C. albicans strains used in this study were SC5314 (WT, ATCC) and ATP2 null mutant (atp2Δ/Δ) (Li et al., 2018). Strains were maintained on YPD (1% yeast extract, 2% peptone, 2% dextrose) agar at 30°C and stored in 50% glycerol at -80°C. For each assays, C. albicans cells were cultured in YPD liquid medium overnight at 30°C and 150 rpm, then collected, washed and normalized to appropriate concentrations.

Murine macrophage-like cell line RAW264.7 (ATCC) was used for all macrophage assays. Cells were maintained in Dulbecco’s modified Eagle medium (DMEM) supplemented with 10% heat-inactivated fetal bovine serum, 100 U/ml penicillin and 100 μg/ml streptomycin, at 37°C with 5% CO₂.

The end-point dilution assay was performed to detect the killing effect of macrophages on C. albicans as described previously (She et al., 2013). RAW264.7 cells were seeded into 96-well plates at a density of 1×10⁵ cells/well and adhered for 1h. Then 2×10⁵ cells/well C. albicans were added and cocultured for 6, 12, 24 and 48 h. At each time point, the co-cultures were treated with 0.1% Triton X-100 for 2 min. A serial dilution was performed and plated onto YPD agar. Numbers of colony-forming units (CFU) were counted after 48 h incubation and the macrophage killing rate was determined by comparing cocultures to C. albicans control cultures without macrophages. Assays were performed in triplicate, and experiments were repeated three times.

C. albicans-induced macrophage damage was assessed by measuring the release of lactate dehydrogenase (LDH) using a Non-Radioactive Cytotoxicity assay (Abcam) (She et al., 2013). RAW264.7 (1×10⁵ cells) were co-incubated with C. albicans (1×10⁵ cells) for 2, 4, 6, and 8 h. At each time point, the absorbance at 490 nm was recorded according to the manufacturer’s protocol. The release of LDH relative to maximum LDH release was then calculated and corrected for spontaneous release of LDH by the C. albicans or macrophages alone. The experiment was performed in triplicate.

Activated RAW264.7 cells were seeded in 96-well plates at 1×10⁶ cells/ml (She et al., 2013). C. albicans cells were stained with 1.25 mM fluorescein isothiocyanate (FITC, Sigma) for 15 min, then diluted to 2 × 10⁶ cells/ml in RPMI medium and cocultured with macrophages for 30, 90, 120 min. At each time point, 100 μl supernatant was removed and 100 μl trypan blue was added to quench the fluorescence of unengulfed strains. The fluorescence signal intensity (E_x/E_m=495/525) of each well were detected with a microplate reader (Thermo Fisher Scientific). Assays were performed in triplicate, and experiments were repeated three times.

10-week-old female BALB/c mice (20 g) were injected intraperitoneally with 200 μl of clodronate- or PBS-liposome (http://clodronateliposomes.org) both 24 h before and 24 h after intravenous C. albicans infections (Wirnsberger et al., 2016). Mice were infected with 2×10⁵ CFU of C. albicans cells intravenously. At 24 and 72 h after infection, organs were taken out and weighed, then grinded and measured fungal burden (CFU/g) by the end point dilution assay. There were three mice in each group, and experiment was repeated twice.

10-week-old female BALB/c mice (20 g) were infected with 1× 10⁶ CFU of C. albicans cells intravenously. Mice were fasted for 12 h before the imaging time point and only water was provided. Each mouse was injected 1% pentobarbital (15 ul per g mouse) intraperitoneally and 5 μCi/g of [¹⁸F]FDG intravenously 45 min before imaging. PET and CT images were obtained by the Inveon micro-PET/CT (Siemens) small-animal PET imagers with a static acquisition time of 15 min (Davis et al., 2009). After PET/CT scan, the ex vivo biodistribution of organs was detected by γ-counter (Perkin-Elmer) (Rolle et al., 2016). There were three mice in each group, and experiments were repeated three times.

Total RNA from co-cultured macrophages were extract using Trizol (Invitrogen), and total RNA from C. albicans were extract using the E.Z.N.A. Yeast RNA kit (Omega Bio-Tek) (Li et al., 2017). Approximately 0.8 μg of RNA was used to synthesize cDNA (Qiagen, Venlo, Netherlands). RT-qPCR was done in triplicate as previously described using Bio-Rad iQ5. Primers used are shown in Table S1 . The expression levels of C. albicans genes were normalized to 18S rRNA levels, while the macrophage genes were normalized to GAPDH levels. The 2^–ΔΔCT (where CT is the threshold cycle) method was used to determine the fold change in gene transcription. Each sample was performed in triplicate and experiments were repeated three times.

C. albicans cells were collected and washed with PBS, then inoculated in 100 ml of macrophage-mimicking media (YNB liquid medium supplemented with 2% glucose, CAA, GLcNAc, oleic acid, or lactate) with an initial OD₆₀₀ of 0.02. Shake cultures were grown at 30 °C and OD₆₀₀ of each strain was measured every 2 h (Li et al., 2018). Cell viability was detected by the end-point dilution assay as describe previously. C. albicans cells were washed and resuspended in media mentioned above to a density of 1×10⁶ cells/ml. Cultures were grown at 30 °C and 100 μl of liquid was taken out to perform end-point dilution assay every 2 h. All experiments were performed in triplicate and repeated three times.

RAW264.7 cells (1×10⁶ cells) were seeded to glass coverslips and incubated overnight (Vylkova and Lorenz, 2014). 500 nM MitoTracker Deep Red FM (Molecular Probes) was added and incubated for 30 min. C. albicans cells were stained with FITC as mentioned before, and cocultured with macrophages for 3 h. Images were taken using a confocal laser scanning microscope (Carl Zeiss LSM880), under the set of E_x644/E_m655 (macrophages) and E_x488/E_m525 (C. albicans). The percentage of hyphae cells were counted manually (the number of germinated cells versus the number of total cells). At least 50 cells per strain were counted. Experiments were repeated three times.

The hyphae formation assay was performed as described previously (Li et al., 2017). C. albicans cells were washed and resuspended in YNB liquid medium supplemented with 2% glucose, CAA, GLcNAc, oleic acid, or lactate to a density of 1×10⁵ cells/ml. Cells were incubated in 24 well plates at 37°C for 3 h and images were taken by an inverted microscope (Olympus IX81).

The end-point dilution assay was performed to detect the killing effect of H₂O₂ on C. albicans (Wu et al., 2018). C. albicans cells were washed and resuspended in YNB liquid medium supplemented with 2% glucose, CAA, GLcNAc, oleic acid, and lactate to a concentration of 5×10⁴ cells/ml. Then, 6 mM H₂O₂ was added except for the growth control well and incubated at 37°C for 6 h. A serial dilution was performed and plated onto YPD agar. CFU were counted after 48 h incubation and the mortality rate was determined by comparison with fungi recovered without exposure to H₂O₂.

C. albicans protein was extracted by a liquid nitrogen grinding method as described previously (Kitahara et al., 2015). C. albicans cells were diluted into 1 L cultures at a starting OD = 0.5, cultured for 8 h at 30°C, and harvested at 4 °C. The samples were ground into cell powder, then lysis buffer (8 M urea, 1% Triton-100, 10 mM dithiothreitol, and 1% protease inhibitor) was added, followed by sonication three times on ice using a high-intensity ultrasonic processor (Scientz).

The proteins were reduced with 5 mM dithiothreitol for 30 min at 56°C and alkylated with 11 mM iodoacetamide for 15 min. The sample was diluted with urea, and trypsin was added at the mass ratio of trypsin: protein = 1:50 for the first overnight digestion and 1:100 for a second 4 h digestion. The tryptic peptides were fractionated by high pH reverse-phase HPLC using a Betasil™ C18 column (5 μm particles, 250 mm length, Thermo Scientific). The protein digests were then labeled with a Tandem Mass Tag kit (Thermo Scientific) (Kitahara et al., 2015). The peptides were desalted using a Strata X C18 SPE column (Phenomenex) and vacuum-dried. Then the peptides were separated into 60 fractions using a gradient of 8%–32% acetonitrile (pH 9.0) over 60 min. Finally, the peptides were combined into 10 fractions and dried by vacuum centrifugation.

The detailed protocol for LC‐MS/MS was described previously (Herrero-de-Dios et al., 2018; Truong et al., 2019). The tryptic peptides were dissolved in 0.1% formic acid (solvent A) and loaded on an analytical column (75 µm × 150 mm). The constant flow rate was set at 500 nL/min with the step gradients of mobile B (0.1% formic acid in 90% acetonitrile): 9%–26% for 38 min, 26%–35% for 14 min, 35%–80% for 4 min, and held at 80% for 4 min. The peptides were subjected to NSI source followed by tandem MS analysis in Q ExactiveTM Plus (Thermo) coupled online to the UPLC. The mass range was 400–1500 m/z, and tandem mass spectra were recorded in high sensitivity mode. In each cycle, a maximum of 20 precursors were selected for fragmentation, with the dynamic exclusion for 15 s. Protein identification and quantification were performed via the Maxquant search engine (v.1.5.2.8). The data were searched against a protein sequence database downloaded from UniProtKB for C. albicans SC5314 (total 6040 entries). The quantitative method was set to TMT-10plex, and the FDR for protein identification and PSM identification was adjusted to < 1% (Truong et al., 2019). The mass spectrometry proteomics data have been deposited to the ProteomeXchange Consortium via the PRIDE (Perez-Riverol et al., 2019) partner repository with the dataset identifier PXD024039.

All BABL/c mice were obtained from Guangdong Medical Laboratory Animal Center, Foshan, Guangdong, China. The animal experiments were performed in accordance with the recommendations in the Guide for the Care and Use of Laboratory Animals of the National Academy of Science. The protocol was carried out with permission from the Laboratory Animal Ethics Committee of Jinan University (NO. 2019670).

Escape from macrophage clearance is the basis by which C. albicans establishes systemic infection (Lionakis et al., 2013; Ngo et al., 2014; Weiss and Schaible, 2015). To investigate the impact of ATP2 on this pivotal procedure, we examined the proportion of C. albicans (WT and atp2Δ/Δ) killed by macrophages (RAW264.7) in a coculture system via the end-point dilution assay. When C. albicans and macrophages were cocultured at a ratio of 2:1 (multiplicity of infection [MOI] of 2), the atp2Δ/Δ and WT cells were phagocytosed at comparable rates (P > 0.05) ( Figure 1A ), but more atp2Δ/Δ cells were killed by macrophages and less macrophages were damaged at each time point than WT cells (P < 0.05) ( Figures 1B, C ). The mortality rate of atp2Δ/Δ was 98.2% at 24 h, but only 32.5% in the WT at this time point (P < 0.05) ( Figure 1B ). Then we compared the CFUs of WT and atp2Δ/Δ in medium without macrophages and found that atp2Δ/Δ grew only 31.1% less than WT at 24 h ( Figure S1 ). This suggests that the deletion of ATP2 in C. albicans resulted in a significant reduction in the ability to escape macrophage clearance.

To further study the effect of ATP2 on the escape of C. albicans from macrophage clearance in vivo, we constructed macrophage normal/depleted mice with PBS/clodronate liposome, and detected the fungal burden at different time points after systemic infection with WT or atp2Δ/Δ. The results showed that the WT cells were not cleared in the normal mice (PBS liposome), and that the fungal burden in the kidney at 72 h was higher than that at 24 h post-infection (P < 0.05); however, the fungal burden in the liver, spleen and brain did not significantly differ from that at 24 h (P > 0.05). In contrast, the atp2Δ/Δ mutant cells were almost completely cleared 72 h after infection, with no detectable pathogens in the liver, kidney and spleen. Cells remained in the brain were far less than WT ( Figures 2A–D ) (P < 0.05).

However, in the macrophage-depleted mice (clodronate liposome), the atp2Δ/Δ cells stably persisted in the liver, kidney, spleen, and brain at 24 h and 72 h after infection instead of being cleared. The number of WT cells was increased in the spleen ( Figure 2H ) and brain ( Figure 2G ) but unchanged in the liver ( Figure 2E ) and kidney ( Figure 2F ) 72 h post-infection compared with those at 24 h. These results suggest that C. albicans cells lacking ATP2 were unable to resist macrophage clearance in vivo.

The clearance of pathogens typically requires macrophages to recruit large numbers of phagocytes, including neutrophils and NK cells, to enhance the killing efficiency (Ngo et al., 2014; Netea et al., 2015). Is the efficient clearance of atp2Δ/Δ associated with increased recruitment of phagocytes? To answer this question, we injected mice with [¹⁸F]FDG, which accumulates in activated macrophages and neutrophils, and detected the [18F]FDG signal distribution by microPET/CT and γ counter in real time, dynamically and overall (Davis et al., 2009; Rolle et al., 2016). Surprisingly, the atp2Δ/Δ-infected mice never showed an increase in the [18F]FDG signal intensity throughout the body, and there was no significant difference between the atp2Δ/Δ and NS groups ( Figures 3A–D ) (P > 0.05). However, the [18F]FDG intensity in the WT-infected mice was significantly increased, especially in the target organ, i.e., kidney, where the signal intensity reached 7-fold of that observed in the atp2Δ/Δ and NS groups at 72 h ( Figures 3E ) (P < 0.05). The brain, heart, liver, and colon signals in the WT group were also stronger than those in the atp2Δ/Δ and NS groups ( Figures 3C, F ) (P < 0.05).

In alignment with the PET results, we did not observe significant inflammatory foci in the kidney sections from the atp2Δ/Δ-infected mice, whereas multiple foci of inflammatory cells containing mostly polymorphonuclear leukocytes (PMNs) were observed in the WT-infected mice ( Figure 4 ). The inflammatory scores obtained from renal immune cell infiltration and tissue destruction were significantly lower than those in the WT group ( Figure 4 ) (P < 0.05). These results suggest that the clearance of the atp2Δ/Δ mutant cells did not induce abnormal inflammatory cells recruitment in situ.

Once engulfed, C. albicans cells rapidly adapt the glucose-deficient environment, shifting to gluconeogenic growth and escaping from macrophages within 6-8 h (Lorenz et al., 2004; Wartenberg et al., 2014). To observe the effect of ATP2 on the adaptation of C. albicans to the environment, we first examined the growth and viability of WT and atp2Δ/Δ in the media that simulated glucose-deficient environment within macrophages. The results showed that the OD₆₀₀ of WT group in macrophage-mimicking media (YNB medium with 2% CAA, GlcNAc, lactate, or oleic acid) increased, while the OD₆₀₀ of atp2Δ/Δ group remained at the initial level ( Figures 5A–D ). More importantly, consistent with the growth results detected by OD₆₀₀, the viable cells of WT were increased in these macrophage-mimicking media over time, reaching 3-5 times the initial value at 8 h. However, the number of viable atp2Δ/Δ in these media decreased by 20-40% at 8 h ( Figures 5E–H ). These results suggest that the atp2Δ/Δ cells stop proliferating, but can partially remain viable for a short period of time in the glucose-deficient environment.

C. albicans can also adapt to or even alkalize the acidic environment within macrophages, creating more favorable conditions for its survival and escape (Vylkova and Lorenz, 2014). We examined the effect of WT and atp2Δ/Δ on the pH value of the environment in macrophage-mimicking media. The results showed that the pH values of the CAA ( Figure 5I ), GlcNAc ( Figure 5J ) and lactate ( Figure 5L ) medium increased simultaneously with the OD₆₀₀ when the WT cells were cultured. The pH of the oleic acid medium gradually decreased when the oleic acid decomposed by WT cells ( Figure 5K ). In contrast, the pH values of the media with atp2Δ/Δ cells were not statistically different at each time point, the atp2Δ/Δ cells were unable to change the environment pH in all macrophage-mimicking media ( Figures 5I–K ). The above results suggest that deletion of ATP2 impaired the adaptation of C. albicans to the glucose-deficient environment within macrophages.

Morphogenesis is an important contributing factor for the escape from macrophages (Peroumal et al., 2019; Rogiers et al., 2019). To investigate the hyphae formation of these viable atp2Δ/Δ cells, we observed the status of FITC-stained C. albicans (green) cocultured with MitoTraker-preloaded macrophages (red) when cocultured for 3 h. The results showed that WT cells completed morphogenesis with a germination rate of 45.4%, whereas the atp2Δ/Δ mutant cells were trapped inside macrophages in the yeast phase ( Figures 6A, B ). Then, we extracted RNA from phagocytic C. albicans to conduct RT-qPCR and found that the expression levels of the genes involved in hyphal formation (ECE1, HGC1, HWP1 and ALS3) were significantly lower than those in the WT group (P < 0.05) ( Figure 6C ).

Interestingly, we found that the unengulfed atp2Δ/Δ mutant could undergo morphogenesis in the medium with a germination rate of 18.1% ( Figure 6D ). Under this condition, only HGC1 was expressed at a significantly lower level than that in the WT, while the expression levels of ECE1, HWP1 and ALS3 did not significantly differ ( Figure 6E ). The above results suggest that deletion of ATP2 prevents C. albicans from morphogenesis inside macrophages, but not outside of them.

Why atp2Δ/Δ cells were unable to form hyphae in macrophages? We performed hyphae formation assay in classic hyphae induction media (YPD+10% FBS, Spider, Lee’s), DMEM, and macrophage-mimicking media. The results showed that more than 90% of the WT cells formed elongated hyphae in all the media mentioned above, while the atp2Δ/Δ cells only formed short hyphae in media containing glucose (DMEM, YPD+10% FBS, and Lee’s) with germination rates around 10% ( Figures 7A–D ), and could not reach a level close to that of WT even with extended incubation time ( Figure S2 ). In Spider medium (no glucose inside) and macrophage-mimicking media, the atp2Δ/Δ cells didn’t form hyphae ( Figures 7A–D ). Then we supplemented macrophage-mimicking media with 0.2% glucose, which was close to glucose level in blood, and found that hyphae formation of atp2Δ/Δ was partially restored ( Figures 7E, F ). The above results suggest that the hyphae formation of atp2Δ/Δ is reduced, but whether it forms hyphae also related to the presence or absence of glucose.

Oxidative stress within the phagosome creates a toxic environment that induces oxidative stress and programmed cell death in C. albicans (Dantas Ada et al., 2015). We further investigated the resistance of atp2Δ/Δ to oxidative stress (6 mM H₂O₂) in different environments. The results showed that 39.3% and 64.7% of the WT and atp2Δ/Δ cells were killed in the glucose medium at 6 h, respectively (P < 0.05) ( Figure 8A ). However, in the macrophage-mimicking media mentioned above, almost all atp2Δ/Δ cells were killed under the same H₂O₂ concentration, while the WT cells exhibited only slightly increased sensitivity to H₂O₂ in oleic acid ( Figure 8A ).

We further tested the transcription levels of the CAT1 gene (which protects C. albicans from oxidative stress) in the atp2Δ/Δ and WT cells under different conditions (Dantas Ada et al., 2015). Consistent with the phenotype, the expression level of CAT1 in the atp2Δ/Δ cells was decreased by 31.6% in the glucose medium compared with that in the WT ( Figure 8B ). However, the CAT1 levels in the atp2Δ/Δ cells were greatly downregulated in the alternative carbon source but were upregulated (CAA and lactate), unchanged (GlcNAc), or slightly downregulated (oleic acid) in the WT cells (P < 0.05) ( Figure 8B ). It can be seen that atp2Δ/Δ cells are more susceptible to oxidative stress, especially in the glucose-deficient environments.

To further investigate the effect of ATP2 on multiple abilities associated with C. albicans escape from macrophages, we performed a proteomics study and found that deletion of ATP2 resulted in 112 proteins up-regulated and 268 proteins down-regulated (P < 0.05 and fold change > 1.5). Here, we focused on proteins involved in hyphae formation, stress responses, alternative carbon source utilization, etc.

When referred to the WT strain, several proteins required for hyphae formation were down-regulated, for example Ihd1p, Opt3p, and Ole1p were 0.21-, 0.54-, and 0.35-fold down-regulated in the atp2Δ/Δ mutant ( Figure 9 ). In addition, putative glutathione peroxidase (Gpx2p) involved in Cap1p-dependent oxidative stress response was 0.28-fold down-regulated in atp2Δ/Δ mutant ( Figure 9 ). However, Sod1p, which play a protective role against oxidative stress, was upregulated 3.3-fold in atp2Δ/Δ mutant ( Figure 9 ). Proteins related to cell wall remodeling, GPI-anchored cell wall proteins (Exg2p, Rbt5p, and Crh11p) were up-regulated in atp2Δ/Δ mutant. Proteins associated with DNA damage repair were generally down-regulated.

As for proteins involved in alternative carbon source utilization, all enzymes (Icl1p, Mls1p, Aco1p, Cti1p, and Mdh1p) involved in the GC pathway were significantly repressed in the atp2Δ/Δ cells, two of the eight (Pck1p and Fbp1p) gluconeogenesis enzymes were upregulated, and there was no significant change in the β-oxidation pathway ( Figure 9 ). The proteomic results were generally consistent with the phenotype, proteins related to multiple functions related to C. albicans escape from macrophages were repressed to some extent.