The fungal strains used in this study (Table 1) were stored at -80°C in tubes containing a mixture of 75% YPD and 25% glycerol. Unless otherwise specified, strains were retrieved from the storage every seven days and cultured on solid YPD plates (1% yeast extract, 2% peptone, 2% dextrose, 1.5% agar, pH 5.6) at 30°C for 48 to 72 hours. Prior to each experiment, colonies from each strain were used to inoculate liquid YPD and incubate overnight at 30°C under constant agitation. For flow cytometry, lipid extraction and in vivo assays, the cultured strains were washed three times in sterile saline and suspensions of 10⁷ CFU/mL were inoculated into liquid minimal medium (MM) composed of Dextrose 15 mM, MgSO_4–10 mM, Glycine 13 mM, KH₂PO₄ 29.4 mM, and Thiamine 3 µM (pH 5.5). Cultures were incubated at 30°C under agitation for 15 hours before use.

The annotated sequence of the choline-phosphate cytidylyltransferase (PCT1) gene from the C. neoformans H99 genome was obtained from the FungiDB database (http://fungidb.org). Based on the nucleotide sequence CNAG_00424 (CnPCT1), primers for polymerase chain reaction (PCR) were designed using the Primer3 tool (https://fungidb.org/fungidb/app) (Supplementary Table 1), for gene deletion in the parental strain and in CnOPI3 deleted mutant (obtained from Madhani’s knock out collection). Gene deletion was carried out using double joint-PCR (DJ-PCR) method for selection cassette construction, as described by Kim et al (Kim et al., 2009). (Supplementary Figure 1), followed by biolistic transformation for genomic integration. The selection cassette was generated using plasmid pPZP-HYG2, which carries HPH, (conferring hygromycin resistance) under control of C. neoformans actin promoter and TRPC terminator (Walton et al., 2005).

Mutant strains reconstitutions were performed via co-transformation using PCR-amplified PCT1 or OPI3 gene fragments in combination with pJAF1 plasmid. This plasmid contains the neomycin resistance marker, Neo^R, driven by C. neoformans actin promoter and terminator (Fraser et al., 2003). Confirmation of gene deletions or reconstitutions were performed by selection marker screening and diagnostic PCRs (Supplementary Figures 1, 2).

The growth rate of each mutant was assessed in liquid YPD medium (Yeast 1%, peptone 2%, dextrose 2%, pH 5.6) and in liquid MM, either supplemented or not with 10µM choline chloride (Sigma), 10µM L-α-glycerophosphorylcholine (Sigma), 500 µM phosphatidylcholine (Sigma) or 10% fetal bovine serum. Wild-type and mutant cells were inoculated at a final cell density of 5 × 10⁴ cells/mL into the respective media. Then, 200 μL of each inoculum was dispensed into wells of a 96-well microplate. Each experiment was realized in technical duplicates or triplicates and two biological replicates. The plates were incubated at 30°C or 37°C in a microplate spectrophotometer (Epoch 1, Gen5 software v.3.11, Biotek), under agitation at 190 rpm for 96 to 120 hours. Optical density at 600 nm (OD 600 nm) was recorded every 30 minutes.

Susceptibility to various stressors was assessed by preparing a starting inoculum of 2 × 10⁸ CFU/mL, which was then serially diluted to a final concentration of 2 × 10² CFU/mL. A volume of 5 μL from each dilution was spotted onto YPD or MM agar plates supplemented with different stress-inducing agents: Congo Red (0.5%), Calcofluor White (1.5 mg/mL), sodium chloride (1 M), sodium nitrite (4 mg/mL hydrogen peroxide (2 mM), sodium dodecyl sulfate (0.01%), and/or sorbitol (2 M). Plates were incubated at 30°C and photographed after colony formation.

The melanization phenotype was evaluated following a protocol adapted from Eisenman (Eisenman et al., 2011). A total of 1 × 10⁷ cells from each strain were inoculated onto MM agar plates supplemented with 1 mM L-DOPA (Sigma-Aldrich), with or without 1 mM choline chloride or 10 µM GPC. The plates were incubated at 30°C and 37°C for one week protected from light. Photographs were taken at defined time intervals: 24 h, 48 h, 72 h, and 96 h. To quantify melanization, mean gray values of the colonies were analyzed using ImageJ v.1.53r software (https://imagej.net/ij/). Statistical analysis was performed using two-way ANOVA with Tukey’s correction for multiple comparisons. Data represents the average of two biological replicates.

Polysaccharide capsule size and titan cell formation was evaluated as described by Hommel et al (Hommel et al., 2018). Briefly, 1 × 10⁶ cells were inoculated into liquid MM and incubated at 30°C for 20 hours under agitation. Then, 1 mL of the culture was centrifuged and resuspended in India ink for optical microscopy analysis. Slides were prepared with 5 μL of the cell suspension mixed with India ink and imaged using a Nikon ECLIPSE Si microscope with a RoHS, BC1200 camera under a 40 × objective. The capsular diameter of 100 cells per strain was measured by subtracting the cell body diameter from the total diameter. The diameter of 200 cells per strain was also measured and titan cells were defined as those with ≥10 μm. All measurements were conducted using the image analysis and processing software ImageJ v.1.53r. Statistical analysis was performed using Kruskal-Wallis test with Tukey’s corrections for multiple comparisons. Three biological replicates were performed.

Cell viability in nutrient-poor medium was evaluated by incubating previously grown strains in liquid MM at a concentration of 1 × 10⁷ CFU/mL for 24 hours at 30°C under agitation. At hourly intervals after inoculation, samples were collected from each strain, serially diluted, and spotted onto YPD agar plates for CFU enumeration. Regression analysis was performed to determine statistically significant differences in viability among strains. Three biological replicates were performed with two technical replicates.

Lipid extraction and total inorganic phosphate (Pi) measurement was carried out using Mandala buffer as previously described and adapted to this study (Singh et al., 2011). Strains were cultured overnight in liquid YPD and then transferred to liquid MM for 15 hours. Strains were then washed with PBS and 5 × 10⁸ yeasts per sample were centrifuged and frozen in liquid nitrogen. Pellets were resuspended in 2 mL Mandala buffer (dH₂O:ethanol:diethyl ether:pyridine:NH₄OH (15:15:5:1:0.018; v/v)) and 50 mg of glass beads were added. Samples were vortexed for 30 seconds, sonicated three times for 30 seconds, vortexed again, and incubated in a 60°C water bath for 15 minutes. Following homogenization and centrifugation at 1,300 × g for 10 minutes at 4°C, the supernatant was transferred to a new tube and dried using SpeedVac. The pellet was resuspended in methanol and incubated at 37°C for 60 minutes, then centrifuged at 1,300 × g for 10 minutes at room temperature. The resulting supernatant was transferred to a new tube and 1 mL of chloroform and dH₂O were added. They were then homogenized and centrifuged for 5 minutes to collect the lower organic phase containing the phospholipids.

For total phosphate (Pi) quantification, 120 μL of each lipid extract was digested with perchloric acid and incubated at 150°C for 2 hours. After cooling to room temperature, 900 μL of dH₂O, 500 μL of 0.9% ammonium molybdate and 200 μL of 9% ascorbic acid were sequentially added with vortexing. Samples were incubated in water bath at 80°C for 30 minutes and absorbance was measured at 797 nm. A standard curve was constructed using NaH₂PO₄ (0-50 μM), and sample Pi concentrations were determined accordingly. Three biological replicates with nine technical replicates were analyzed.

Phosphatidylcholine quantification was performed using a commercial assay kit (MAK049, Sigma) in two biological replicates with three technical replicates. A qualitative evaluation of phospholipid profile was conducted using thin-layer chromatography (TLC). Silica gel plates pre-washed with chloroform:methanol:acetic acid:dH₂O (60:50:1:4) eluent solution (Cartwright, 1993). After drying, 20 μL of wild-type sample was spotted onto the plate and mutant strains sample volume were normalized by total Pi quantification. After incubation in eluent solution, plates were developed using iodine vapor and plates were then photographed. Standards of PC (15:0-19:1-d7-PC, Avanti^®), and PE (16:0-18:1-PE, kindly donated by Dr. Gabriel Silva Vignoli Muniz), were prepared at 10 mg/mL and 1 mg/mL, respectively, in chloroform:methanol (2:1). A quantitative analysis was conducted utilizing ImageJ v.1.53r to evaluate median pixel intensity (MPI) of the PC spot areas of each strain and results were normalized by wild-type PC spot area MPI. Six independent replicates were performed. Data were analyzed using one-way ANOVA with Dunnett’s correction, using the wild-type KN99α strain as the control.

To evaluate lipid droplet formation, fungal strains were cultured overnight in YPD and then transferred to liquid MM for 15 hours at 30° C under agitation. After incubation, cells were washed with PBS and adjusted to a final concentration of 1 ×10⁷ CFU/mL. Lipid droplets were stained by incubating each sample with lipophilic dye Nile Red (Sigma) (Rostron and Lawrence, 2017) at a final concentration of 0.2 μM for 10 minutes at 30°C in the dark. Chitin content was assessed by staining with Calcofluor White (Fluorescent Brighetner 18, Sigma) at a final concentration of 100 μg/mL for 10 minutes. Mitochondrial activity was evaluated using MitoTracker™ Orange CMTMRos (ThermoFisher) at a final concentration of 0.2 μM under the same conditions. Stained and fixed cells were analyzed by both flow cytometry and fluorescence microscopy. For microscopy, samples were mounted on glass slides and imaged using a LEICA TCS SP5 confocal microscope equipped with a CorrL 405 63x/1.4-0.60 oil immersion objective. Fluorescence was detected across four excitation/emission channels (385–720 nm). For flow cytometry, stained cells were analyzed using BD VERSE flow cytometer equipped with PE-Texas Red (610/20 nm), Pacific-Blue (450/50 nm) and PE-A (586/15 nm) filters. Forward and side scatter parameters were used to gate single-cell populations and fluorescence intensity was recorded for at least 20,000 events per sample. Data were analyzed using FlowJo software (Treestar). Two biological replicates were carried out with two technical replicates. Statistical analyses were conducted using one-way ANOVA with Tukey’s correction for multiple comparisons.

C. neoformans mutant strains (KN99α, pct1Δ, opi3Δ, and opi3Δpct1Δ) were cultured for extracellular vesicle (EV) isolation as described by Marina et al. (2020). Yeast cells in the exponential growth phase were first grown in YPD medium under agitation (150 rpm) for 48 hours at 30°C. Subsequently, cells were inoculated onto Sabouraud Dextrose Broth (SDB) plates at pH 5.5 to a final density of 4 × 10⁷ cells/mL and incubated for 24 hours at 30°C. After growth on SDB plates, cells were collected from the agar surface and resuspended in phosphate-buffered saline (PBS). The suspension was centrifuged at 5,000 × g for 15 minutes, and the supernatant was filtered through a 0.45 μm membrane and ultracentrifuged at 100,000 × g for 1 hour. The resulting pellet containing EVs was resuspended in 500 μL of PBS.

EV content was quantified indirectly by measuring ergosterol and protein concentrations using the Amplex Red Cholesterol Assay Kit (Thermo Fisher Scientific) for ergosterol and the Micro BCA Protein Assay Kit (Thermo Fisher Scientific) for protein, according to the manufacturer’s instructions. Vesicle size distribution and surface characteristics were analyzed by dynamic light scattering (DLS) using a Zetasizer instrument (Malvern Panalytical) to determine the hydrodynamic diameter of EVs. Data were analyzed using one-way ANOVA with Dunnett’s correction, using the wild-type KN99α strain as the control.

Phagocytosis assays were performed following a protocol adapted from Nicola and Casadevall (Nicola and Casadevall, 2012). Murine bone marrow-derived cells were cultivated in RPMI-1640 (Gibco) medium supplemented with 10% fetal bovine serum (FBS, Gibco), 20 ng/mL granulocyte-macrophage colony-stimulating factor (GM-CSF, ImmunoTools), and 50 μM β-mercaptoethanol (Sigma) for eight days. On the last day of differentiation, cells were recovered and the macrophages were counted and adjusted to 5 × 10⁴ cells/well, which were resuspended in RPMI-1640 + 10% FBS and transferred to a 96-well plate and incubated for 24 hours at 37°C in a 5% CO₂ atmosphere.

Before macrophage infection, the wild-type and mutant strains were grown in liquid YPD for 24 hours. Cells were washed three times with PBS, counted, and resuspended in RPMI-1640. Fungal cells were opsonized in RPMI with 10 μg/mL of the anti-GXM monoclonal antibody 18B7 and inoculated into macrophage-containing wells at a multiplicity of infection (MOI) of 5. Plates were incubated for 2 hours and 24 hours at 37°C in a 5% CO₂ atmosphere. After 2 hours of co-culture, wells were washed with PBS. In one plate, macrophages were lysed with 100 μL of distilled water. Lysates were diluted, plated on YPD agar, and incubated at 30°C for 48 hours to count colony-forming units (CFUs). In the second plate, 100 μL of RPMI 1640 + 10% FBS (with 500 IU of IFN-γ and 500 ng/mL LPS for the stimulated group) was added and incubated for an additional 24 hours. After incubation, macrophages were lysed, and CFUs were determined. The fungal viability was calculated using the formula: Viability (%) = {CFU (24h)/mean CFU (2h)} × 100. Data were analyzed using two-way ANOVA with Dunnett’s correction, using the wild-type KN99α strain as the control. All experiments included three technical replicates and were performed in three independent biological replicates.

To investigate mitochondrial activity and cell reactive oxygen species (ROS) production, a fluorescence spectrophotometric analysis using Mitotracker™ Orange CMTMRos and 2’,7’-dichlorofluorescein diacetate, DCFDA (Sigma-Aldrich) stains, was carried out. The mutants and wild-type strains were grown in YPD and transferred to minimal medium for 15 hours as previously described. After that, yeast cells were washed with PBS, counted, and cell concentration was adjusted to 1 × 10⁶ yeast/mL in 500 μL for staining with either Mitotracker orange (0,2 μM) or DCFDA (10 μM) and incubated in the dark at 37°C for 30 minutes. Stained samples were then washed twice with PBS and 100 μL of each replicate was transferred to a 96 well plate for fluorimetric analysis in a Varioskan LUX (Thermo Fisher Scientific) plate reader. Mitotracker orange fluorescence was detected using 554/576 nm excitation/emission settings, while DCFDA was detected using 495/525 nm. Microscopic analysis of Mitotracker labeling was done using an EVOS M7000 Cell Imaging Systems (Invitrogen) microscope, coupled with Red Fluorescent Protein range emission (580 nm to 650 nm) LED light source. Microphotographs were taken with an exposure of 60 ms and 10.0 db gain. Each experiment was realized in technical triplicates and two biological replicates. One-Way ANOVA test with Dunnett correction was used for statistical analysis.

An in vivo virulence assay was conducted using G. mellonella larvae, following the protocol described by Mylonakis et al (Mylonakis et al., 2005). Larvae in their final development stage were weighed, and only those exceeding 200 mg were selected for the experiment. On the day of infection, C. neoformans cells previously grown for 15 hours in MM were resuspended in PBS, and the inoculum was adjusted to a final concentration of 5 × 10⁶ cells/mL. Each larva was injected with 10 μL of cell suspension into the last left proleg with a sterile Hamilton syringe. Following the injection, larvae were incubated at 37°C. For each fungal strain, 12 larvae were used. The negative control group was inoculated with 10 μL of PBS under the same conditions. Larval survival was monitored daily for 10 days. Data analysis was performed using the Mantel-Cox test.

The mice (Mus musculus), aged between 8 and 12 weeks and from the C57BL/6 strain, were housed under standard conditions at the animal facility of the Bi−Institutional Translational Medicine Platform/Fiocruz, with water and food provided ad libitum. All experimental procedures were approved by Ethics Committee of the University of São Paulo (Process No. 1305/2024) and conducted in accordance with institutional and national guidelines for animal care and use. Mice were infected via intratracheal route with 1 ×10⁴C. neoformans yeast cells previously cultured for 15 hours in liquid MM. For the survival curve analysis (n=5 per group), animals were monitored daily for clinical signs of morbidity. Mice exhibiting severe symptoms were euthanized according to the ethical recommendation. Briefly, mice were anesthetized with ketamine hydrochloride and xylazine hydrochloride (80–100 and 5–10 mg/kg, respectively), administered intramuscularly, followed by. euthanasia via intravenous administration of an overdose of the same anesthetic agents (ketamine hydrochloride, 240–300 mg/kg, and xylazine hydrochloride, 15–30 mg/kg). At 28 days post-infection, surviving animals were euthanized and visceral organs were exposed using sterile surgical instruments. The pulmonary lobes and brain tissue were collected to assess fungal burden. Samples were homogenized in PBS, and serial diluted, and plated on YPD plates for CFU counting. Data were analyzed using the Mantel-Cox test for survival growth curve and one-way ANOVA for CFU burden. Three independent experiments were realized.

All statistical analyses were performed using GraphPad Prism 8.3 (GraphPad Inc., San Diego, CA, USA). A p value < 0.05 was considered statistically significant.

Deletion of PCT1 appears to impair C. neoformans’ ability to assimilate choline from the medium. This was evident from the impaired growth of the double mutant opi3Δpct1Δ in liquid MM supplemented with choline chloride (Figure 2). However, this mutant exhibited normal growth in YPD, similar to the wild-type, suggesting that it could acquire an alternative metabolite required for PC biosynthesis from rich medium. One likely candidate is the GPC pathway, a third route for PC production previously described in the literature (Głąb et al., 2016; King et al., 2024), potentially compensating for loss of PCT1. Supporting this, supplementation of liquid MM with only 10 μM GPC fully rescue the growth defect of opi3Δpct1Δ to the wild-type levels (Figure 2, Supplementary Figure 3). Interestingly, only high concentrations (500 μM) of PC (L-α-phosphatidylcholine type XVI-E) were sufficient to restore growth of the double mutant (Supplementary Figure 3). Furthermore, temperature did not affect growth of OPI3 and or PCT1 single mutants; however, GPC supplementation completely rescued growth defect of opi3Δpct1Δ observed in MM at both 30 and 37°C (Figure 2). These results indicate that while the Kennedy pathway is critical for choline assimilation in PC synthesis, the GPC reacylation pathway can compensate and plays a significant role in maintaining phospholipid synthesis in C. neoformans.

Lee et al. (2024) reported that sorbitol and polyethylene glycol (PEG) function as chemical chaperones, alleviating membrane-associated protein misfolding in opi3Δ mutant and rescuing growth under nutrient-limited condition (Lee et al., 2024). To determine whether a similar effect occurs in opi3Δpct1Δ mutant, growth was assessed in liquid MM supplemented with sorbitol. Sorbitol supplementation fully restored the growth defect of the double mutant (Supplementary Figure 3). To further investigate whether deletions of OPI3 and PCT1 compromises cell wall and plasma membrane integrity, strains were spotted onto MM agar plates containing 1M sorbitol and supplemented with stress-inducing agents. These included sodium dodecyl sulfate (SDS), a membrane-destabilizing detergent, and the cell wall-perturbing agents Calcofluor White and Congo Red, which bind to chitin polymers (Ram and Klis, 2006). The opi3Δpct1Δ double mutant showed pronounced growth defect on SDS-supplemented medium at 30°C (Figure 3) indicating alterations in membrane homeostasis. In contrast, the mutant exhibited wild-type growth on plates containing Calcofluor White and Congo Red despite reduction in cell wall chitin content after 15 hours under nutrient limited conditions, possibly indicating an active cell wall remodeling in response to phosphatidylcholine deprivation (Supplementary Figure 4). Single mutant strains did not exhibit any growth defects under these conditions.

To assess osmoregulatory capacity, all strains were grown on YPD and MM agar plates supplemented with either 1 M NaCl or 2M sorbitol. Under these conditions, no significant differences were observed between the mutants and the wild-type strain (Figure 3). Disruption of phospholipid metabolism has been reported to increase sensitivity to oxidative stress due to impaired mitochondrial function and respiration (Grant et al., 1997; Chang and Doering, 2018; Konarzewska et al., 2019). However, consistent with the osmostress assays, no growth defects were detected for either mutant or wild-type strains, when cultured on media supplemented with sodium nitrite (NaNO₂) and hydrogen peroxide (H₂O₂) (Figure 3).

The ability to rapidly produce melanin is a critical C. neoformans virulence attribute, enabling the pathogen to survive inside host macrophage by protecting against reactive oxygen species (ROS), thereby favoring systemic dissemination (Wang and Casadevall, 1994; de Sousa et al., 2022). Melanin production was analyzed after 96 hours of induction at 37°C in MM agar plates supplemented with L-DOPA with or without GPC. Both opi3Δ and the opi3Δpct1Δ mutants exhibited reduced melanization, even in the presence of GPC (Figure 4). Additional virulence-related phenotypes, including polysaccharide capsule expansion and titan cell formation – key traits for immune evasion and lung infection establishment (Okagaki and Nielsen, 2012; Decote-Ricardo et al., 2019), were severely impaired under nutrient-limited conditions but fully restored upon GPC supplementation (Figure 5). Furthermore, GPC supplementation also increased the cell diameter of the opi3Δpct1Δ double mutant in comparison to the wild-type (Figure 5B), indicating physiological alterations linked to titanization. Other virulence-associated factors such as urease and phospholipase secretion were also investigated, but no differences were detected between mutants and wild-type strain (Supplementary Figure 5).

Cell viability of PC mutants was accessed in MM by culturing cells and determining CFUs over time. The results indicated that although the opi3Δpct1Δ mutant was capable of growth, its growth rate was markedly slower than the wild-type, with a notable decline in viability occurring around 15 hours post-inoculation. In contrast, opi3Δ mutant maintained viability for a longer period before exhibiting a gradual reduction (Figure 6A). These observations guided the design of subsequent experiments, including PC quantification, using viable cells in a PC-depleted state. Under these conditions, wild-type, opi3Δ, pct1Δ and opi3Δpct1Δ strains were cultivated and their phospholipids were extracted for analysis. Quantification of total Pi and PC levels revealed a significant reduction in PC content in the opi3Δ and opi3Δpct1Δ mutants, and to a lesser extent in the Kennedy pathway mutant, pct1Δ; however, total Pi levels were increased only in the opi3Δpct1Δ mutant, suggesting accumulation of other phospholipids species (Figure 6B). Chromatographic and quantitative analysis showed that loss of Opi3 was the major determinant of PC depletion in C. neoformans, while PCT1 appears to play a constitutive role in PC biosynthesis beyond its function in the salvage pathway (Figures 6C,D). Notably, deletion of PCT1 resulted in elevation of PE levels in the opi3Δpct1Δ mutant (Figures 6C, D), reflecting a compensatory mechanism to preserve membrane homeostasis under PC-depletion condition.

Previous studies with yeasts have shown that deletion of OPI3 gene disrupts lipid homeostasis and leads to increased lipid droplet size (Lee et al., 2024). Lipid droplets are cellular organelles composed of a phospholipid monolayer surrounding a neutral lipid core rich in triacyclglycerols (TAGs) and sterols (Vevea et al., 2015). They play roles in various physiological functions such as lipid storage, membrane homeostasis, stress responses and autophagy (Graef, 2018). To investigate potential alterations in lipid metabolism, PC synthesis mutant strains were cultured in liquid MM for 15 hours and analyzed by flow cytometry and fluorescence microscopy using Nile Red staining. The results revealed that the opi3Δpct1Δ strain displayed a mean fluorescence index (MFI) approximately twice that of the wild-type, which indicates a significant accumulation of neutral lipids. Reintroduction of either OPI3 or PCT1, significantly reduced the Nile Red fluorescence (Figure 7). Notably, the opi3Δ mutant did not show a significant increase in lipid accumulation, possibly due to compensation by the intact Kennedy pathway utilizing the TAG precursor, diacylglycerol, for PC biosynthesis.

Previous studies have shown that PC depletion in S. cerevisiae yeast cells negatively impacts mitochondrial morphology and function (Bao et al., 2021). To evaluate mitochondrial activity and ROS levels in the disrupted PC synthesis mutant strains, Mitotracker orange and DCFDA staining fluorimetric analysis were carried out (Supplementary Figure 5, Figure 8). Mitotracker is a membrane potential-dependent probe that labels mitochondria in live cells, while DCFDA is used to detect cellular ROS and H₂O₂ (Maghzal et al., 2012). Our study observed no impact in mitochondrial activity after 15 hours under nutrient limited conditions (Figure 8), however, intracellular levels of ROS were elevated in the opi3Δpct1Δ double mutant (Figure 8B). Although endoplasmic reticulum stress and fatty acid degradation are known to increase cellular ROS levels (Morano et al., 2012), occurrence of mitochondrial defects should not be discarded even though our results pointed to no alterations in membrane potential.

Analysis of EV size distributions revealed distinct profiles among the mutants compared to the wild-type (KN99α). The wild-type strain exhibited a narrow size distribution, with vesicles predominantly around 100–150 nm in diameter. The pct1Δ mutant displayed a slightly broader distribution, with vesicles mainly between 100–200 nm, while the opi3Δ mutant showed a broader range, from approximately 100–300 nm. The opi3Δpct1Δ double mutant exhibited the widest distribution, ranging from 100 up to 400 nm, indicating an additive effect of the two mutations on vesicle heterogeneity (Supplementary Figure 6A). Despite these differences in vesicle size, total protein (Supplementary Figure 6B) and sterol content (Supplementary Figure 6C) per 10⁹ yeast cells were not significantly different among the strains, confirming that the observed size changes were not due to variations in overall vesicle yield.

To evaluate virulence-related phenotypes, the survival and proliferation of opi3Δ, pct1Δ and opi3Δpct1Δ mutant strains were first investigated after co-culture with murine macrophages. Following phagocytosis, the killing rate of OPI3 deleted mutants (opi3Δ and opi3Δpct1Δ) was significantly higher compared to the pct1Δ mutant and wild-type strains (Figure 9A). However, in vivo assays using G. mellonella larvae revealed that the double mutant was hypovirulent, while the opi3Δ and pct1Δ single mutant were slightly less virulent compared to the wild-type strain (Figure 9B). To determine whether these results would be translated to a mammalian host, mice were infected with the mutant strains and survival curves along with fungal burdens were analyzed. The opi3Δpct1Δ strain was confirmed to be hypovirulent, as infected mice survived longer than those infected with wild-type, opi3Δ, and pct1Δ strains (Figure 9C). Interestingly, fungal burden revealed that although the double mutant colonized the lungs at similar levels to the wild-type, its presence in the brain was markedly reduced. These findings suggest that while opi3Δpct1Δ mutant can establish pulmonary infection, its ability to disseminate and/or survive in central nervous system is virtually ablated, resulting in reduced overall pathogenicity.