Nutrient sensing and utilization are critical for microbes to adapt to their environment and evade predators. The encapsuled yeast Cryptococcus neoformans has a unique ability to sense and utilize the sugar inositol as both a signaling molecule and as a carbon source. This trait is advantageous not only in the natural environment but also to promote its pathogenesis, including invasion of the central nervous system. In this study, we have identified a novel inositol transporter-like receptor (transceptor) Itr4 that coordinates inositol and glucose metabolism by regulating inositol transporters (ITRs) and glucose sensor Hxs1. Itr4 binds inositol and regulates the function of other ITRs and its loss leads to a significant decrease in inositol uptake activity, resulting in a lack of mating, reduced capsule size, reduced blood–brain barrier (BBB) penetration, and virulence attenuation. Mutagenesis analysis of Itr4 protein identified an essential N-terminal tail and two amino acids, Q388 and Q389, as sites of functional importance. The ITR4^{Q388A, Q389A} allele showed a dominant phenotype with increased inositol uptake, enlarged capsule and significantly attenuated virulence. In summary, we identify a new C. neoformans inositol transceptor, Itr4, the first such example in eukaryotes, with novel regulatory roles in both inositol and glucose metabolism and fungal virulence.

Itr4 is a group II ITR homolog required for polysaccharide capsule formation

The production of a polysaccharide capsule is a key virulence factor that plays a critical role in Cryptococcus pathogenesis. Previously, we demonstrated that inositol regulates both capsule size and polysaccharide structure [21,25]. We also identified the expanded gene family of inositol transporters (ITRs) and inositol oxygenases (MIOs) and revealed that both the two major group I ITRs (ITR1A and ITR3C) and MIOs are required for fungal virulence [25]. However, the function of the group II ITRs (ITR4, ITR5 and ITR6) has not been studied in detail. To further understand the role of the group II ITRs in capsule regulation, we generated the itr4∆, itr5∆, and itr6∆ single mutants and their triple mutants (itr4∆ itr5∆ itr6∆) and measured the size of their capsules. We found that while the itr5∆ and itr6∆ single mutants did not exhibit obvious capsule difference, both the itr4∆ single mutant and the triple mutant cells had significantly reduced capsule and cell size (Fig 1A and 1B). The capsule defect in itr4∆ was fully complemented by reintroducing the ITR4 allele. In addition to having a capsule, thermotolerance and melanin production are two other classical virulence factors in C. neoformans. Our phenotypic analysis showed that all three itr single and triple mutants had normal growth at 37 °C with proper melanin production (S1A and S1B Fig).

We also evaluated the possibility that group II ITRs could be involved in stress responses. In our in vitro phenotypic assays, the itr null mutants showed normal growth under osmotic stress (1.5 M NaCl or 1.0 M Sorbitol) and oxidative stress (3.0 mM H₂O₂) (S1C Fig). We further examined the cell integrity of itr mutants by applying chemicals that target cell integrity, including SDS, calcofluor white (CFW), and Congo Red. SDS disrupts the plasma membrane and lyses cells with a defective membrane, while CFW specifically binds to chitin and Congo Red binds to glucan and chitin in the cell wall to disrupt cell wall integrity. Our results showed that the itr4∆ mutant had a growth defect on SDS but not CFW or Congo Red, indicating a defect in plasma membrane integrity (S1D Fig).

Itr4 regulates sexual reproduction

We previously found that inositol is important for stimulating Cryptococcus mating to help the fungus complete its life cycle and several ITRs in Group I are required for fungal mating [11,26]. To determine the potential involvement of group II ITRs in fungal mating, we tested their single and triple mutants in both unilateral (mutant × wild type) and bilateral (mutant × mutant) mating assays on MS medium. Significant reduction of mating hyphal production and defects in sporulation were observed in the unilateral mating assays of the itr4∆ single mutant and itr4∆ itr5∆ itr6∆ triple mutant, while the mating was completely blocked in their bilateral mating assays. No obvious mating defect was observed in unilateral and bilateral mating assays of itr5∆ and itr6∆ single mutants. The mating defect of itr4∆ could be fully complemented (Fig 1C), demonstrating that Itr4 is essential for fungal mating.

Itr4 regulates inositol uptake activity

To understand whether the mating and capsule formation defect of the itr mutants were related to the defect in inositol uptake activity, we performed inositol uptake assays for parental wild type KN99α and the group II itr single mutants, as well as the itr4∆ itr5∆ itr6∆ triple mutant. Consistent with our previous report [21,23], our results showed that the itr4∆ mutant had significantly reduced inositol uptake activity when compared to KN99α. Similar to itr4∆, the triple mutants also manifested reduced inositol uptake activity, while itr5∆ and itr6∆ single mutants showed wild-type level of inositol uptake activity (Fig 1D). These results also indicate that Itr4 is the main ITR in group II with important biological function.

Itr4, but not Itr5 or Itr6, is required for full fungal virulence

Our previous studies showed that Itr4 is required for full cryptococcal virulence, and the outcome of infection in animal models is mouse strain-dependent, due to difference in host immune responses [23,24,27]. Here we evaluated the potential involvement of every group II ITR in fungal virulence in a murine infection model of systemic cryptococcosis. Ten C57BL/6 mice per group were inoculated with 1 × 10⁴ yeast cells per mouse of each strain intranasally, and their survival rates were determined. Our results showed that while mice infected with the itr4∆ mutant, or the itr4∆ itr5∆ itr6∆ triple mutant had delayed mortality, the ones infected with the itr5∆ or it6∆ mutants showed a survival rate similar to KN99α-infected ones. The virulence attenuation exhibited in itr4∆ was fully rescued by reintroducing the ITR4 allele (itr4∆ + ITR4) in the same murine model (Fig 1E). These results demonstrate that Itr4, but not Itr5 or Itr6, was required for full virulence, which is consistent with our previous report that Itr4 is required for full fungal virulence in C57BL/6 mice [23]. Our previous studies also demonstrated that the ITR4 gene showed a significant polymorphism in clinical isolates from patients with cryptococcal meningitis [23]. Therefore, we focused on the functional analysis of Itr4 to better understand the inositol regulation of Cryptococcus pathogenesis.

Itr4 is an inositol transceptor regulating inositol sensing

To better characterize Itr4 function, we expressed ITR4 in a Saccharomyces mutant strain with deletions of both ITR1 and ITR2 (Sc itr1 itr2) that lacks inositol uptake activity [21]. Full-length cDNAs of ITR4 was cloned into the yeast expression vector pTH74 under the control of the ADH1 promoter to generate a ITR4:GFP fusion construct. The strains expressing ITR4:GFP localized most of their GFP fluorescence signals to the plasma membrane. The control strain expressing the empty vector had diffuse GFP fluorescence signal in cytosol (Fig 2A).

We measured inositol uptake activity for the Sc itr1 itr2 mutant strain expressing C. neoformans ITR4 by using ³H-labeled inositol (myo-[2-³H] inositol) at 30 °C as described previously [21]. The yeast strain expressing ITR4 had detectable signal but was significantly lower than the parental wild type BY4741 or the Sc itr1 itr2 strain expressing the major inositol transporter ITR3C (Fig 2B), and close to the background noise. To determine whether the weak signal was due to inositol uptake or binding, we performed inositol binding assay at 0 °C and detected a weak ³H signal in the ITR4 yeast-expressing strain that was similar to the signal detected in our uptake assay at 30 °C (Fig 2C). This result showing a lack of inositol uptake in the heterologous system indicates that Itr4 may not function as a typical transporter rather as an inositol sensor (transceptor). Transceptors are membrane proteins that combine transporter and receptor functions by mediating nutrient uptake while simultaneously initiating signaling responses [28].

Itr4 is not required for glucose uptake activity in a heterologous expression system

There was a possibility that Itr4 may transport other sugars, e.g., glucose. To examine whether Itr4 functions as a glucose transporter, the ITR4:GFP fusion construct was expressed under the control of the ADH1 promoter in the S. cerevisiae hxt mutant strain EBY.VW1000, in which all 20 hexose transporters are deleted [29]. The EBY.VW1000 strain thus cannot grow on YPD medium but can grow on maltose as a carbon source. The expression of ITR4 was confirmed by GFP signals and qRT-PCR. Our results showed that Itr4:GFP was localized on the plasma membrane as expected (Fig 2D).

The S. cerevisiae control strain CEN.PK2.1C, EBY.VW1000 expressing the empty vector or P_ADH1-ITR4:GFP were used for growth assay on YPD or YP+Maltose. On YPD there was no growth of the EBY.VW1000 expressing either empty vector or ITR4:GFP, while the growth defect was rescued on YP+Maltose condition, indicating the ITR4-expressing strain can’t utilize glucose for their growth, likely due to the lack of glucose uptake by Itr4 (Fig 2E). To further confirm this, we examined the glucose uptake by Itr4 in this heterologous expression strain and found that no glucose uptake was detected in strains expressing ITR4 or empty vector, while a control strain expressing Cryptococcus high-affinity glucose transporter Hxs1 showed glucose uptake activity, confirming that Itr4 is not a glucose transporter (Fig 2F).

We posit that Itr4 may function as an inositol transceptor to regulate inositol uptake and metabolism. Such a sensory function could explain why the itr4∆ mutant, in which all other 9 ITRs remain present, showed significant defects in inositol uptake, virulence factor development, fungal mating, as well as fungal virulence (Fig 1). To test this hypothesis, we determined the potential role of Itr4 in the regulation of other ITR family members by comparing the expression changes of remaining ITR genes in KN99a and its itr4∆ mutant. Our qRT-PCR results showed that the ITR4 expression level was highly induced in inositol condition compared to glucose condition (Fig 2G), and that most ITR genes had reduced expression in itr4∆ compared to wild type, indicating that they were upregulated by Itr4 (Fig 2H). Based on this result and also results from Itr4 inositol uptake and binding activity shown in Fig 2C, we confirm that Itr4 is an inositol transceptor.

Receptors often form homodimer or heterodimer structure to sense signals and activate downstream pathways. To test the hypothesis that Itr4 may directly interact with other ITRs to form dimer structure to regulate inositol uptake, we performed protein-protein interaction assays using the split-ubiquitin system, a membrane-based yeast two-hybrid system [21]. We detected a strong interaction between Itr4 with itself, a weak interaction with Itr3B, and no interaction with two major inositol transporters Itr1A and Itr3C, based on both yeast growth on SD drop out medium and ß-galactosidase activity (Figs 2I and S2B). These results indicate that Itr4 can form a homodimer and also a heterodimer with Itr3B (Fig 2I). Predicted protein dimer structure models were presented in Fig 2J.

In aggregate, our results demonstrate that Itr4 is an inositol-specific transceptor to sense inositol and regulate inositol uptake in C. neoformans. This assessment is consistent with the result that the itr4∆ single mutant showed significant defects in inositol uptake and exhibited strong inositol-related phenotypes.

Itr4 N-terminal tail is required for maintaining the Itr4 membrane localization and function

To better understand the potential domains and amino acid sites required for Itr4 function, we generated a structure model for the Itr4 protein with the AlphaFold 2 program [30] (S3 Fig). Interestingly, protein structural modeling and protein sequence topological analysis showed that Itr4 has a unique long N-terminal cytosolic leader sequence (120 amino acids) before the first transmembrane domain (Figs 3C and S3). Structural modeling suggests that this region may be required for the Itr4 dimerization (S2C Fig). Previous studies reported that glucose transceptors, e.g., Snf3 and Rgt2 in S. cerevisiae, have a long C-terminal tail that is required for their function as a glucose sensor by binding to membrane-associated casein kinases [31]. To investigate the potential function of the Itr4 N-terminal leader sequence, we expressed an ITR4 gene lacking this region (ITR4^∆11-119) in the Sc itr1 itr2 mutant strain (Fig 3A). While the strain expressing ITR4:GFP showed plasma membrane localization, the strain expressing ITR4^∆11-119 was no longer localized to the plasma membrane and instead showed an exclusively cytosolic GFP signal (Fig 3A). The inositol uptake assay also showed that the Sc itr1 itr2 strain expressing ITR4^∆11-119 did not have uptake activity (Fig 3B).

In addition, expression of the ITR4^∆11-119 in the C. neoformans itr4∆ mutant did not rescue the capsule defect phenotype, indicating a critical role for the N-terminal sequence in Itr4 function (Fig 4A). To test whether the intracellular fluorescent signal observed in the fluorescently tagged ITR4^∆11-119 alleles due to protein misfolding and subsequential degradation, we monitored the stability of fusion proteins in a western blot. Our data showed that in the S. cerevisae strain expressing ITR4^∆11-119:GFP allele, the fusion protein was stable (S5A Fig). However, we failed to detect the ITR4^∆11-119:mCherry protein that was expressed in C. neoformans itr4∆ background (S5B Fig). This data indicate that the N-terminus-truncated Itr4 protein may not be stable, leading to mislocalized mCherry signal. We posit that the N-terminal sequence is essential for proper Itr4 dimerization, cellular localization and function. It remains to be studied whether this region may be required for sensing inositol and activating downstream signaling, a function similar to the C-terminal tail of Snf3 and Rgt2 in S. cerevisiae [31].

Identification of key amino acids required for Itr4 function

Utilizing the Itr4 structural model, we further identified two predicted inositol binding sites (Q388 and Q389) (Figs 3C and S3A). The F497 residue was in close proximity to the ligand in the bulding pocket but not directly interact with inositol in our structural model, and was selected as a control. To determine their potential importance for Itr4 function, we generated a ITR4^{Q388A, Q389A} allele and a ITR4^F497A allele and expressed them in the Sc itr1 itr2 heterologous system. Membrane localization was detected for both mutant alleles (Figs 3A and S4A). Inositol uptake assays showed that yeast strain expressing the ITR4^{Q388A, Q389A} mutation allele had increased levels of inositol uptake activity compared to the strain expressing the wild-type ITR4 (Fig 3B), while the ITR4^F497A allele did not exhibit any difference (S4B Fig). These data indicate that these two glutamine amino acids may function as a gatekeeper to limit inositol transport, and their mutations partially open the channel.

To test this model in silico, we performed molecular dynamics (MD) simulation analysis to simulate the interaction between inositol and Itr4 and Itr4^{Q388A, Q389A} protein. Our data suggest that Q388 and Q389 are critical residues of inositol recognition and retention within the Itr4 transporter (Fig 3C–3H). Mutation on these residues decreases ligand stability in the binding pocket, facilitating inositol translocation through the transmembrane channel and supporting the role of the Q388A/Q389A variant as a transporter state that permits enhanced inositol passage across the membrane, which is consistent with the results of inositol uptake assay (Fig 3B). Our structure modeling also showed that the Q388A, Q389A mutation led to a large central cavity with lower interacting energy potential that may be more amenable to transport (S3 Fig). In addition, the Q388A/Q389A mutation also showed an increased frequency in forming hydrogen bond with inositol at Q262, suggesting a possibly more stable or persistent binding with inositol at that site (Fig 3E and 3H). It is likely the mutation also enhanced inositol binding at a secondary site, leading to stronger activation of downstream signaling. Therefore, the Q388A, Q389A point mutation produced a dominant active Itr4 allele that can both sense and transport inositol.

The ITR4^{Q388A, Q389A} allele yields larger capsule but typical GXM particles

The functional importance of Itr4 N-terminus and the putative inositol-binding sites in C. neoformans were assessed using an mCherry-tagged strain for Itr4, under the control of its native promoter (P_ITR4-ITR4:mCherry). Utilizing the Actin promoter, we also generated strains overexpressing ITR4 (P_ACT1-ITR4:mCherry), ITR4^∆11-119 (P_ACT1-ITR4^∆11-119:mCherry) or ITR4^{Q388A, Q389A} (P_ACT1-ITR4^{Q388A, Q389A}:mCherry) alleles. The expression and stability of ITR4, ITR4^∆11-119 and ITR4^{Q388A, Q389A} alleles were evaluated by the mCherry signal under fluorescence microscopy (Fig 3I) and direct protein detection in a western blot using anti-mCherry antibody (S5 Fig). While the strains expressing ITR4 or ITR4^{Q388A, Q389A} retained membrane localization, the strain expressing the ITR4^∆11-119 allele lost most of plasma membrane localization and instead most mCherry signal was localized in cytosol. Interestingly, detection of the Itr4^∆11-119:mCherry protein using mCherry antibody was not successful, indicating that this protein may be unstable and could be misfolded and degraded (S5 Fig). The control strain expressing P_ITR4-ITR4:mCherry manifested a weak plasma membrane fluorescence signal, suggesting a lower expression level compared to that under the control of the Actin promoter (Fig 3I and 3J).

We further measured inositol binding and uptake activities for the itr4∆ strain expressing of the ITR4 and ITR4^{Q388A, Q389A} alleles by using ³H-labeled myo-inositol at 0 and 30 °C as described previously [21]. Our results showed that the itr4∆ strain had lower binding and uptake signals than wild type, while the itr4∆ expressing ITR4^{Q388A, Q389A} alleles had significantly higher binding activity (Fig 3K). This result indicates that the ITR4^{Q388A, Q389A} allele may also enhance the inositol binding to the ITR4 ligand-binding site in addition to increasing inositol transport activity, thus promoting inositol signaling activation.

To investigate the Itr4 localization changes in response to different carbon sources, strains expressing P_ITR4-ITR4:mCherry were grown in YPD overnight and reinoculated to YNB + 0.1% inositol, YNB + 1% inositol and YNB + 1% glucose media and incubated for 2 hours. Our data showed that, when P_ITR4-ITR4:mCherry-expressing cells grown on YNB + 1% glucose, the mCherry fluorescence signal was weakly localized to the plasma membrane, and its expression increased when switched to YNB + 0.1% inositol medium. When cells were transferred to YNB + 1% inositol condition, the fluorescent signal also increased, but was detected mostly in cytosol, indicating an internalization of the Itr4:mCherry protein. The percentage of cells with membrane localization under different conditions was quantified (Fig 3L and 3M).

Consistent with our previous studies demonstrating that inositol can significantly affect the capsule [25], we found that the itr4∆ cells produced smaller capsule sizes (Fig 4A). To understand how Itr4 regulates capsule growth, we examined capsule in the itr4Δ mutat and its ITR4^OE, ITR4^∆11-119 and ITR4^{Q388A, Q389A} expressing strains under in vitro capsule-inducing conditions [32]. We observed a significantly increased capsule size in the ITR4^OE and ITR4^{Q388A, Q389A} strains and a reduced capsule size in the itr4Δ and ITR4^∆11-119 strain (Fig 4A and 4B). The large capsule size of the strain expressing the ITR4^{Q388A, Q389A} allele is consistent with our hypothesis that it is a dominant active allele.

We further analyzed capsule structural changes in the itr4∆ backgrounds by purifying the extracellular polysaccharides (EPS) secreted from KN99, itr4∆ and ITR4^{Q388A, Q389A} strains. Our 1D [¹H] NMR measurement showed no change in EPS structure or Glucuronoxylomannan (GXM) O-acetylation pattern between KN99, itr4∆ and ITR4^{Q388A, Q389A} strains (S6 Fig). Further analysis by SEC-MALS (Size Exclusion Chromatography coupled with Multi-Angle Light Scattering) shows that while the structure of the EPS polymers did not change, the size of the polymers in both the itr4∆ and ITR4^{Q388A, Q389A} strains were roughly half that of the parental strain (Fig 4C; S7 Table). In addition, the ITR4^{Q388A, Q389A} allele also showed less capsule density than either the parent or itr4∆ mutant. These data suggest that the EPS polymer size, rather than polymer structure, contributes to the changes in biological function of the itr4∆ mutant strains and their interactions with the host.

Itr4 is required for C. neoformans to overcome the anti-fungal activity of macrophages

A key characteristic of C. neoformans as a facultative intracellular pathogen is its ability to survive and replicate within the phagolysosome of macrophages [33]. The capsule has antiphagocytic effects during the macrophage-pathogen interaction [34]. To investigate whether Itr4 was required for fungal cell survival in macrophages, we performed Cryptococcus-macrophage interaction assays using the macrophage-like cell line J774. After coincubation with C. neoformans for 4 h, J774 cells were fixed with methanol and stained with Giemsa staining, and the phagocytosis rate was determined by the number of yeast cells inside each J774 cell. Macrophage-like cells co-cultured with KN99a resulted in ~30% of J774 cells containing ingested yeasts, with ~2.0 yeast cells on average per host cell. These phagocytosis numbers increased when co-cultured with the itr4∆ mutant or ITR4^∆11-119, but decreased when co-cultured with ITR4^OE or ITR4^{Q388A, Q389A}-expressing yeast cells, suggesting that Itr4 promotes phagocytic evasion (Fig 5A and 5B). We found that the expression of the Th1 cytokine IFN-γ was significantly increased in the itr4Δ, ITR4^OE and ITR4^∆11-119 co-incubated macrophages (Fig 5C). These observations suggest that increased IFN-γ upon interaction is important during phagocytosis, which is consistent with our previously reported data [24].

Itr4 is required for inositol-mediated fungal traversal across the blood–brain barrier in an in vitro model

Previously, we demonstrated that C. neoformans Itr1a and Itr3c are two major ITRs for inositol uptake and are required for dissemination across the BBB during brain infection [13]. To determine whether the inositol sensor Itr4 was required for fungal transversal across the BBB, we examined the transmigration ability of the itr4∆ mutant in the in vitro BBB model as previously described [13]. In the absence of additional inositol, the itr4∆ transmigration rate decreased by ~50% at 3 and 24 h incubation period compared to the wild type, indicating that Itr4 is required for cryptococcal crossing of the BBB (Fig 5D). Under this condition, the Transendothelial Electrical Resistance (TEER) measurement indicated that the membrane was intact (Fig 5E). With inositol treatment, cryptococcal traversal was enhanced regardless of the presence of ITR4; however, the itr4∆ mutant showed significantly less transmigration than the parental control. The transmigrated wild-type cells increased by ~3- and ~5-fold in the presence of inositol at 3 and 24 h incubation, respectively, compared to in the absence of supplemental inositol. In contrast, the itr4∆ transmigration rate was similar to the control (Fig 5F). We also tested the ITR4^{Q388A, Q389A} strain for its transmigration efficiency using the in vitro BBB model. Our data showed that this mutant had reduced transmigration, likely due to its enlarged capsule size, but an increased transmigration rate was observed in the presence of additional inositol (Fig 5G).

To address the concern that the CFU of different strains may be influenced by growth rate difference, we compared the growth rate of wild type and the itr4∆ mutant in medium with or without addition of inositol. The results showed that all tested strains have similar growth rates regardless of the presence of additional inositol (Fig 5H). Thus, the data are consistent with our interpretation that Itr4 plays an important role in responding to inositol availability and contributes to cryptococcal traversal across the BBB.

The ITR4^{Q388A, Q389A} allele shows attenuated virulence

In vitro analyses showed that Itr4 is important for capsule formation, mating, macrophage phagocytosis and BBB crossing, suggesting a potential role in host-pathogen interactions. We then examined the virulence of the itr4Δ, ITR4^OE, ITR4^∆11-119 and ITR4^{Q388A, Q389A} strains in a murine inhalation infection model. Our results showed that mice infected with itr4Δ, ITR4^∆11-119, ITR4^OE or ITR4^{Q388A, Q389A} strain all manifested prolonged survival than mice infected with wild-type control, indicating a virulence attenuation (Fig 6A). These results indicate that the ITR4^{Q388A, Q389A} allele had significant virulence attenuation.

To better understand the mechanism of altered virulence of these itr4 mutant strains, fungal burdens in lungs, brains, and spleens of 3 mice per group were examined at day 7 post-infection (7 DPI). While significantly reduced CFU was recovered in the lungs infected with the ITR4^{Q388A, Q389A} allele compared to wild-type infected ones, no CFUs were recovered in brains and spleens of those animals, indicating a lack of dissemination at this time point (Fig 6B). Histopathology results of infected organs demonstrated less severe lesion development in lungs and brains infected by the itr4∆, ITR4^OE or ITR4^{Q388A, Q389A} strain (S7 and S8 Figs), which are consistent with their virulence attenuation. At the endpoint, while all mice contained high levels of CFUs in the infected lungs, several of the itr4∆, ITR4^OE, ITR4^∆11-119 and ITR4^{Q388A, Q389A} strains were cleared of their brain or spleen infection, with some of them containing low levels of fungal CFUs compared to wild-type control group (S8 Fig).

We isolated yeast cells from bronchoalveolar lavage fluid (BALF) of mice infected with different strains and analyzed their capsule size to examine the role of Itr4 in capsule size regulation during infection. Compared to wild-type KN99a, we detected significantly increased capsule size in ITR4^OE and ITR4^{Q388A, Q389A}, but smaller capsule in itr4∆ and ITR4^∆11-119 cells isolated from infected mice, which is consistent with the observations at in vitro conditions (Fig 6C and 6D). Altogether, these data confirm that Itr4 is involved in capsule size regulation during infection in vivo, and a significantly enlarged capsule may have slowed down the yeast dissemination, leading to virulence attenuation.

To understand how these capsuel changes in different strain background impact host immune responses, we analyzed the expression of chemokine genes in lung tissues infected with different strains at 7 DPI by qRT-PCR. Uninfected C57BL/6 mice were analyzed as a control. We found that the expression of IFN-γ was significantly increased in the lungs of mice infected with itr4Δ, and significantly decreased in ITR4^{Q388A, Q389A} infected mice (S7A Fig). These observations suggest that increased influx of IFN-γ upon infection with the itr4Δ mutant is associated with containment of the infection, which is consistent with our previously reported data [24]. Our lung gene expression analysis also showed that genotype-dependent differences in cytokine and macrophage polarization gene expression coincided with fungal burden and survival outcomes. Compared to the parental strain, itr4Δ infected lungs showed higher expression of IFN-γ, IL-12a, IL-12b, and TNF-α, followed by increased CCR2 and NOS2 expression but decreased ARG1 expression. In contrast, the ITR4^{Q388A, Q389A} strain showed reduced expression of CCR2 and NOS2 genes, with lower ARG1 and IL-10 levels. IL-4 expression remained high across groups (S7A Fig). Together, these data indicate that differential modulation of lung innate and chemokine responses distinguishes Itr4 alleles and aligns with their respective CFU and survival phenotypes in animal models.

Involvement of ITR4 in in vivo brain infection in a mouse intravenous model

Given that Itr4 was required for BBB crossing in our in vitro BBB model (Fig 5D and 5F), we evaluated the involvement of Itr4 in BBB crossing in vivo by infecting mice intravenously with the itr4Δ, ITR4^OE, and ITR4^{Q388A, Q389A} strains, respectively. All these manifested significant virulence attenuation (Fig 6E), consistent with our observation in the inhalation animal infection model.

To determine the rate of fungal transversal across the BBB to cause brain infection in the intravenous infection model, lungs, brains, and spleens from mice injected with different itr4 mutants were isolated at 3 DPI and their fungal burdens determined by yeast CFUs. Our results showed that there was a significant difference in fungal burden in the brain between mice infected by wild-type and the itr4Δ mutant at 3 DPI. On the other hand, the fungal burden in all three organs was significantly decreased in ITR4^OE and ITR4^{Q388A, Q389A} infected mice (Fig 6F). Our results thus demonstrate that the inositol transceptor Itr4 is required for fungal cells to either cross the BBB or grow in the brain after transmigration.

To examine the role of Itr4 in capsule size regulation during CNS infection, we homogenized the brain tissues infected with different strains and analyzed their capsule size. Similar to what we have observed in vitro conditions, we detected significantly enlarged capsule of the ITR4^OE and ITR4^{Q388A, Q389A} cells isolated from infected mouse brains, while the capsule size was smaller in the itr4∆ mutant (Fig 6G and 6H). Histopathology of infected mouse brain sections also indicated that the ITR4^OE and ITR4^{Q388A, Q389A} capsules were larger than the wild-type and itr4∆ mutant strains (S7 and S8 Figs). Altogether, these results further confirmed that fungal Itr4 plays an important role in capsule size regulation and fungal pathogenesis, including Cryptococcus traversal of the BBB in CNS infection.

Itr4 regulates genes involved in inositol metabolism during inositol incubation

To further understand how Itr4 regulates Cryptococcus cells responding to inositol, we performed RNA-Seq assays to analyze transcriptional profiles of the Cryptococcus wild-type KN99a and itr4∆ mutant that were either treated with or without inositol for 24 h. Our RNA-seq data identified a large number of differentially expressed genes (DEG) when compared between KN99a and itr4∆ under YNB with either only 1% glucose or both 1% glucose and 1% inositol presence (Fig 7A). Analyzing the differentially expressed genes (DEGs) using Ingenuity Pathway Analysis (IPA) identified several pathways that were upregulated by Itr4. A KEGG (Kyoto Encyclopedia of Genes and Genomes) pathway enrichment analysis highlighted the overrepresented pathways, including: ‘Carbon metabolism’ (map01200, p = 0.0018), ‘Valine, leucine and isoleucine degradation’ (map00280, p = 0.0022), ‘Citrate cycle (TCA cycle)’ (map00020, p < 0.0087), ‘Pyruvate metabolism’ (map00620, p = 0.0117), and ‘Cell cycle - yeast’ (map04111, p = 0.0155) (Fig 7C). A GO enrichment analysis plot (Fig 7D) revealed the top Biological Process (BP) terms include ‘DNA replication’ (GO:0006260, p < 0.01),‘intracellular signal transduction’ (GO:0035556, p < 0.05), and ‘organic acid metabolic process’ (GO:0006082, p < 0.01).

When analyzing the genes involved in inositol function, we found that genes involved in inositol pyrophosphate biosynthesis and inositol de novo biosynthesis were mostly upregulated in the itr4Δ mutant background, while many genes involved in inositol uptake and inositol catabolism were down-regulated (Fig 7B; S8 Table). The Itr4 regulation of other ITR genes observed in RNA-seq data is consistent with our qRT-PCR data (Fig 2H). Cryptococcus can use inositol as a carbon source. Conversion of inositol to glucuronic acid by inositol oxygenases (MIOs) is the first step of the only known pathway for inositol catabolism [25]. We found that the gene encoding inositol oxygenase 1 (MIO1) was induced significantly less by inositol treatment in itr4∆ than in KN99a (Fig 7B; S8 Table). These gene expression data support a role for Itr4 in regulating the expression of genes involed in inositol acquisition and inositol catabolism.

Interestingly, the glucose transceptor homolog gene HXS1 (High affinity glucose transporter, CNAG_03772) [35] was significantly repressed (>125-fold) in the itr4∆ cells, so was the MIG1 homolog, while the SNF1 homolog was significantly induced. Because MIG1 and SNF1 are two key carbon catabolite repression (CCR)-associated genes in S. cerevisiae, our data indicate that Itr4 may also regulate glucose signaling. Indeed, several hexose transporter homologs were also differentially expressed in wild-type and itr4∆ cells (Fig 7B; S8 Table). In addition, SCW1 (cell wall integrity protein 1, CNAG_05663) [36], a gene required for cell wall integrity, and BAT1 (Branched-chain-amino-acid transaminase, CNAG_05664) [36], a gene involved in amino acid metabolism, were completely silenced in the itr4∆ mutant background under the same culture condition (S4 Table). These data may indicate a connection between inositol sensing and glucose uptake, cell wall integrity and amino acid metabolism, suggesting a crosstalk among different metabolic pathways, and underscores the dynamic role inositol plays in the biology of Cryptococcus.

Fig 8. Proposed model for the role of Itr4 in coordinating carbon utilization and fungal virulence.

Itr4 as an inositol receptor that is required to coordinate inositol and glucose metabolism by regulating ITRs and glucose sensor Hxs1. Itr4 regulates fungal pathogenesis in part through its regulation of capsule development. Itr4 serves as a critical sensor in the process of evolutionary specialization in this model. Created in BioRender. Wang, Y. (2026) https://BioRender.com/z8h6fdo.

https://doi.org/10.1371/journal.pbio.3003709.g008

Ethics statement

All studies were conducted following biosafety level 2 (BSL-2) protocols and procedures approved by the Institutional Animal Care and Use Committee (IACUC) of Rutgers University under protocol 999901066. Animal studies were compliant with all applicable provisions established by the Animal Welfare Act and the Public Health Services (PHS) Policy on the Humane Care and Use of Laboratory Animals.

Strains, media, and growth conditions

C. neoformans strains used in this study are listed in S1 Table. S. cerevisiae strains and plasmids used in this study are listed in S2 Table. All primers used in this study are listed in S3 Table.

Strains were grown at 30 °C on yeast extract-peptone-dextrose (YPD) agar medium and synthetic (SD) medium. YNB medium without inositol was purchased from MP Biomedicals. Niger-seed medium was used to test for melanin production. Dulbecco modified Eagles (DME) medium for assessing capsule production was prepared as previously described [9,55]. A V8 juice agar medium (50 mL V8 original juice, 0.05% KH₂PO₄, and 2% agar, pH 5.0) and a modified Murashige and Skoog (MS) medium minus sucrose (Sigma-Aldrich) were used for mating and sporulation assays. Minimal medium (MM, 15 mM D-glucose, 10 mM MgSO₄, 29.4 mM KH₂PO₄, 13 mM Glycine, and 3.0 μM Thiamine, pH 5.5) was used for capsule induction. All other media were prepared as described previously [26,56,57].

Generation of itr4∆, itr5∆ and itr6∆ mutants

Mutants for ITR4, ITR5 and ITR6 were generated in both KN99α and KN99a strains backgrounds by overlap PCR as previously described [58]. The 5′ and 3′ regions of the ITR4 gene were amplified from KN99α genomic DNA with primer pairs CX1005 and CX1006, and CX1007 and CX1008, respectively. The 5′ and 3′ regions of the ITR5 gene were amplified from KN99α genomic DNA with primer pairs CX998 and CX999, and CX1000 and CX1001, respectively. The 5′ and 3′ regions of the ITR6 gene were amplified from H99 genomic DNA with primer pairs CX991 and CX992, and CX993 and CX994, respectively (S3 Table). Each target gene replacement cassette was generated by overlap PCR. Purified overlap PCR products were precipitated onto 10 µl gold microcarrier beads (0.6 µm; Bio-Rad), and strains KN99α or KN99a were ballistically transformed as described previously. Stable transformants were selected on YPD medium containing G418 (200 mg/L) or NAT (100 mg/L). To screen for itr4∆, itr5∆ or itr6∆ mutants, diagnostic PCR was performed by analyzing the 5′ junction of the disrupted mutant alleles with primers CX1009 and JH8994, CX1002 and JH8994, CX995 and JH8994. Positive transformants were identified by PCR screening with primers CX1010 and CX1011, CX1003 and CX1004, CX996 and CX997, respectively.

To generate complemented strains of the itr4Δ single mutant, a genomic DNA fragments that contained a 1.5 kb upstream region of ITR4 ORF, the ITR4 ORF, and its 500 bp downstream region was amplified by PCR with primers CX1222/CX1223. The ITR4 PCR fragment was inserted into vector pJAF13 using in-fusion cloning. The construct pJAF13-ITR4 was ballistically transformed in itr4∆ mutant strains. The selecting colonies grown on YPD with both nourseothricin and G418 were further confirmed by PCR. Phenotypic assays were performed to identify transformants in which the itr4Δ phenotype was complemented.

To generate overexpression strain of the itr4∆ mutant, the ITR4 full-length genomic DNA was amplified with primers CX1800/CX1801 (S3 Table). The fragment was cloned into the BamHI site of vector pCXU200 containing the Cryptococcus actin promoter to generate a plasmid which contains ITR4:mCherry fusion fragment. Linearized plasmid was introduced into itr4Δ mutant strain to generate overexpression strain that expresses Itr4:mCherry protein.

Generation of triple mutants and complemented strains

In mating assays, C. neoformans cells of opposite mating types were mixed and co-cultured on MS agar medium at 25 °C in the dark for 10 days and filamentation was examined by light microscopy. Spore production was visualized by microscopy and photographed. Basidiospores were dissected from mating assays performed on MS medium.

To generate itr4∆ itr6∆ double mutants, a mating assay between α itr4∆::NAT and a itr6∆::NEO mutants was conducted and spores were isolated with an MSM system (Singer Instrument, England). Genomic DNA was isolated from all progenies that grew on YPD with both nourseothricin and G418. PCR was used to screen for ITR4 gene deletion with primers CX1009/JH8994 and CX1010/CX1011, and for ITR6 gene deletion with primers CX995/JH8994 and CX996/CX997 (S3 Table). To generate itr4∆ itr5∆ itr6∆ triple mutants, a mating assay between α itr4∆::NAT itr6∆::NEO double mutant and a itr5∆::NEO mutant was conducted and spores were isolated. Triple mutants were screened first by PCR on cultures initiated from single spores. All mutants were confirmed by PCR. The mating type of each confirmed mutant strain was determined by PCR using mating type-specific primers and by genetic crossing.

Assays for melanin and capsule production

Assays for melanin and capsule production were performed as previously described [55,59]. In brief, melanin production was assayed by inoculating C. neoformans strains into 2 ml of YPD liquid medium, incubating the culture overnight at 30 °C and spotting 5 µl of each culture with 10 times series dilutions on Niger-seed agar medium. The agar plates were incubated at 30 or 37 °C for 2 days, and pigmentation of fungal colonies was assessed and photographed. To examine capsule production, 30 µl of overnight cultures were inoculated on DME agar medium and incubated at 37 °C for 3 days. Capsule was visualized with India ink negative staining and observed with a 100× Olympus CX41 equipped with an Infinity digital camera (Olympus). Capsule sizes and cell body sizes of more than 100 cells were measured and calculated.

Assays for stress responses and cell integrity

Each strain was incubated overnight at 30 °C in YPD and diluted to an optical density at 600 nm (OD600) of 2.0. The cells were washed, resuspended, and serially diluted (1:10) in dH₂O and spotted (5 μl) on YPD agar plates containing 1.5 M NaCl, or 1.0 M KCl for osmotic shock, or 2.5 and 5.0 mM H₂O₂ for oxidative stress. To test cell integrity, cells were also spotted on YPD agar plates containing 0.05% SDS, 0.03% SDS, 0.5% Congo Red, or 250 μg/ml CFW. Plates were incubated at both 30 and 37 °C for two days and photographed.

Inositol uptake assay

The inositol uptake assay was performed following a previously published method [19,20]. In brief, the Cryptococcus strains were grown in YPD liquid cultures overnight at 30 °C. Cells were diluted in YPD to an OD₆₀₀ of 1.0, grown at 30 °C, and collected at an OD₆₀₀ of 5.0 by centrifugation at 2,600g for 5 min. Cells were then washed twice with PBS at 4 °C and resuspended in 2% glucose to reach a final concentration of 2 × 10⁸ cells/ml as determined by the use of a hemacytometer. For the uptake assay, the reaction mixture (200 l) contained 2% glucose, 40 mM citric acid-KH₂PO₄ (pH 5.5), and 0.15 µM myo-[2-³H]-inositol (MP Biomedicals) (1 µCi/µl). An additional 200 µM concentration of unlabeled inositol (Sigma-Aldrich) was added to the reaction mixtures for competition assays. Equal volumes of the reaction and cell mixtures (60 µl each) were warmed to 30 °C and mixed for the uptake assay, which was performed for 10 min at 30 °C. As negative controls, mixtures were kept at 0 °C (on ice) during the 10-min incubation. Aliquots of 100 µl were removed and transferred onto prewetted Metricel filters (1.2-µm pore size) on a vacuum manifold. The filters were washed four times each with 2 ml of ice-cold water. The washed filters were removed and added to liquid scintillation vials for measurements on a PerkinElmer TRI-CARB 2900TR scintillation counter.

Glucose uptake assay

Full-length cDNAs of the ITR4, ITR5 and ITR6 genes were amplified from C. neoformans KN99a total cDNA and were cloned into the yeast expression vector pTH74 to generate GFP fusion constructs, under the control of the ADH1 promoter. ITR4, ITR5 and ITR6 expression plasmids were introduced into an S. cerevisiae strain EBY.VW1000 that lacks all 20 HXT transporters [29]. The expression of Cryptococcus ITR4, ITR5 or ITR6 in this yeast heterologous system was verified by both GFP localization and qRT-PCR using gene-specific primers. Yeast strains were tested for growth on different medium at 30 °C.

The S. cerevisiae control strain CEN.PK2.1C, the S. cerevisiae hxtΔ mutant strain EBY.VW1000, and EBY.VW1000 expressing empty vector, ITR4:GFP, ITR5:GFP or ITR6:GFP genes from C. neoformans were grown in YP containing 2% maltose liquid cultures overnight at 30 °C. Collected cells were washed with dH₂O, resuspended in YP containing 2% maltose, and incubated for 2 h. Then cells were suspended in PBS at a final concentration of 2 × 10⁸ cells/ml for uptake assay. Each 100 μl cell suspension was mixed with 100 μl labeled glucose (³H-glucose) solution at room temperature. Samples (100 μl) were removed after 30 s, 1 min, 5 min, and 10 min, and mixed with 1 ml ice-cold water to stop the reactions. Cells were immediately collected on fiber filters, washed three times with 10 ml of ice-cold water, and transferred to scintillation vials for measurement.

Detection of ITR gene expression using qRT-PCR

To test how the ITRs respond to the presence of environmental myo-inositol, both in vitro and in vivo, we measured the mRNA levels for ITR genes under different conditions via quantitative real-time PCR (qRT-PCR). Cultures of C. neoformans wild-type strain KN99a and its mutant strains were grown on YPD for 24 h at 30 °C. Collected cells were washed with dH₂O and pellets were used for total RNA extraction. Total RNAs were extracted using Trizol reagents (Invitrogen) and purified with Qiagen RNeasy cleanup kit (Qiagen) following the manufacturer’s instructions. Purified RNAs were quantified using a Nanodrop instrument (Thermo Scientific). First-strand cDNAs were synthesized using a Superscript III cDNA synthesis kit (Invitrogen) following the instructions provided by the manufacturer. Expression levels of ITRs and GAPDH were analyzed with the comparative C_T method using Brilliant SYBR green QPCR reagents (Takara) as described previously [11,21].

Protein-protein interaction assay using the split-ubiquitin system

The split-ubiquitin system was utilized to investigate the interaction between Itr4 and Itr1A, Itr3C and Itr3B as previously described [56]. Vectors and yeast strains were included in the DUALmembrane Kit 2 (Dualsystem Biotech, Switzerland). ITR4 full-length and ITR4^∆11-119 cDNA were cloned into yeast expression vector pCCW (the N-terminal half of the ubiquitin Cub protein was fused to the C-terminus of ITR4 and ITR4^∆11-119), and ITR1A, ITR3C, ITR3B full-length cDNA and ITR4^∆11-119, ITR4^{Q388A, Q389A} cDNA were cloned into the pDL2Nx vector (the mutated C-terminal half of ubiquitin NubG protein was fused to the ITR1A, ITR3C, ITR3B, ITR4, ITR4^∆11-119 and ITR4^{Q388A, Q389A} N-terminus). All cDNA sequences were confirmed by DNA sequencing. Cub and NubG fusion constructs were co-transformed into host yeast strain NMY32. Two constructs, pCCW-Alg5 and pAI-Alg5, and pCCW and pDL2Nx serve as positive and negative control. Interaction was determined by the growth of yeast transformants on medium lacking histidine or adenine, and also by measuring β-galactosidase enzyme activity assays using Chlorophenolred-β–D-galactopyranoside (CRPG; Calbiochem, San Diego, CA) as substrate.

Fluorescent signal quantification and data analysis

Plasmids expressing P_ACT1-ITR4:mCherry fusion protein, P_ACT1-ITR4^∆11-119:mCherry fusion protein and P_ACT1-ITR4^Q388A,Q389A:mCherry fusion protein were linearized and transformed into itr4Δ mutant to generate the itr4Δ ITR4^OE, itr4Δ ITR4^∆11-119 and itr4Δ ITR4^{Q388A, Q389A} strains. Strains were cultured in YPD overnight and were observed with a Nikon fluorescence microscope.

The percentage of mCherry signals localized within the cell (% cytosolocalized mCherry) was calculated in accordance with data analysis seen in Hankins and colleagues [60]. The outside of each cell was outlined using the freehand selection tool in ImageJ and then the total fluorescence of the cell was measured (Fluor_total). The area just within the plasma membrane was then outlined to give the total internal fluorescence excluding the plasma membrane (Fluor_int). The percentage of fluorescence localized within the cell was found by dividing Fluor_int by Fluor_total.

Cryptococcus-macrophage interaction assay

The Cryptococcus-macrophage interaction assay was done as previously described [52]. Macrophage-like J774 cells were cultured in DME medium with 10% heat-inactivated FBS at 37 °C with 5% CO₂. J774 cells (5 × 10⁴) in 0.5 ml fresh DME medium were added into each well of a 48-well culture plate and incubated at 37 °C in 5% CO₂ overnight. To activate macrophage cells, 50 units/ml gamma interferon (IFN-ϒ; Invitrogen) and 1 µg/ml lipopolysaccharide (LPS; Sigma) were added into each well. C. neoformans overnight cultures were washed with phosphate-buffered saline (PBS) twice and opsonized with 20% mouse complement. Cryptococcus cells (2 × 10⁵) were added into each well (yeast/J774 ratio, 4:1). To assess the phagocytosis rate, the cells were washed with PBS after a 1.5 h coincubation and fixed with methanol for 30 min. Giemsa stain was added to the wells at a 1:5 dilution, and the plates were incubated overnight at 4 °C. Cells were washed once with PBS and analyzed using an inverted microscope. To assess intracellular proliferation of C. neoformans, nonadherent extracellular yeast cells were removed by washing with fresh DME medium after a 4 h coincubation and cultures were incubated for another 0, 4, and 24 h. At indicated time points, the medium in each well was replaced with distilled water (dH₂O) to lyse macrophage cells for 30 min at room temperature. The lysate was spread on YPD plates, and CFU were counted to determine intracellular proliferation.

Human brain microvascular endothelial cells (HBMEC) transmigration assay

Primary isolates of HBMEC were cultured as previously described [61]. HBMEC were routinely grown in RPMI 1640 supplemented with 10% heat-inactivated fetal bovine serum, 10% Nu-serum, 2 mM glutamine, 1 mM sodium pyruvate, penicillin (100 units/ml), streptomycin (100 mg/ml), essential amino acids, and vitamins. The cells were incubated at 37 °C in a humidified incubator with 5% CO₂. For all experiments, the same medium used for routine cell culture was used as the experimental medium.

The in vitro human BBB model was generated and used for fungal transmigration assays as previously described [62,63]. HBMEC were seeded on Transwell polycarbonate tissue culture inserts with a pore diameter of 8.0 μm (Corning Costar) and cultured until their TEER reached over 350 V/cm², as measured by an Endohm volt/Ω meter in conjunction with an Endohm chamber (World Precision Instruments). The same medium used for cell culture was used as the experimental medium. Yeast cells were washed with phosphate-buffered saline (PBS) and resuspended in HBMEC culture medium. 10⁵ Cryptococcus cells were added to the top compartment and then incubated at 37 °C. At 3 and 24 h, the medium in bottom compartments was collected and immediately replaced with fresh medium. Fungal cell numbers in the collected medium were addressed by CFU counts to determine the number of transmigrated viable yeast cells. To determine the specificity of myo-inositol, the transmigration assay was performed in the presence of myo-inositol, D (+) galactose or D (+) maltose (1 mM each) in the bottom compartments prior to addition of Cryptococcus cells (10⁵) in the top compartment of Transwells. Results are presented as the total number in the bottom chamber. Each set was triplicated and repeated three times independently [13]. The statistical analysis of the data from our in vitro studies was done with a two-tailed Student t test. Statistical significance was determined at P < 0.001.

RNA-seq quantification analysis

C. neoformans cells of the WT and itr4∆ mutant strains were cultured overnight at 30 °C in YPD liquid medium and then inoculated in YNB + 1% Glucose + 1% Inositol condition for 24 h. The cells were then collected and total RNAs were extracted using Trizol reagents (Invitrogen) and purified with MACHEREY-NAGEL RNA purification kit following the manufacturer’s instructions. Purified RNAs were quantified using a Nanodrop instrument (Thermo Scientific). RNA-Seq analysis was performed by Novogene according to the company’s protocol (https://www.novogene.com/us-en). For the RNA-seq, cDNA library preparation, Illumina sequencing, and data analysis were performed by Novogene (Sacramento, CA). Sequencing was performed using Illumina NovoSeq 6000 to generate 150 bp paired-end reads. Approximately 50 million raw reads were generated. Adapter and low-quality reads or reads with uncertain nucleotides constitute more than 10 percent of either read (N > 10%), and reads with low-quality nucleotides (base quality less than 5) constitute more than 50 percent of the read, were removed. Clean reads were mapped to the annotated genome of C. neoformans H99 (NCBI:txid235443) using HISAT2

software (version 2.0.5). Differential expression analyses were conducted using the DESeq2 package (version 1.20.0) [64]. For each experimental condition included in the RNA-seq analysis, we performed three independent biological replicates. Significance was determined using multiple unpaired T tests with two-stage linear step-up.

Murine infection

Survival curves of infected mice in a Cryptococcus murine inhalation model as previously described [21,25]. Cryptococcus strains were grown at 30 °C overnight and cultures were washed twice with 1× phosphate-buffered saline (PBS) by centrifugation and resuspended at a final concentration of 2 × 10⁵ CFU/ml. Groups of 10 female C57BL/6 mice (Jackson Laboratory, ME) were used for each infection. For the intranasal inhalation model, mice were intranasally infected with 10⁴ yeast cells of each strain in 50 µl PBS as previously described. For the intravenous injection model, 5 × 10⁴ yeast cells in 100µl volume for each strain were inoculated via tail vein injection. Over the course of the experiments, animals that appeared moribund or in pain were sacrificed by CO₂ inhalation. Survival data from the murine experiments were statistically analyzed between paired groups using the log-rank test and the PRISM program 6.0 (GraphPad Software) (P values of <0.001 were considered significant).

Isolation and measurement of yeast cells from mouse BALF

Yeast strains were grown at 30 °C overnight and cultures were washed twice with PBS buffer and resuspended to a final concentration of 2 × 10⁷ cells/mL. Groups of 5 female C57BL/6 mice were intranasally infected with 1 × 10⁶ yeast cells of each strain as previously described [65,66]. BALF samples were harvested at day 3 after inoculation. BALF was collected in 3 ml of 1× PBS buffer using a catheter inserted into the trachea of animal post-euthanasia, and airway-infiltrating cells were lavaged with ∼0.7 ml of 1× PBS at one time to a total volume of 3 ml. Cells were spun down and host cells were lysed by adding sterile ddH₂O and incubating for 45 min at room temperature. Cryptococcal cells were examined under microscope. Capsule sizes and cell body sizes of more than 500 cells were measured and calculated.

Histopathology and immune response analysis for infected organs

Infected animals were sacrificed at the end of the experiment according to the Rutgers University IACUC-approved animal protocol. To compare the fungal burdens and host inflammatory responses, the lungs, brains, and spleen were dissected and fixed in 10% formalin solution for section preparation at Rutgers University Histology Core Facility. Tissue slides were treated either with hematoxylin and eosin (H&E) staining for bronchus-associated lymphoid tissue, or with Grocott’s methenamine silver (GMS) staining for fungal morphology observation in vivo. Infected lungs, brains, and spleens were also isolated and homogenized (Ultra-Trra T8, IKA) in 3 ml cold 1× PBS buffer for 1 min for each type of organ. The tissue suspensions were serially diluted and plated onto YPD agar medium with ampicillin and chloramphenicol, and colonies were counted after 3 days of incubation at 30 °C.

Lung tissue total RNAs were extracted using Trizol reagents (Invitrogen) and purified with a RNeasy cleanup kit (Qiagen) following the manufacturer’s instructions. Purified RNAs were quantified using a Nanodrop instrument (Thermos Scientific). First-strand cDNAs were synthesized using a Superscript III cDNA synthesis kit (Invitrogen) following the instructions provided by the manufacturer. Expression levels of Cytokines and ACTIN were analyzed with the comparative C_T method using Brilliant SYBR green QPCR reagents (Takara) as described previously. Relative mRNA levels were determined by quantitative reverse transcription qRT-PCR. Gene expression relative to that of a naive sample was calculated by the ΔΔCT method.

In silico protein structure analysis

Structural modeling and ligand docking: The three-dimensional structures of Itr4 monomer and several dimers were predicted using AlphaFold2 in monomer and multimer modes, respectively [30]. AlphaFold2 was accessed through the COSMIC2 portal. The resulting PDB files were transferred to the Rutgers University Amarel high-performance compute cluster for analysis. The structures were assessed in Molecular Operating Environment (MOE) version 2024.06 and the structures with the fewest clashing residues were prepared in MOE. The Amber:EHT force field was used for all operations within MOE. For docking, the Itr4 monomer and the energy-minimized myo-inositol were analyzed within MOE. The residues within the transmembrane helices were saved as sets and labeled TM1-12 according to their location. A dummy atom was created and residues within 4.5 Å were selected and the ligand-receptor pair with the lowest energy was selected. The resulting pose was captured as an instance within a MOE storyboard, imaged, and the ligand interaction diagram was reported. This storyboard instance was the starting point for subsequent virtual mutations.

Digital Alanine scan: The protein builder tool was used within MOE to mutate Q388 and Q389 to alanine as monosubstituted and disubstituted models. The protein builder minimization tool was used to minimize the energy of the selected residues that were substituted. A molecular surface representing the local electrostatic environment was produced using receptor atoms within 4.5 Å of ligand atoms. The resulting mutant structure was saved as a new storyboard instance and the primary MOE viewer was reset to the original storyboard instance for a subsequent mutation.

Molecular dynamics simulations

MD simulations were performed for the wild-type Itr4/inositol complex and the Itr4Q388A/Q389A/inositol complex. The initial binding poses of inositol were obtained from molecular docking. The prepared protein–ligand complexes were embedded in a DPPC (325K) lipid bilayer using the Membrane Builder module in Desmond (D. E. Shaw Research, NY; Maestro-Desmond Interoperability Tools, Schrödinger, NY, 2023). The place of membrane was further corrected based on the TMHMM results. The membrane-protein systems were then solvated with the SPC water model, and an orthorhombic simulation box was constructed with a minimum buffer distance of 10 Å * 10 Å * 21 Å in x, y, z directions. For MD workflow, the simulation time was set as 200 ns with recording intervals was set as 20 ps resulting in 10,000 frames for further analysis. Other parameters were set as default. Trajectory visualization and structural analyses were performed using the Maestro graphical user interface (GUI). Protein and ligand root-mean-square deviations (RMSDs) were calculated throughout the simulation. Protein–ligand interaction frequencies were analyzed using the Simulation Interaction Diagram (SID) tool implemented in the Desmond module to evaluate the stability of inositol binding and the contribution of individual residues to ligand recognition.

S1 Fig. Phenotypic analysis of group II ITRs.

(A) Cultures were grown overnight in YPD and diluted to an optical density at 600 nm (OD600) of 2.0. Tenfold serial dilutions were made in dH2O, and 5 ul of each was plated. The plates were grown for 2 days at 30 and 37 °C. (B) Melanin detection assay. Color development by itr4∆, itr5∆, and itr6∆ single mutants and their triple mutants (itr4∆ itr5∆ itr6∆) and wild-type KN99α strains was evaluated after cultures were grown on Niger Seed plate for 48 h at 30 and 37 °C. (C) Cultures were grown overnight in YPD and diluted to an OD600 of 2.0. Serial dilutions and spotting assays were made, and the plates were grown for 2 days at 30 and 37 °C for different stress and inhibitor conditions. Wild-type and mutant strains are indicated on the left and the conditions at the top.

https://doi.org/10.1371/journal.pbio.3003709.s009

(TIF)

S2 Fig. Itr4 N-terminal tail is required for maintaining the Itr4 function.

(A) Expression of genes involved in inositol transport, glucose metabolism in the itr4Δ mutant and itr4Δ+ITR4^{Q388A, Q389A}. Relative transcript levels of the indicated genes were determined by quantitative qRT-PCR in wild-type, the itr4Δ mutant and itr4Δ+ITR4^{Q388A, Q389A} strains grown under 1% glucose + 1% inositol condition. Gene expression was normalized to the internal control and is presented relative to wild-type KN99a. Each dot represents an independent biological replicate, and bars indicate the mean ± SD. Statistical significance was determined using an unpaired two-tailed Student t test. “****”, P < 0.0001; “***”, P < 0.001; ns, not significant. (B) Detection of potential interactions between Itr4, Itr3B, Itr4^∆11-119 and Itr4^{Q388A, Q389A} in a yeast split-ubiquitin system. Two constructs, pCCW-Alg5 and pAI-Alg5, and pCCW and pDL2Nx serve as positive and negative control. Interaction was determined by the growth of yeast transformants on medium lacking histidine or adenine, and also by measuring β-galactosidase enzyme activities to verify the interaction. (C) Predicted Itr4 homodimer self-interactions. Depicts interactions of chain 1 N′ residues 97–115 with chain 2 C′ residues 629–630 was shown on left (L), and depicts interactions of chain 1 N′ residues 112–120 with chain 2 C′ residues 605–606 wa shown on right (R). The data underlying this figure can be found in the supplemental information file S2 Data.

https://doi.org/10.1371/journal.pbio.3003709.s010

(TIF)

S3 Fig. Itr4 protein structure modeling and predicted sugar binding sites.

(A) Itr4 protein structure model predicted by AlphaFold2. Green color represents the Itr4 protein ribbon. Red region represents the first 120 amino acids of the N-terminal tail located within the cytoplasm. Blue region represents intracellular loops in Itr4. (B) Schematic of mutation sites in Itr4 protein. The N-terminal region between residues 11 and 119 amino acids was deleted, while residues Q388, Q389 and F497 were replaced by alanine (A). (C) Results from the virtual alanine mutation scan from the protein builder tool in Molecular Operation Environment (MOE). Receptor surface maps within 4.5Å of the ligand, shown as line diagrams, represent the shapes of the binding pockets in the wild-type (WT) and mutated forms of the protein. Red lines represent a negatively charged environment, blue lines represent a positively charged environment, and white lines represent neutral environments. The Q388A mutation had a significantly larger impact on the size of the binding pocket shape and volume than the Q389A mutation. Ligand interaction diagrams represent the non-covalent interactions between the ligand and receptor. Potential energy calculations computed did not account for hydration energies and may not fully capture the reality of the living cell.

https://doi.org/10.1371/journal.pbio.3003709.s011

(TIF)

S4 Fig. Heterologous expression and uptake assays of ITR4 in Saccharomyces strains lacking inositol transporter.

(A) Localization of ITR4, ITR4^∆11-119, ITR4^{Q388A, Q389A} and ITR4^F497A in S. cerevisiae was determined by overexpressing of their GFP fusion protein in Sc itr1 itr2 mutant lacking two inositol transporters. (B) Inositol uptake analysis of Cryptococcus ITR4, ITR4^∆11-119, ITR4^{Q388A, Q389A} and ITR4^F497A expressed in a yeast heterologous system. Yeast cells were mixed with ³H-labeled inositol and incubated at 30 °C for 10 min. Error bars indicate standard deviations of data from at least three independent experiments. Sc, S. cerevisiae. The data underlying this figure can be found in the supplemental information file S2 Data.

https://doi.org/10.1371/journal.pbio.3003709.s012

(TIF)

S5 Fig. Western blot analysis to detect the GFP fusion protein expression of itr4 and its muant alleles.

(A) western blot analysis to examine the expression and stability of Itr4^Δ11–119, Itr4 and the Itr4^{Q388A, Q389A} GFP fusion proteins in S. cerevisiae with a GFP antibody. (B) western blot analysis to examine the expression and stability of Itr4^Δ11–119, Itr4^OE and the Itr4^{Q388A, Q389A} mCherry fusion proteins in C. neoformans with a mCherry antibody.

https://doi.org/10.1371/journal.pbio.3003709.s013

(TIF)

S6 Fig. Structural characterization of GXM polymers.

(A) SEC-MALS analysis of extracellular polysaccharide (EPS) from KN99a, itr4∆, and ITR4^Q388,389A strains in glucose growth conditions. (B) 1D [¹H] NMR analysis of EPS from all 3 strains. (C) Characterizing GXM motif expression and GXM O-acetylation.

https://doi.org/10.1371/journal.pbio.3003709.s014

(TIF)

S7 Fig. Itr4 is required for fungal virulence in in vivo murine models of systemic cryptococcosis.

(A) Expression profile of cytokine genes and genes involved in macrophage polarization was examined by qRT-PCR in mouse lung at day 7 after infection. Samples from naïve mice were analyzed as a control. Relative expression levels of indicated genes were calculated using the ΔΔCt method and normalized to Actin gene. (B and C) Histological examination of lung apex and parenchyma sections. The tissue sections were fixed and stained with H&E and GMS. Fungal cells appear dark brown or black with GMS staining. Yeast cells are highlighted with black arrow. Bar, 20 µm. The data underlying this figure can be found in the supplemental information file S2 Data.

https://doi.org/10.1371/journal.pbio.3003709.s015

(TIF)

S8 Fig. Itr4 s required for fungal dissemination to brain to cause brain infection in an in vivo murine infection model.

(A) End-point fungal burdens in the lungs, brains, spleens of animals infected with indicated strains in a murine intranasal inhalation model of systemic cryptococcosis. (B) End-point fungal burdens in the lungs, brains, spleens of animals infected with indicated strains in a murine intravenous injection infection model of systemic cryptococcosis. Each symbol represents one mouse. Bars represent the mean values ± standard errors of the means. “****”, P < 0.0001; “***”, P < 0.001 “**”, P < 0.01; “*”, P < 0.05 (determined by One-way ANOVA nonparametric test for multiple comparisons). (C and D) Histological examination of brain sections. The brain sections were fixed and stained with H&E and GMS. Fungal cells appear dark brown or black with GMS staining. Yeast cells are highlighted with black arrow. Bar, 20 µm. The data underlying this figure can be found in the supplemental information file S2 Data.

https://doi.org/10.1371/journal.pbio.3003709.s016

(TIF)