α (Supplementary Fig. 4h). Production of the proinflam- infection. In these conditions, macrophages infected with the
matory cytokine IL-1β, known to be transcriptionally regulated ΔpksP strain displayed increased levels of secreted lactate (Fig. 5g),
by HIF-1α9, was also impaired in HIF-1c BMDMs (Fig. 4g), a a finding in accordance with the upregulation of glycolytic genes
finding in line with the lower levels of cytokines produced by (Supplementary Fig. 5d), and indicating that the inhibition of
human macrophages treated with ascorbate (Supplementary calcium/CaM signaling by melanin is required for the transcrip-
Fig. 4i). In support of the link between fungal melanin and tional induction of glycolysis. To further illustrate the regulation
mTOR/HIF-1α signaling, infection of macrophages with of immunometabolic signaling by fungal melanin, we next
melanin-coated live ΔrodA/pksP conidia restored both p- evaluated the influence of calcium signaling on mTOR expres-
p70S6K (Fig. 4h) and HIF-1α translocation to the nucleus sion. In line with the impaired phosphorylation of the mTOR
(Fig. 4e) to levels comparable with those obtained after infection target p70S6 kinase (Fig. 4b), mTOR expression was also
with the Δku80 strain. Taken together, these results suggest that decreased during infection with ΔpksP conidia, a defect that
signaling via mTOR and HIF-1α is required for the melanin- was rescued by the CaM antagonist W7 (Supplementary Fig. 5e).
mediated activation of glycolysis in macrophages. Because melanin blocks CaM recruitment to conidia-containing
phagosomes43, we next assessed mTOR recruitment to the
phagosome. Macrophages infected with the ΔpksP strain
Remodeling of calcium signaling enables host glycolysis. Since displayed a lower percentage of mTOR+ phagosomes when
the ability of fungal melanin to sequester calcium inside the compared with cells infected with Δku80 conidia (Fig. 5h, i).
phagosome is a major inhibitory mechanism of intracellular Importantly, however, mTOR recruitment was restored following
signaling pathways43, we evaluated the contribution of calcium CaM inhibition, a finding implying a direct link between
signaling to macrophage metabolism during infection. In line melanin-mediated restraining of calcium signaling and activation
with the interference of fungal melanin with calcium responses, of mTOR and the downstream metabolic reprogramming of
live imaging of macrophages preloaded with the calcium indicator macrophages. Furthermore, macrophages infected with ΔpksP
Fluo-4-AM revealed that infection with ΔpksP or ΔrodA/pksP, but conidia and treated with either thapsigargin or W7 displayed an
not Δku80 conidia, triggered a sustained accumulation of cyto- enhanced ability to produce cytokines (Fig. 5j). Collectively, these
solic calcium early after infection (Fig. 5a, b; Supplementary results highlight a mechanism whereby the inhibition of calcium/
Fig. 5a). Importantly, stimulation of macrophages with live CaM signaling enables efficient immunometabolic responses.
ΔrodA/pksP conidia coated with purified melanin abrogated
cytosolic calcium flux in a similar fashion to the Δku80 strain,
confirming melanin as a master regulator of calcium-mediated Discussion
responses. This effect was dependent on fungal viability since The metabolic reprogramming of innate immune cells is required
stimulation with inactivated ΔrodA/pksP conidia or melanin to generate protective inflammation and drive antimicrobial
ghosts instead failed to promote cytosolic calcium accumulation defenses12,20,46. We establish fungal melanin as an essential
(Supplementary Fig. 5b). PAMP required for the induction of glycolysis in macrophages
To ascertain the specific contribution of melanin-mediated and the resulting antifungal immune responses. DHN–melanin is
regulation of calcium signaling to host cellular metabolism, we a major determinant of the fungal interaction with the innate
next performed calcium depletion and signaling inhibition during immune system35, endowing A. fumigatus with the ability to
infection and evaluated the induction of glycolysis. Strikingly, the survive killing by phagocytes, namely by blocking phagosome
depletion of extracellular calcium sources had no effect on biogenesis43,47 and acidification of phagolysosomes48, and pre-
calcium signaling and metabolism after infection with ΔpksP venting phagocyte apoptosis49. We now show that the host has in
conidia (Fig. 5c). In contrast, incubation with the cell-permeable turn evolved strategies to counter the inhibitory mechanisms
EGTA-AM further inhibited calcium responses to ΔpksP conidia, deployed by melanin, by sensing its removal during intracellular
suggesting that intracellular calcium sources regulate the meta- swelling or cell wall remodeling inside the phagosome and using
bolic reprogramming. Because calcium concentrations in differ- these signals to rewire cellular metabolism and promote anti-
ent subcellular compartments exert distinct functional effects44, fungal immune responses.
we performed the depletion of endoplasmic reticulum (ER) The innate immune system is equipped to respond to the
calcium stores with thapsigargin, a sarco/ER calcium ATPase expression of virulence factors from fungi16. The identification of
(SERCA) inhibitor, during infection. Upon inhibiting SERCA, the the MelLec receptor, which recognizes DHN–melanin and acti-
impaired ability of macrophages to secrete lactate in response to vates antifungal defenses, highlights the evolution of the receptor
infection with ΔpksP conidia was rescued (Fig. 5d). To further repertoire in innate cells towards the efficient recognition of
differentiate store depletion versus calcium entry, we evaluated fungal melanin32. Because the induction of glycolysis by melanin,
lactate secretion following calcium supplementation. In these but not MelLec expression, is evolutionarily conserved between

a
Ctrl Δku80 ΔpksP ΔrodA/pksP Melanin-coated
ΔrodA/pksP
b c P = 0.0011
Ctrl 20 P = 0.035
240 Δku80
15
P = 0.0021
Fluorescence intensity
ΔpksP
Lactate (mM)
Lactate (mM)
15 P = 0.013
210 ΔrodA/pksP
10
Melanin-coated
180 ΔrodA/pksP 10
(ΔAU)
P = 0.0008
P = 0.0005
150 5
P = 0.029
P = 0.014
120 0 0
Ca2+ free Ctrl
90 Ctrl Δku80
Δku80 Δku80 + EGTA-AM
ΔpksP ΔpksP
10
20
30
40
0
50
60
15
25
35
45
55
5
ΔpksP + EGTA-AM
Time (min)
d P < 0.0001 e f g P = 0.0021

P = 0.041
25 P = 0.0021 20 P < 0.0001 15 20
P = 0.032
20
Lactate (mM)
Lactate (mM)
Lactate (mM)
Lactate (mM)
15 P < 0.0001 15 P = 0.0096
P < 0.0001
15 10
10 10
10
5
5 5 5
0 0 0 0
Ca2+ free Ca2+ free Ca2+ free
P
P
Δp 80
Δp 80
u l
u l
Δk Ctr
Δk Ctr
ks
ks
Ctrl Ctrl + CaCl2 Ctrl
Δku80 Δku80 + CaCl2 Δku80
Δku80 + TG ΔpksP + CaCl2 siNC siSTIM1 Δku80 + W7
ΔpksP ΔpksP
ΔpksP + TG ΔpksP + W7
Δku80 ΔpksP
Δku80 + W7 ΔpksP + W7
mTOR DAPI A. fumigatus
i j IL-1β TNF IL-6

mTOR+ phagosomes (%)
P = 0.0049
30 P < 0.0001
20 30 P = 0.0002
250
P = 0.045
Cytokine (pg/mL)
P = 0.0021 P = 0.0011 200

P = 0.014
15
20 20
150
10 P = 0.024
100
10 10
5 50
0 0 0 0
Δku80
–
–
TG
TG
TG
7
7
W
Δku80 + W7
ΔpksP ΔpksP ΔpksP ΔpksP
ΔpksP + W7
human and murine macrophages, MelLec appears to be largely pathogens that are sensed by the host to weigh the intensity of
redundant for their metabolic reprogramming. Instead, the specific antimicrobial responses50. It is noteworthy that the bac-
germination associated with the active removal of melanin in terial vita-PAMP cyclic-di-adenosine monophosphate triggers
live fungi within the phagosome is required for host cells to stress-mediated autophagy to restrain phagocyte death51, and
reorient their metabolism towards protection. As such, melanin these molecular pathways are also well-known targets for the
can be considered a vita-PAMP, molecules expressed by viable action of melanin during infection47. How is the outer layer of the

Fig. 5 Calcium signaling regulates glucose metabolism in response to A. fumigatus. a Micrographs of macrophages preloaded with Fluo-4-AM and left
untreated (Ctrl) or infected with the Δku80, ΔpksP, ΔrodA/pksP, or melanin-coated ΔrodA/pksP strains for 15 min (representative of three independent
experiments). Scale bars, 35 µm. b Quantification of Fluo-4-AM fluorescence in 5 min intervals for 1 h, and expressed as the difference in arbitrary units
(ΔAU) between infected and uninfected cells. c Lactate secretion by macrophages left untreated (Ctrl) or infected with the Δku80 or ΔpksP strains for 24 h
using calcium-free medium (n = 5) or 500 µM EGTA-AM (n = 4). d Lactate secretion by macrophages left untreated (Ctrl) or infected in calcium-free
medium for 24 h and in the presence of 2 µM thapsigargin (TG) (n = 5) or e 2 mM CaCl2 (n = 5). f Lactate secretion by macrophages left untreated
(Ctrl) or infected and silenced with a STIM1 siRNA (siSTIM1) (n = 4). A scrambled siRNA was used as negative control (siNC). g Lactate secretion by
macrophages left untreated (Ctrl) or infected for 24 h in the presence of 25 µM W7 (n = 5). h Immunofluorescent staining for mTOR and i quantification of
mTOR+ phagosomes in infected macrophages, without or with 25 µM W7 for 2 h (representative of two independent experiments). The white arrows
indicate the recruitment of mTOR to the phagosome. Scale bars, 10 µm. Data on quantification were determined by analyzing at least 200 phagosomes.
j Production of IL-1β, TNF, and IL-6 by macrophages infected with the ΔpksP strain for 24 h without or with 2 µM TG or 25 µM W7 (n = 5). Data are
expressed as mean values ± SEM. P-values were calculated using Student’s two-tailed t test or one-way ANOVA and two-way ANOVA with Tukey’s
multiple comparisons test.
cell wall remodeled to shed melanin during germination and and HIF-1α signaling. Our findings are supported by proteomics
which signals are generated remains unclear. Could there be an of conidia-containing phagolysosomes55, in which melanin was
intracellular receptor other than MelLec involved in melanin revealed as a major regulator of mTOR activator/regulator
recognition? It is possible, since fungal ligands, namely β-(1,3)- (LAMTOR), a member of the Ragulator/LAMTOR complex
glucan, directly activate the inflammasome upon their release known to regulate mTOR56. Importantly, mTOR suppresses
during infection52, and melanin could behave analogously. In factor inhibiting HIF-1 (FIH-1), a negative regulator of HIF-1α57,
fact, both the metabolic reprogramming of macrophages and and FIH-1 may represent one central node congregating mTOR
inflammasome activation53 rely exclusively on infection by live and HIF-1α signaling towards the activation of glycolysis in
conidia, and melanin particles have been occasionally detected in response to infection. Myeloid HIF-1α has been shown to be
the cytosol of infected cells (Chamilos, unpublished), raising the required for protection against aspergillosis, a phenotype attrib-
intriguing possibility that inflammasome activation and induction uted to impaired chemokine production and enhanced neutrophil
of glycolysis in macrophages are molecular events coupled via apoptosis, resulting in a net decrease of neutrophil numbers58.
melanin-mediated signaling. Because HIF-1α also regulated cytokine production and the
The recognition of β-1,3-glucan from the cell wall is required metabolic activity of human dendritic cells infected with A.
for the metabolic reprogramming of myeloid cells in response to fumigatus59, it may deliver broader implications to antifungal
C. albicans11,20. Surface exposure of β-1,3-glucan on A. fumigatus immunity, with consequences to different cell types and involving
appears instead largely redundant for macrophage glycolysis distinct effector mechanisms.
given its broad expression on the surface of albino conidia47. Our What are the molecular signals that bridge melanin removal
data nevertheless suggest a minor contribution from signals other and the mTOR/HIF-1α axis culminating with the metabolic
than melanin. Whether these include other cell wall poly- reprogramming of macrophages? We demonstrate that the
saccharides or fungal-derived products requires further investi- immunometabolic reprogramming induced by A. fumigatus
gation. The differential activation of glucose metabolism upon depends on calcium sequestration inside the phagosome by
exposure of host cells to distinct fungal PAMPs also reflects melanin which in turn triggers glycolysis-promoting signals.
important differences in the outcome of the host–fungus inter- Calcium-mediated regulation of glycolysis has been known for
action. In certain cases, fungi can hijack nutritional weaknesses of several decades; for example, calcium overload was observed
host cells that are a direct consequence of immunometabolic concomitantly to glycolytic inhibition in heart tissue60, whereas
shifts to their own benefit; the terminal commitment of macro- calcium depletion promoted increased levels of glycolytic
phages to glycolysis after infection with C. albicans is subverted intermediates in cerebral cortex slices61. We demonstrate that
by the efficient shift of fungal metabolism to rapidly deplete calcium/CaM signaling orchestrates mTOR recruitment to the
glucose and trigger cell death22. Because of lacking nutrients in phagosome, an effect directly regulated by fungal melanin.
the lung microenvironment, A. fumigatus resorts to different Phagosome and entotic vacuole fission depend on mTOR
strategies to survive, which include the cross-pathway control localization to vacuolar membranes surrounding engulfed
system and degradation of proteins to acquire amino acids and cells62 and, likewise, we propose that the subcellular redis-
the glyoxylate cycle to produce carbohydrates from lipids avail- tribution of mTOR to melanin-containing phagosomes is
able at the site of infection54. In contrast to C. albicans, the required for its activation. Because the spatial organization of
metabolic reprogramming of macrophages in response to A. mTOR is coordinated through a variety of sensors and reg-
fumigatus represents instead an advantageous mechanism of host ulators that converge on Rag GTPases63, the phagosomal reg-
defense, highlighting the immunometabolic repurposing of ulation of LAMTOR by melanin55 suggests similar mechanisms
immune cells toward enhanced glycolysis as an attractive ther- during fungal infection, in line with the ability of melanized
apeutic strategy in aspergillosis. conidia to inhibit phagosomal recruitment of CaM, and prevent
Despite the remarkable differences in the nature of the fungal the specialized autophagy pathway LC3-associated phagocytosis
ligands and the mechanisms through which these deliver signals (LAP)43. By deploying melanin, the fungus promotes calcium
required to induce glycolysis, the pathways that are activated in sequestration inside the phagosome and inhibits the recruitment
response to different fungal pathogens appear to converge on a of LAPosome machinery components, including the NADPH
common signaling axis involving mTOR and HIF-1α. This oxidase and CaM, to block LAP and avoid elimination43,47. Our
pathway represents a major signaling hub that regulates meta- data demonstrate that melanin, by blocking the activation of
bolic changes underlying trained immunity in myeloid cells in LAP, contributes to fungal persistence within the phagosome,
response to β-1,3-glucan11,17–19. Although trained immunity allowing conidial germination and removal of melanin, pro-
induced by melanin has not been explored, we provide evidence viding the necessary signals to redirect macrophage metabolism
that this pigment is endowed with the ability to regulate mTOR towards glycolysis and efficient fungal clearance.

Our study provides mechanistic insights into the interplay stimulating factor (GM-CSF, Miltenyi Biotec) or 20 ng/mL of macrophage colony-
between A. fumigatus and the glucose metabolism of immune stimulating factor (M-CSF, Miltenyi Biotec). The culture medium was replaced
every 3 days, and acquisition of macrophage morphology was confirmed by
cells. It has, however, limitations, such as the likely involvement visualization in a BX61 microscope (Olympus).
of metabolic pathways other than glycolysis in the activation of
antifungal immunity. Cellular metabolism consists of highly
Generation of BMDMs. Bone marrow-derived macrophages (BMDMs) were
interconnected pathways that feed into each other and it is obtained from femur and tibia bones of male or female 8-week-old C57BL/6 or
challenging, if not at all impossible, to analyze them in an entirely Hif1afl/fl-LysMcre+/+, hereafter referred as HIF-1c, mice. Briefly, bone marrow was
independent way. Nevertheless, our results suggest a dominant harvested and cultured in Dulbecco’s modified medium (DMEM) with 1% peni-
role for glycolysis and represent a promising first step toward the cillin/streptomycin, 1% L-glutamine, and 10% FBS (Gibco, Thermo Fisher Scien-
tific), supplemented with 20 ng/mL M-CSF (Peprotech) for 7 days at 37 °C and 5%
elucidation of the metabolic profiles involved in the activation of CO2, with the addition of 20 ng/mL of M-CSF at day 4 of differentiation. Acqui-
antifungal immunity. Current therapeutic limitations and con- sition of macrophage morphology was confirmed by visualization in a BX61
cerns over the emergence of antifungal resistance are inspiring the microscope (Olympus).
search for novel host-directed therapies. Understanding how
metabolic networks coordinate immune cell function may lead to Cell stimulations and treatments. Unless otherwise indicated, MDMs or BMDMs
innovative therapeutic approaches or metabolic adjuncts to (5 × 105/well in 24-well plates) were infected with A. fumigatus conidia at a 1:2 or
reorient host cells towards immune protection against infection. 1:10 effector-to-target ratio or stimulated with 100 µg/mL of melanin ghosts or
with 50, 75, 100, and 200 µg/mL 1,8-DHN for 24 h at 37 °C and 5% CO2. For
experiments involving glucose depletion, RPMI-1640 medium without glucose
Methods (Thermo Fisher Scientific) was used. To interfere with cellular metabolism, MDMs
Ethics statement. The functional experiments involving cells isolated from the were pretreated for 1 hr with 5, 10, or 20 mM 2-DG, 30 µM 3PO, 500 nM 6-AN,
peripheral blood of healthy volunteers at Hospital of Braga, Portugal, were 10 µM wortmannin, or 10 nM rapamycin, and for 3 h with 50 µM (+)-sodium L-
approved by the Ethics Subcommittee for Life and Health Sciences (SECVS) of the ascorbate. For experiments involving calcium manipulation, MDMs were pre-
University of Minho, Portugal (no. 014/015). Experiments were conducted treated for 30 min with 2 µM thapsigargin, or treated for 10 min with 25 µM W7
according to the principles expressed in the Declaration of Helsinki, and partici- (all from Sigma-Aldrich) in calcium-free DMEM (Thermo Fisher Scientific) or for
pants provided written informed consent. 1 hr with 500 µM EGTA-AM (Thermo Fisher Scientific) after infection. In some
conditions, MDMs were cultured for 1 h with calcium-free DMEM followed by the
addition of 2 mM CaCl2 during infection. To assess cell viability, 4 h prior to the
Mice. Eight-week-old gender- and age-matched C57BL/6 mice were bred under end of infection, 50 µL of alamarBlueTM Cell Viability Reagent (Thermo Fisher
specific pathogen-free condition and kept at the Life and Health Sciences Research Scientific) were added to each well. Viability was assessed by the quantification of
Institute (ICVS) Animal Facility. Mice were fed ad libitum and kept under light/ relative fluorescence units (RFU) using a Varioskan Flash fluorescent plate reader
dark cycles of 12 h, temperature of 18–25 °C and humidity of 40–60%. Animal (Thermo Fisher Scientific) with a fluorescence excitation wavelength of 570 nm and
experimentation was performed following biosafety level 2 (BSL-2) protocols an emission wavelength of 600 nm. In some experiments, viability was evaluated by
approved by the Institutional Animal Care and Use Committee (IACUC) of flow cytometry using annexin-V/propidium iodide 6 h after infection. In all
University of Minho, and ethical and regulatory approvals were consented by experiments involving MDMs, the data were assessed in triplicates, and are shown
SECVS (no. 074/016). All procedures in vivo followed the EU-adopted regulations as the mean value for each individual.
(Directive 2010/63/EU), and were conducted according to the guidelines sanc-
tioned by the Portuguese ethics committee for animal experimentation, Direção-
Geral de Alimentação e Veterinária (DGAV). siRNA-mediated gene silencing. MDMs (5×104/well in 96-well plates) were
incubated for 72 h at 37 °C and 5% CO2 in Accell Delivery Media in the presence of
either 1 µM STIM1 siRNA or a non-targeting siRNA control (siNC) (Dharmacon).
Aspergillus strains and culture conditions. The A1163 Δku8064 and B-523365
After incubation, the transfection media was removed, and cells were infected with
strains of A. fumigatus were used as wild-type strains. The ΔrodA, ΔpksP, and A. fumigatus conidia at a 1:10 effector-to-target ratio. Pooled replicates from three
ΔrodA/pksP deletion mutants43,47 and the mutant strains in the DHN–melanin different individuals were collected after 24 h to measure lactate secretion and
biosynthetic pathway (ΔpksP, Δayg1, Δabr1, Δabr2, Δarp1, and Δarp2) in the B-
cytokine production. The mRNA knockdown was confirmed by qPCR.
5233 background66 were generated previously. Wild-type CBS110.46 and
CBS386.75 strains with albino conidia38 were also used, as indicated. All strains
were grown on 2% malt extract agar or YAG agar for 7 days at 28 °C. The conidia RNA sequencing. MDMs (5 × 105/well in 24-well plates) were infected with A.
were harvested from agar slants using phosphate buffer saline (PBS) (Gibco, fumigatus conidia at a 1:2 effector-to-target ratio, and pooled replicates from three
Thermo Fisher Scientific) with 0.05% Tween 20 (Sigma-Aldrich), followed by different individuals were collected after 2 and 6 h. Uninfected MDMs were cul-
gentle agitation and subsequent filtration through a 40-μm-pore size cell strainer tured in parallel as controls. Sample processing, sequencing, and analysis were
(Falcon). The concentration of conidia/mL was determined by counting in a performed at IMGM Laboratories GmbH (Germany). Briefly, the total RNA was
Neubauer chamber. Swollen conidia and germ tubes were obtained after incubation isolated using the RNeasy Mini Kit (Qiagen) according to the manufacturer’s
in Sabouraud liquid culture medium at 37 °C for 4 h and 6 h, respectively. Heat instructions, including on-column DNase digestion. The total RNA was eluted in
inactivation of conidia was performed by incubation for 30 min at 90 °C, whereas 30 μL of RNase-free water. The quality of the total RNA was analyzed with the 2100
UV inactivation was performed by exposing conidia to UV light for 3 h. Bioanalyzer using RNA 6000 Nano and Pico LabChip kits (Agilent Technologies).
Library preparation was performed using the TruSeq® Stranded mRNA HT tech-
nology, according to the manufacturer’s protocol. All single libraries were pooled
Extraction of A. fumigatus melanin and conidial coating. Melanin from A1163
into a final sequencing library with an equal DNA amount per sample. The final
Δku80 conidia was isolated using a combination of proteolytic (proteinase K;
sequencing library generated by pooling was quantified using the highly sensitive
Sigma-Aldrich) and glycohydrolytic (Glucanex; Novo) enzymes, denaturant (gua-
fluorescent dye-based Qubit® dsDNA HS Assay kit (Invitrogen) before sequencing
nidine thiocyanate) and hot, concentrated HCl (6 M) to treat conidia, resulting in
at a final concentration of 1.8 pM and with a 1% PhiX v3 control library spike-in
an electron-dense layer similar in size and shape to the original conidial melanin
(Illumina) on the NextSeq500 sequencing system (Illumina). For the clustering and
layer but without the underlying cell components, named melanin ghosts67.
sequencing of samples, a high-output single-end 75 cycles (1 × 75 bp SE) run was
Melanin-coated ΔrodA/pksP conidia were obtained by overnight coating of conidia
performed under the control of the NextSeq Control Software (NCS, Illumina).
with different concentrations of melanin ghosts, previously sonicated with pulses of
Quality control was carried out using NCS and Real Time Analysis 2.4.11 softwares
5 s during 1 min, at room temperature (RT).
applying the FastQ only pipeline. Read data were imported into the CLC Genomics
Workbench (CLC bio, Qiagen), and reads were mapped against the human
Isolation of PBMCs and generation of MDMs. Peripheral blood mononuclear reference genome (GRCh37.p13) with subsequent counting and distribution of
cells (PBMCs) were enriched from buffy coats or whole blood by density gradient reads across genes and transcripts. The expression values were then processed to
using Histopaque®-1077 (Sigma-Aldrich), washed twice in PBS and resuspended in reads per kilobase million (RPKM), a normalized measure of relative abundance of
RPMI-1640 culture medium with 2 mM glutamine (Gibco, Thermo Fisher Scien- transcripts68, followed by analysis using the EdgeR Bioconductor package69 to
tific) supplemented with 10% human serum (Sigma-Aldrich), 10 U/mL penicillin/ identify differentially expressed genes with a fold change value ≥2 or ≤ −2 with a
streptomycin and 10 mM HEPES (Thermo Fisher Scientific) (cRPMI). Monocytes false discovery rate (FDR)-corrected p-value < 0.05. Heatmaps were created for the
were isolated from PBMCs using positive magnetic bead separation with anti- most significantly represented genes of a specific functional class using the Mor-
CD14+-coated beads (MACS Miltenyi) according to the manufacturer’s instruc- pheus tool (Broad Institute; https://software.broadinstitute.org/morpheus/). For
tions. Isolated monocytes were resuspended in cRPMI medium, and seeded at a pathway analysis, the annotated hallmark gene sets from the Molecular Signatures
concentration of 1 × 106 cells/mL in 24-well and 96-well plates (Corning Inc.) and Database (MSigDB)33 were used and enrichment analysis was performed using the
eight-well chamber slides (LAB-TEK, Thermo Fisher Scientific) for 7 days in the Gene Ontology Biological Processes category in the Gene Ontology Consortium
presence of 20 ng/mL recombinant human granulocyte macrophage colony- software (http://www.geneontology.org/).

Quantification of glucose and lactate by HPLC. After infection, supernatants and digested at 37 °C for 30 min in iDMEM culture medium containing 1 mg/mL
were removed, centrifuged, and transferred to HPLC tubes. Glucose and lactate of collagenase D (Sigma-Aldrich). Subsequently, the tissue was forced through a
levels were determined using a Gilson pump system (Gilson) with a 54 °C 70-μm cell strainer, and contaminating red blood cells were lysed. Leukocytes were
HyperREZ XP Carbohydrate H+ 8 μM (Thermo Fisher Scientific) column and a isolated by Percoll (GE Healthcare Bio-Sciences Ab) density gradient, and finally
refractive index detector (IOTA 2, Reagents). The mobile phase consisting of resuspended in FACS buffer (PBS containing 2% FBS and 2 mM EDTA). To assess
0.0025 M H2SO4 was filtered and degasified for at least 45 min before use. Standard cell viability, cells were stained for 30 min in the dark with the Zombie Violet
solutions were prepared in Milli-Q water (Millipore). All data were analyzed using fluorescent dye (BioLegend) and resuspended in FACS buffer. For surface marker
the Gilson Uniprot Software, version 5.11. staining, cell suspensions were stained for 30 min on ice while protected from light
with the indicated antibodies. Pellets were washed and resuspended in fresh FACS
buffer prior to analysis. Gating for myeloid subpopulations in the lung was per-
LC–MS/MS-targeted metabolomics. MDMs (2 × 106/well, in six-well plates)
formed to analyze the composition of lung-infiltrating cells using a combination of
were infected with A. fumigatus conidia at a 1:2 effector-to-target ratio for 6 h at
the following antibodies: BV510 anti-mouse CD45 (clone 30-F11), BV605 anti-
37 °C in 5% CO2. To detach the cells, the bottom of the well was scraped lightly
mouse CD11c (clone N418), PE-Cy7 anti-mouse CD11b (clone M1/70), and APC
with a cell scraper. The resulting cell suspensions, obtained from pooled replicates
anti-mouse F4/80 (clone BM8) (all from BioLegend)22. Data were obtained on a
from three different individuals, were centrifuged at 4 °C, the supernatant was
BD FACS LSRII instrument (Becton Dickinson), and processed using FlowJo (Tree
discarded, and the resulting pellet was immediately frozen in liquid nitrogen to
Star Inc). For the sorting of macrophage populations in the lung, CD45 + positive
quickly quench the metabolism. To maximize overall metabolite yield, 280 μL of
cells were isolated using positive magnetic bead separation with anti-CD45+ coated
MeOH:MTBE (Methyl tert-butyl ether) (4:1, v/v) was added to the pellet, followed
beads (MACS Miltenyi) before surface marker staining. Macrophages were sorted
by three cycles of freezing/thawing in liquid nitrogen and in cold-water bath,
using a combination of the following antibodies: PerCP/Cy5.5 anti-mouse CD11c
respectively, for 10 s each. A sonication at 15 W for 6 min was then performed,
(clone N418), APC anti-mouse CD11b (clone M1/70), PE anti-mouse Siglec-F
followed by 1 min of vortexing and a centrifugation at 4 °C. From the resulting
(clone E50-2440), APC-Cy7 anti-mouse Ly-6G (clone 1A8), FITC anti-mouse
supernatant, a volume of 250 μL was carefully transferred to a new Eppendorf tube.
CD64 (clone X54-5/7.1), and PE/Cy7 anti-mouse CD45 (clone 30-F11) (from
In addition, 280 μL of MeOH:H2O (4:1, v/v) was added to the pellet, and the same
BioLegend or BD Biosciences)70. Data were obtained on a BD FACSAria II
procedure above was performed. At the end, 250 μL of the resulting supernatant
instrument and analyzed with the FACSDiva software (Becton Dickinson).
was mixed with the first 250 μL of supernatant. The samples were stored at −80 °C
until further analysis. To perform the metabolomics analysis, standards of analy-
tical grade (Sigma-Aldrich) were used for external calibration and adjusted to the Measurement of ADP/ATP ratio. MDMs (1 × 105/well in 96-well plates) were
corresponding levels in the samples. Ultrapure water, purified using a Milli-Q infected with A. fumigatus conidia for 6 h at a 1:10 effector-to-target ratio at 37 °C
system (Millipore), was used for buffer preparation. Liquid chromatography–mass in 5% CO2. After infection, the ADP/ATP ratio was measured using an assay kit
spectrometry (LC–MS)- grade methanol and acetonitrile (both from Thermo (Sigma-Aldrich), according to the manufacturer’s instructions. Briefly, culture
Fisher Scientific) were used. Tributylamine (≥99.5%) (Sigma-Aldrich) stock solu- medium was removed, and the ATP reagent was added to each well. Plate was
tions of all the standards were prepared at 1000 ppm in water, and stored at incubated for 1 min at RT, and luminescence was read to obtain ATP levels. After
−20 °C. The LC–MS/MS analyses were performed in an Agilent 1290 Infinity 10 min of incubation at RT, the ADP reagent was added to each well, and lumi-
(Agilent Technologies) using a 1290 Infinity Binary Pump (1200 bar) and a 1260 nescence was read. The ADP/ATP ratio was calculated by subtracting background
Infinity Quaternary Pump (400 bar). The LC system was coupled to an Agilent values from ADP values and dividing the result by ATP values. Luminescence was
6460 triple-quadrupole mass spectrometer using an electrospray ionization (ESI) read in a Fluoroskan FL Microplate Luminometer (Thermo Fisher Scientific).
interface working in multiple reaction monitoring (MRM) mode. The chromato-
graphic method, the MS parameters and the setup arrangement were based on that
Phagocytosis assay. To determine phagocytosis, MDMs (5 × 105/well in 24-well
described by Agilent Technologies, Inc, with minor modifications. Briefly, this
plates) were infected with fluorescein isothiocyanate (FITC)-labeled conidia of A.
method uses tributylamine as an ion-pairing reagent, with buffer A composed of
fumigatus at a 1:5 effector-to-target ratio. The infection was synchronized for 30
97% water and 3% methanol, 10 mM tributylamine, 15 mM glacial acetic acid
min at 4 °C, and phagocytosis was initiated by shifting the co-incubation to 37 °C at
(VWR), and buffer B composed of 10 mM tributylamine, 15 mM glacial acetic acid,
5% CO2 for 1 h. Phagocytosis was stopped by washing wells with PBS, and
prepared in methanol. The transitions showing the highest signal to noise ratios
extracellular conidia were stained with 0.25 mg/mL Calcofluor White (Sigma-
were used for the quantification of the analytes in samples.
Aldrich) for 15 min at 4 °C to avoid further ingestion. Wells were then washed
twice with PBS, and cells were fixed with 3.7% (v/v) formaldehyde/PBS for 15 min.
Mouse infection. For the in vivo studies, at day 3 before infection and until the The number of MDMs with ingested green conidia was enumerated by examining
end of experimental protocol, 100 mg/Kg of 2-DG was administered to mice daily the slides by fluorescence microscopy (Olympus), and data were expressed as
via intraperitoneal route (i.p.). Control groups consisted of mice to which an equal percentage of MDMs that internalized one or more conidia.
volume of vehicle (sterile PBS) was administered. At day 0, mice were challenged
with 1 × 108 live conidia of A. fumigatus (A1163 Δku80 strain) using a noninvasive
Conidiacidal activity assay. Following differentiation, MDMs (1 × 105/well in 96-
intranasal (i.n.) infection procedure upon anesthesia with 75 mg/Kg of ketamine
well plates) were washed twice with cRPMI medium, and a suspension of A.
(Ketamidor®, Ritcher Pharma) and 1 mg/Kg of medetomidine (Domtor®, Ecuphar).
fumigatus conidia was added at a ratio of 10:1 effector-to-target ratio. The cells
At 24 h post infection, mice were killed, and the lungs were PBS-perfused and
were then incubated for 1 h at 37 °C and 5% CO2 to allow the internalization of
excised, excluding the trachea and major bronchi. For assessment of fungal burden,
conidia. Medium containing the non-ingested conidia was removed, and wells were
lung single-cell suspensions were serially diluted and plated on solid growth media.
washed twice with pre-warmed PBS. To measure the conidiacidal ability, MDMs
were allowed to kill the ingested conidia for 2 h at 37 °C in 5% CO2. To determine
Determination of systemic glucose. On the day of the experiment, mice were the conidiacidal activity of splenocytes, cells (1 × 106/well in 96-well plates) were
starved for 3–4 h in the morning before measurements of blood glucose levels at infected with A. fumigatus conidia at a 1:5 effector-to-target ratio for 2 h at 37 °C
0 h, by snipping the very end of the tail to collect a drop of blood in a Glucocard®+ and 5% CO2. After incubation, culture plates were snap frozen at −80 °C, and
(Arkray), which was read using the ACCU-CHEK Performa glucometer (Roche). thawed at 37 °C to cause cell lysis and release of ingested conidia. Serial dilutions of
Afterwards, mice were infected intranasally with 1 × 108 conidia of the Δku80 cell lysates were plated on solid growth media and, following a 24 h incubation at
strain with ad libitum access to food. After 20 h of infection, mice were again 37 °C, the number of colony-forming units (CFUs) was enumerated, and the
starved for 3–4 h before glucose measurement in the blood. The systemic glucose percentage of CFU inhibition was calculated.
concentration was calculated as the glucose (mM) per gram of mice body weight.
Measurement of ROS production. MDMs (1 × 105/well) were plated in 96-well
Isolation of splenocytes. Infected animals were sacrificed at day 1 post infection TC-treated dark clear bottom plates (Sigma-Aldrich) and infected with live A.
with posterior spleen excision. The excised spleen was minced into small fragments fumigatus conidia at a 1:3 effector-to-target ratio. Infections were synchronized by
with a plunger end of a syringe and forced through a 70-μm cell strainer (Corning centrifugation at 1000 rpm for 5 min, and then 10 μM dihydrorhodamine 123
Inc.) Upon washing the cells with cold sterile PBS, the resulting cell pellet was (DHR) (Thermo Fisher Scientific) was added to each well, and the production of
resuspended in 2 mL of pre-warmed ACK lysis buffer (0.15 M NH4Cl, 10 Mm reactive oxygen species (ROS) was measured for a 24-h period using the Varioskan
KHCO3, and 0.1 mM EDTA). Afterwards, the cells were centrifuged at 1600 rpm Flash fluorescent plate reader (Thermo Fisher Scientific). Excitation was performed
for 5 min at RT. Finally, the cells were counted and adjusted to a final con- at a wavelength of 480 nm, and emission was measured at a wavelength of 528 nm.
centration of 5 × 106 cells/mL of pre-warmed RPMI (Gibco, Thermo Fisher Sci-
entific) supplemented with 10% FBS (Gibco, Thermo Fisher Scientific) and 200 µL
Cytokine measurements. MDMs and BMDMs (5 × 105/well in 24-well plates) and
of the cellular suspension were seeded in round bottom 96-well plates
splenocytes (1 × 106/well in 96-well plates) were infected with A. fumigatus conidia
(Corning Inc.).
at a 1:10 effector-to-target ratio for 24 h at 37 °C and 5% CO2. PBMCs (5 × 105/well
in 96-well plates) were infected with UV-inactivated conidia at a 1:4 effector-to-
FACS analysis and sorting. After infection, the lungs were excised and collected target ratio for 7 days at 37 °C and 5% CO2. After infection, supernatants were
in incomplete Dulbecco’s Modified Eagle Medium (iDMEM) culture medium collected and cytokine levels were quantified using ELISA MAX Deluxe Set kits
(Gibco, Thermo Fisher Scientific). Perfused lungs were chopped in small fragments (BioLegend), according to the manufacturer’s instructions. Quantitative cytokine

measurements were also performed on the supernatants of lung single-cell sus- used, and imaging was performed at RT. Unless otherwise stated, mean projections
pensions 24 h post infection. of image stacks were obtained using the LCS Lite software and processed with
Adobe Photoshop CS2. Phagosomes surrounded by a rim of fluorescence of the
indicated protein-marker were scored as positive43,47. At least 200 phagosomes
RNA isolation and qRT-PCR. The total RNA from MDMs (5 × 105/well in 24-well were analyzed for each condition in two independent experiments.
plates) was isolated at different time points after infection with A. fumigatus at a
1:10 effector-to-target ratio using the PureLinkTM RNA Mini Kit (Thermo Sci-
entific) according to the manufacturer’s instructions. The total RNA from lungs Statistical analysis. The data were expressed as means ± SEM. Statistical sig-
was extracted using the GRS Total RNA Kit-Tissue (Grisp) at different time points nificance of differences were determined by two-tailed Student’s t test, one-way
after infection, according to the manufacturer’s instructions. The concentration ANOVA, or two-way ANOVA with post hoc tests for multiple comparisons (P <
and quality of total RNA in each sample were determined by spectrophotometry 0.05 was considered statistically significant). Analyses were performed in GraphPad
using the ND-100 UV-visible light spectrophotometer (NanoDrop). One micro- Prism software.
gram of the total RNA was retro-transcribed using the first-strand cDNA Synthesis
Kit (Nzytech). Quantitative PCR was performed in an Applied Biosystems 7500 Data availability
Fast qPCR system (Applied Biosystems, Thermo Fisher Scientific), using the RNA-seq data have been deposited in Gene Expression Omnibus (GEO) with the
PowerUp SYBR Green Master Mix (Applied Biosystems, Thermo Fisher Scientific). accession code GSE128661. The Molecular Signatures Database version 7.1, a collection
Data were analyzed using the 7500 Software v2.0.6 software (Applied Biosystems, of annotated gene sets for use with GSEA software, was used in the study [https://www.
Thermo Fisher Scientific). Amplification efficiencies were validated, and the gsea-msigdb.org/gsea/msigdb/index.jsp]. All other data, materials, and reagents are
expression levels of the transcripts were normalized using the ACTB (human), Ubb
available on request from the corresponding author. The source data underlying
(mouse) and 18S (A. fumigatus) genes.
Fig. 1d–h, Fig. 2, Fig. 3, Figs. 4b–d and 4f–h, Fig. 5, and Supplementary Fig. 1, Fig. 2,
Fig. 3, Fig. 4, and Fig. 5c–e are provided as a Source Data file.
Western blot analysis. Human MDMs (5 × 105/well in 24-well plates) were
infected with A. fumigatus conidia for the indicated time points at a 1:10 Received: 19 July 2019; Accepted: 14 April 2020;
effector-to-target ratio at 37 °C in 5% CO2. After infection, cells were lysed in
RIPA buffer (50 mM Tris, 250 mM NaCl, 2 mM EDTA, 1% NP-40, 10% glycerol,
pH 7.2, and a mixture of protease inhibitors [Roche Molecular Biochemicals]).
Cell lysis was performed at 4 °C for 30 min (with shaking), and samples were
then centrifuged. The protein content was determined using the Bradford dye-
binding (Bio-Rad) method. Laemmli buffer (Bio-Rad) was added to 20 μg of
protein, and samples were boiled and separated on a 12% SDS-PAGE gel, and References
transferred to nitrocellulose membranes (Bio-Rad). Western blotting was per- 1. O’Neill, L. A. & Pearce, E. J. Immunometabolism governs dendritic cell and
formed according to the manufacturer’s instructions, using the following pri- macrophage function. J. Exp. Med. 213, 15–23 (2016).
mary antibodies: rabbit anti-phospho-p70S6 Kinase, rabbit anti-p70S6 Kinase, 2. Van den Bossche, J., O’Neill, L. A. & Menon, D. Macrophage
rabbit anti-Akt, rabbit anti-phospho-Akt, rabbit anti-mTOR (all from Cell immunometabolism: where are we (going)? Trends Immunol. 38, 395–406
Signaling), mouse anti-β-actin (Abcam), all diluted 1:1000, and rabbit anti-HIF- (2017).
1α antibody (Abcam) diluted 1:500. Secondary antibodies used were anti-rabbit 3. Millet, P., Vachharajani, V., McPhail, L., Yoza, B. & McCall, C. E. GAPDH
and anti-mouse, both diluted to 1:5000. The blots were developed using che- binding to TNF-alpha mRNA contributes to posttranscriptional repression in
miluminescence (SuperSignal™ West Femto Maximum Sensitivity Substrate; monocytes: a novel mechanism of communication between inflammation and
Thermo Fisher Scientific), and detected with ChemiDocTM XRS system (Bio- metabolism. J. Immunol. 196, 2541–2551 (2016).
Rad). Signal intensities and quantifications were determined with the ImageLab 4. Palsson-McDermott, E. M. et al. Pyruvate kinase M2 regulates Hif-1alpha
4.1 analysis software (Bio-Rad). activity and IL-1beta induction and is a critical determinant of the warburg
effect in LPS-activated macrophages. Cell Metab. 21, 65–80 (2015).
Live cell imaging. To perform live imaging of calcium, MDMs were seeded in 5. Jha, A. K. et al. Network integration of parallel metabolic and transcriptional
eight-well-chamber slides (LAB-TEK, Thermo Fisher Scientific) (3 × 105/well) and data reveals metabolic modules that regulate macrophage polarization.
loaded with 3 μM of the calcium indicator Fluo-4AM (Thermo Fisher Scientific) Immunity 42, 419–430 (2015).
according to the manufacturer’s protocol. Briefly, MDMs were placed in serum-free 6. Lampropoulou, V. et al. Itaconate links inhibition of succinate dehydrogenase
HBSS (without Ca2+, MgCl2, and phenol red) and loaded with Fluo-4AM for 30 with macrophage metabolic remodeling and regulation of inflammation. Cell
min at 37 °C, 5% CO2. Cells were then washed twice with PBS, and infections with Metab. 24, 158–166 (2016).
the indicated fungal strains were performed in cRPMI medium. Live cell imaging 7. Mills, E. L. et al. Succinate dehydrogenase supports metabolic repurposing of
was performed on an Olympus FV1000 Plus Confocal Microscope (60× Luc Plan mitochondria to drive inflammatory macrophages. Cell 167, 457–470 e413
FL 0.70 NA objective) at 37 °C with 5% CO2. All analyses and processing were (2016).
made using ImageJ software (Fiji). To visualize the temporal changes in calcium, 8. Van den Bossche, J. et al. Mitochondrial dysfunction prevents repolarization
raw sequences were processed, and mean pixel intensity at each frame was mea- of inflammatory macrophages. Cell Rep. 17, 684–696 (2016).
sured. The data were first plotted as fluorescence intensity versus time (Z profile) 9. Tannahill, G. M. et al. Succinate is an inflammatory signal that induces IL-
and subsequently converted to relative scale (ΔF/F baseline). To perform imaging 1beta through HIF-1alpha. Nature 496, 238–242 (2013).
of HIF-1α, MDMs were seeded in eight-well-chamber slides (LAB-TEK, Thermo 10. Stienstra, R., Netea-Maier, R. T., Riksen, N. P., Joosten, L. A. B. & Netea, M. G.
Fisher Scientific) (3 × 105/well) and infected for 2 h with the indicated FITC-labeled Specific and complex reprogramming of cellular metabolism in myeloid cells
fungal strain at 37 °C in 5% CO2. After washing twice with PBS, the cells were fixed during innate immune responses. Cell Metab. 26, 142–156 (2017).
with 3.7% (v/v) formaldehyde/PBS for 15 min, permeabilized with 0.3% TritonTM 11. Cheng, S. C. et al. mTOR- and HIF-1alpha-mediated aerobic glycolysis as
X-100 (Sigma-Aldrich) in PBS for 10 min, blocked with 4% BSA (Sigma-Aldrich) metabolic basis for trained immunity. Science 345, 1250684 (2014).
for 1 h, and incubated overnight with primary antibody rabbit anti-HIF-1α anti- 12. Lachmandas, E. et al. Rewiring cellular metabolism via the AKT/mTOR
body (dilution 1:100, Abcam). MDMs were then washed twice with PBS and pathway contributes to host defence against Mycobacterium tuberculosis in
incubated with goat anti-Rabbit IgG (H + L) Cross-Adsorbed Secondary Antibody human and murine cells. Eur. J. Immunol. 46, 2574–2586 (2016).
Alexa Fluor® 568 conjugate (dilution 1:1000, Thermo Fisher Scientific) for 1 h at 13. Wickersham, M. et al. Metabolic stress drives keratinocyte defenses against
RT. Nuclei were stained with DAPI (dilution 1:1000, Thermo Fisher Scientific) for Staphylococcus aureus infection. Cell Rep. 18, 2742–2751 (2017).
10 min. Images were captured at ×100 magnification on an Olympus FV1000 Plus 14. Lachmandas, E. et al. Microbial stimulation of different Toll-like receptor
Confocal microscope. HIF-1α quantification was performed using ImageJ (Fiji), signalling pathways induces diverse metabolic programmes in human
and expressed as nuclear/total HIF-1α staining of at least 60 cells for each con- monocytes. Nat. Microbiol. 2, 16246 (2016).
dition. For immunofluorescence imaging of mTOR on phagosomes, MDMs were 15. Netea, M. G., Joosten, L. A., van der Meer, J. W., Kullberg, B. J. & van de
seeded on coverslips pretreated with polylysine, fixed with 4% PFA for 15 min at Veerdonk, F. L. Immune defence against Candida fungal infections. Nat. Rev.
RT following by 10 min of fixation with ice-cold methanol at −20 °C, washed twice Immunol. 15, 630–642 (2015).
with PBS, permeabilized using 0.1% saponin (Sigma-Aldrich) and blocked for 30 16. van de Veerdonk, F. L., Gresnigt, M. S., Romani, L., Netea, M. G. & Latge, J. P.
min PBS with 2% BSA. After incubation with anti-mTOR antibody (dilution 1:250;
Aspergillus fumigatus morphology and dynamic host interactions. Nat. Rev.
Cell Signaling) for 1 h, slides were washed twice in PBS–BSA and stained with the
Microbiol. 15, 661–674 (2017).
Alexa Fluor 555 secondary antibody (Molecular Probes), followed by DNA staining
17. Arts, R. J. et al. Glutaminolysis and fumarate accumulation integrate
with 10 µM TOPRO-3 iodide (642/661; Invitrogen). After the washing steps, slides
immunometabolic and epigenetic programs in trained immunity. Cell Metab.
were mounted in Prolong Gold antifading media (Molecular Probes). Images were
24, 807–819 (2016).
acquired using a laser-scanning spectral confocal microscope (TCS SP2; Leica),
LCS Lite software (Leica), and a ×40 Apochromat 1.25 NA oil objective using 18. Bekkering, S. et al. Metabolic induction of trained immunity through the
identical gain settings. A low fluorescence immersion oil (11513859; Leica) was mevalonate pathway. Cell 172, 135–146 e139 (2018).

19. Dominguez-Andres, J. et al. The itaconate pathway is a central regulatory 49. Volling, K., Thywissen, A., Brakhage, A. A. & Saluz, H. P. Phagocytosis
node linking innate immune tolerance and trained immunity. Cell Metab. 29, of melanized Aspergillus conidia by macrophages exerts cytoprotective
211–22 (2018). effects by sustained PI3K/Akt signalling. Cell. Microbiol. 13, 1130–1148
20. Dominguez-Andres, J. et al. Rewiring monocyte glucose metabolism via C- (2011).
type lectin signaling protects against disseminated candidiasis. PLoS Pathog. 50. Ugolini, M. & Sander, L. E. Dead or alive: how the immune system detects
13, e1006632 (2017). microbial viability. Curr. Opin. Immunol. 56, 60–66 (2018).
21. Cheng, S. C. et al. Broad defects in the energy metabolism of leukocytes 51. Moretti, J. et al. STING senses microbial viability to orchestrate stress-
underlie immunoparalysis in sepsis. Nat. Immunol. 17, 406–413 (2016). mediated autophagy of the endoplasmic reticulum. Cell 171, 809–823
22. Tucey, T. M. et al. Glucose homeostasis is important for immune cell viability e813 (2017).
during candida challenge and host survival of systemic fungal infection. Cell 52. Briard, B. et al. Fungal ligands released by innate immune effectors promote
Metab. 27, 988–1006 e1007 (2018). inflammasome activation during Aspergillus fumigatus infection. Nat.
23. Segal, B. H. Aspergillosis. N. Engl. J. Med. 360, 1870–1884 (2009). Microbiol. 4, 316–327 (2019).
24. Brown, G. D. et al. Hidden killers: human fungal infections. Sci. Transl. Med. 53. Karki, R. et al. Concerted activation of the AIM2 and NLRP3 inflammasomes
4, 165rv113 (2012). orchestrates host protection against Aspergillus infection. Cell Host Microbe.
25. Maertens, J. A. et al. Isavuconazole versus voriconazole for primary treatment 17, 357–368 (2015).
of invasive mould disease caused by Aspergillus and other filamentous fungi 54. Willger, S. D., Grahl, N. & Cramer, R. A. Jr. Aspergillus fumigatus
(SECURE): a phase 3, randomised-controlled, non-inferiority trial. Lancet metabolism: clues to mechanisms of in vivo fungal growth and virulence. Med.
387, 760–769 (2016). Mycol. 47(Suppl 1), S72–S79 (2009).
26. Herbst, S. et al. Phagocytosis-dependent activation of a TLR9-BTK- 55. Schmidt, H. et al. Proteomics of Aspergillus fumigatus conidia-containing
calcineurin-NFAT pathway co-ordinates innate immunity to Aspergillus phagolysosomes identifies processes governing immune evasion. Mol. Cell.
fumigatus. EMBO Mol. Med. 7, 240–258 (2015). Proteom. 17, 1084–1096 (2018).
27. Shah, A. et al. Calcineurin orchestrates lateral transfer of Aspergillus fumigatus 56. Scheffler, J. M. et al. LAMTOR2 regulates dendritic cell homeostasis through
during macrophage cell death. Am. J. Respir. Crit. Care Med. 194, 1127–1139 FLT3-dependent mTOR signalling. Nat. Commun. 5, 5138 (2014).
(2016). 57. Sakamoto, T. et al. Hypoxia-inducible factor 1 regulation through cross talk
28. Espinosa, V. et al. Inflammatory monocytes orchestrate innate antifungal between mTOR and MT1-MMP. Mol. Cell. Biol. 34, 30–42 (2014).
immunity in the lung. PLoS Pathog. 10, e1003940 (2014). 58. Shepardson, K. M. et al. Myeloid derived hypoxia inducible factor 1-alpha is
29. Cunha, C. et al. Dectin-1 Y238X polymorphism associates with susceptibility required for protection against pulmonary Aspergillus fumigatus infection.
to invasive aspergillosis in hematopoietic transplantation through impairment PLoS Pathog. 10, e1004378 (2014).
of both recipient- and donor-dependent mechanisms of antifungal immunity. 59. Fliesser, M. et al. Hypoxia-inducible factor 1alpha modulates metabolic activity
Blood 116, 5394–5402 (2010). and cytokine release in anti-Aspergillus fumigatus immune responses initiated
30. Cunha, C. et al. IL-10 overexpression predisposes to invasive aspergillosis by by human dendritic cells. Int J. Med. Mycol. 305, 865–873 (2015).
suppressing antifungal immunity. J. Allergy Clin. Immunol. 140, 867–870 e869 60. Corretti, M. C. et al. Glycolytic inhibition and calcium overload as
(2017). consequences of exogenously generated free radicals in rabbit hearts. J. Clin.
31. Gresnigt, M. S. et al. Genetic deficiency of NOD2 confers resistance to invasive Invest. 88, 1014–1025 (1991).
aspergillosis. Nat. Commun. 9, 2636 (2018). 61. Takagaki, G. Control of aerobic glycolysis and pyruvate kinase activity in
32. Stappers, M. H. T. et al. Recognition of DHN-melanin by a C-type lectin cerebral cortex slices. J. Neurochem. 15, 903–916 (1968).
receptor is required for immunity to Aspergillus. Nature 555, 382–386 (2018). 62. Krajcovic, M., Krishna, S., Akkari, L., Joyce, J. A. & Overholtzer, M. mTOR
33. Liberzon, A. et al. The molecular signatures database (MSigDB) hallmark gene regulates phagosome and entotic vacuole fission. Mol. Biol. Cell 24, 3736–3745
set collection. Cell Syst. 1, 417–425 (2015). (2013).
34. Parikh, H. et al. TXNIP regulates peripheral glucose metabolism in humans. 63. Shen, K. & Sabatini, D. M. Ragulator and SLC38A9 activate the Rag GTPases
PLoS Med. 4, e158 (2007). through noncanonical GEF mechanisms. Proc. Natl. Acad. Sci. USA 115,
35. Latge, J. P., Beauvais, A. & Chamilos, G. The cell wall of the human fungal 9545–9550 (2018).
pathogen aspergillus fumigatus: biosynthesis, organization, immune response, 64. da Silva Ferreira, M. E. et al. The akuB(KU80) mutant deficient for
and virulence. Annu. Rev. Microbiol. 71, 99–116 (2017). nonhomologous end joining is a powerful tool for analyzing pathogenicity in
36. Aimanianda, V. et al. Surface hydrophobin prevents immune recognition of Aspergillus fumigatus. Eukaryot. Cell 5, 207–211 (2006).
airborne fungal spores. Nature 460, 1117–1121 (2009). 65. Tsai, H. F., Washburn, R. G., Chang, Y. C. & Kwon-Chung, K. J. Aspergillus
37. Jahn, B. et al. Isolation and characterization of a pigmentless-conidium fumigatus arp1 modulates conidial pigmentation and complement deposition.
mutant of Aspergillus fumigatus with altered conidial surface and reduced Mol. Microbiol. 26, 175–183 (1997).
virulence. Infect. Immun. 65, 5110–5117 (1997). 66. Tsai, H. F., Wheeler, M. H., Chang, Y. C. & Kwon-Chung, K. J. A
38. Sarfati, J. et al. A new experimental murine aspergillosis model to identify developmentally regulated gene cluster involved in conidial pigment
strains of Aspergillus fumigatus with reduced virulence. Nihon Ishinkin Gakkai biosynthesis in Aspergillus fumigatus. J. Bacteriol. 181, 6469–6477 (1999).
Zasshi 43, 203–213 (2002). 67. Bayry, J. et al. Surface structure characterization of Aspergillus fumigatus
39. Langfelder, K., Streibel, M., Jahn, B., Haase, G. & Brakhage, A. A. Biosynthesis conidia mutated in the melanin synthesis pathway and their human cellular
of fungal melanins and their importance for human pathogenic fungi. Fungal immune response. Infect. Immun. 82, 3141–3153 (2014).
Genet. Biol. 38, 143–158 (2003). 68. Mortazavi, A., Williams, B. A., McCue, K., Schaeffer, L. & Wold, B. Mapping
40. Duvel, K. et al. Activation of a metabolic gene regulatory network downstream and quantifying mammalian transcriptomes by RNA-Seq. Nat. Methods 5,
of mTOR complex 1. Mol. Cell 39, 171–183 (2010). 621–628 (2008).
41. Dibble, C. C. & Cantley, L. C. Regulation of mTORC1 by PI3K signaling. 69. Robinson, M. D., McCarthy, D. J. & Smyth, G. K. edgeR: a bioconductor
Trends Cell Biol. 25, 545–555 (2015). package for differential expression analysis of digital gene expression data.
42. Majumder, P. K. et al. mTOR inhibition reverses Akt-dependent prostate Bioinformatics 26, 139–140 (2010).
intraepithelial neoplasia through regulation of apoptotic and HIF-1- 70. Misharin, A. V., Morales-Nebreda, L., Mutlu, G. M., Budinger, G. R. &
dependent pathways. Nat. Med. 10, 594–601 (2004). Perlman, H. Flow cytometric analysis of macrophages and dendritic cell
43. Kyrmizi, I. et al. Calcium sequestration by fungal melanin inhibits calcium- subsets in the mouse lung. Am. J. Respir. Cell Mol. Biol. 49, 503–510 (2013).
calmodulin signalling to prevent LC3-associated phagocytosis. Nat. Microbiol.
3, 791–803 (2018).
44. Raffaello, A., Mammucari, C., Gherardi, G. & Rizzuto, R. Calcium at the Acknowledgements
center of cell signaling: interplay between endoplasmic reticulum, This work was supported by the Northern Portugal Regional Operational Programme
mitochondria, and lysosomes. Trends Biochem. Sci. 41, 1035–1049 (2016). (NORTE 2020), under the Portugal 2020 Partnership Agreement, through the European
45. Liou, J. et al. STIM is a Ca2+ sensor essential for Ca2+-store-depletion- Regional Development Fund (FEDER) (NORTE-01-0145-FEDER-000013), the Fundação
triggered Ca2+ influx. Curr. Biol. 15, 1235–1241 (2005). para a Ciência e Tecnologia (FCT) (SFRH/BD/136814/2018 to S.M.G., SFRH/BD/
46. Howard, N. C. et al. Mycobacterium tuberculosis carrying a rifampicin drug 141127/2018 to C.D.O., PD/BD/137680/2018 to D.A., IF/00474/2014 to N.S.O., IF/
resistance mutation reprograms macrophage metabolism through cell wall 01390/2014 to E.T., IF/00959/2014 to S.C., IF/00021/2014 to R.S., PTDC/SAU-SER/
lipid changes. Nat. Microbiol. 3, 1099–1108 (2018). 29635/2017 and CEECIND/04601/2017 to C.C., and CEECIND/03628/2017 to A.C.), the
47. Akoumianaki, T. et al. Aspergillus cell wall melanin blocks LC3-associated Institut Mérieux (Mérieux Research Grant 2017 to C.C.), and the European Society of
phagocytosis to promote pathogenicity. Cell Host Microbe. 19, 79–90 (2016). Clinical Microbiology and Infectious Diseases (ESCMID Research Grant 2017 to A.C.).
48. Thywissen, A. et al. Conidial dihydroxynaphthalene melanin of the human M.G.N. was supported by a Spinoza grant of the Netherlands Organization for Scientific
pathogenic fungus Aspergillus fumigatus interferes with the host endocytosis Research. A.A.B. was supported by the Deutsche Forschungsgemeinschaft Collaborative
pathway. Front. Microbiol. 2, 96 (2011) Research Center/Transregio TR124 FungiNet (project A1). G.D.B. was funded by the

Wellcome Trust (102705), the MRC Centre for Medical Mycology and the University of Peer review information Nature Communications thanks Darius Armstrong-James, Paul
Aberdeen (MR/N006364/1). Bowyer, Bibhuti Mishra, and the other, anonymous, reviewer(s) for their contribution to
the peer review of this work. Peer review reports are available.
Author contributions Reprints and permission information is available at http://www.nature.com/reprints

S.M.G. designed the study, performed experiments and data analysis, and wrote the
paper. C.D.O., C.F.C., and I.M. performed the animal experiments, V.A. prepared fungal Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in
components, D.A. and G.C. performed confocal microscopy, C.S.R. performed HPLC published maps and institutional affiliations.
measurements, C.B.M., J.G., and S.C. performed flow cytometry and cell-sorting
experiments, and G.C. performed the live recording of calcium levels. N.S.O. analyzed
RNA-seq data, and M.F.G. and C.B. performed targeted metabolomics. A.M. supervised
Open Access This article is licensed under a Creative Commons
the collection of samples from healthy donors. R.t.H., L.L., T.M., P.P., E.T., F.R., L.A.B.J.,
K.L., J.M., J.F.L., A.C.J., and R.S. analyzed the data and contributed to the paper. G.D.B., Attribution 4.0 International License, which permits use, sharing,
A.A.B., and G.C. contributed to the paper. F.L.v.d.V., M.G.N., J.P.L., C.C., and A.C. adaptation, distribution and reproduction in any medium or format, as long as you give
designed the study and wrote the paper. appropriate credit to the original author(s) and the source, provide a link to the Creative
Commons license, and indicate if changes were made. The images or other third party
material in this article are included in the article’s Creative Commons license, unless
Competing interests indicated otherwise in a credit line to the material. If material is not included in the
The authors declare no competing interests. article’s Creative Commons license and your intended use is not permitted by statutory
regulation or exceeds the permitted use, you will need to obtain permission directly from
Additional information the copyright holder. To view a copy of this license, visit http://creativecommons.org/
Supplementary information is available for this paper at https://doi.org/10.1038/s41467- licenses/by/4.0/.
020-16120-z.
© The Author(s) 2020
Correspondence and requests for materials should be addressed to A.C.

2020 Article 16120

Uploaded by

Copyright:

Available Formats

2020 Article 16120

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

2020 Article 16120

Uploaded by

Copyright:

Available Formats

ARTICLE

Phagosomal removal of fungal melanin reprograms

In response to infection, macrophages adapt their metabolism rapidly to enhance glycolysis

NATURE COMMUNICATIONS | (2020)11:2282 | https://doi.org/10.1038/s41467-020-16120-z | www.nature.com/naturecommunications 1

2 NATURE COMMUNICATIONS | (2020)11:2282 | https://doi.org/10.1038/s41467-020-16120-z | www.nature.com/naturecommunications

a 2 h post-infection 6 h post-infection b GO Biological Processes

Adj P value (–log10)

mRNA (relative expression)

HIF1A IL1B P = 0.0002 4

IL10 8 P < 0.0001 20

g Ctrl h Glucose-6P Fructose-6P DHAP Glyceraldehyde-3P 2/3PG PEP Pyruvate

0.020 0.06 0.06 0.15 0.08 0.004

0.10 0.015 0.06 0.003

0.000 0.00 0.0 0.00 0.00 0.00 0.000

NATURE COMMUNICATIONS | (2020)11:2282 | https://doi.org/10.1038/s41467-020-16120-z | www.nature.com/naturecommunications 3

4 NATURE COMMUNICATIONS | (2020)11:2282 | https://doi.org/10.1038/s41467-020-16120-z | www.nature.com/naturecommunications

Conidiacidal activity (%)

e Human Mø f g Mouse alveolar Mø

500 P = 0.0014 80 300 P = 0.0035

100 500 2000 P < 0.0001 100

NATURE COMMUNICATIONS | (2020)11:2282 | https://doi.org/10.1038/s41467-020-16120-z | www.nature.com/naturecommunications 5

a P = 0.0022 b P < 0.0001

mRNA (relative expression)

j GLUT1 HK2 PFKFB3

6 NATURE COMMUNICATIONS | (2020)11:2282 | https://doi.org/10.1038/s41467-020-16120-z | www.nature.com/naturecommunications

a b Δku80 ΔpksP Δku80 ΔpksP c 20

XBP1 P = 0.025 P = 0.021

Normalized pixel density

HIF-1α DAPI A. fumigatus

NATURE COMMUNICATIONS | (2020)11:2282 | https://doi.org/10.1038/s41467-020-16120-z | www.nature.com/naturecommunications 7

8 NATURE COMMUNICATIONS | (2020)11:2282 | https://doi.org/10.1038/s41467-020-16120-z | www.nature.com/naturecommunications

d P < 0.0001 e f g P = 0.0021

i j IL-1β TNF IL-6

P = 0.0021 P = 0.0011 200

NATURE COMMUNICATIONS | (2020)11:2282 | https://doi.org/10.1038/s41467-020-16120-z | www.nature.com/naturecommunications 9

10 NATURE COMMUNICATIONS | (2020)11:2282 | https://doi.org/10.1038/s41467-020-16120-z | www.nature.com/naturecommunications

NATURE COMMUNICATIONS | (2020)11:2282 | https://doi.org/10.1038/s41467-020-16120-z | www.nature.com/naturecommunications 11

12 NATURE COMMUNICATIONS | (2020)11:2282 | https://doi.org/10.1038/s41467-020-16120-z | www.nature.com/naturecommunications

NATURE COMMUNICATIONS | (2020)11:2282 | https://doi.org/10.1038/s41467-020-16120-z | www.nature.com/naturecommunications 13

14 NATURE COMMUNICATIONS | (2020)11:2282 | https://doi.org/10.1038/s41467-020-16120-z | www.nature.com/naturecommunications

Author contributions Reprints and permission information is available at http://www.nature.com/reprints

NATURE COMMUNICATIONS | (2020)11:2282 | https://doi.org/10.1038/s41467-020-16120-z | www.nature.com/naturecommunications 15

You might also like