Survival in macrophages induces enhanced virulence in Cryptococcus

ABSTRACT Cryptococcus is a ubiquitous environmental fungus and frequent colonizer of human lungs. Colonization can lead to diverse outcomes, from clearance to long-term colonization to life-threatening meningoencephalitis. Regardless of the outcome, the process starts with an encounter with phagocytes. Using the zebrafish model of this infection, we have noted that cryptococcal cells first spend time inside macrophages before they become capable of pathogenic replication and dissemination. What “licensing” process takes place during this initial encounter, and how are licensed cryptococcal cells different? To address this, we isolated cryptococcal cells after phagocytosis by cultured macrophages and found these macrophage-experienced cells to be markedly more virulent in both zebrafish and mouse models. Despite producing a thick polysaccharide capsule, they were still subject to phagocytosis by macrophages in the zebrafish. Analysis of antigenic cell wall components in these licensed cells demonstrated that components of mannose and chitin are more available for staining than they are in culture-grown cells or cells with capsule production induced in vitro. Cryptococcus is capable of exiting or transferring between macrophages in vitro, raising the likelihood that this fungus alternates between intracellular and extracellular life during growth in the lungs. Our results raise the possibility that intracellular life has its advantages over time, and phagocytosis-induced alteration in mannose and chitin exposure is one way that makes subsequent rounds of phagocytosis more beneficial to the fungus. IMPORTANCE Cryptococcosis begins in the lungs and can ultimately travel through the bloodstream to cause devastating infection in the central nervous system. In the zebrafish model, small amounts of cryptococcus inoculated into the bloodstream are initially phagocytosed and become far more capable of dissemination after they exit macrophages. Similarly, survival in the mouse lung produces cryptococcal cell types with enhanced dissemination. In this study, we have evaluated how phagocytosis changes the properties of Cryptococcus during pathogenesis. Macrophage-experienced cells (MECs) become “licensed” for enhanced virulence. They out-disseminate culture-grown cells in the fish and out-compete non-MECs in the mouse lung. Analysis of their cell surface demonstrates that MECs have increased availability of cell wall components mannose and chitin substances involved in provoking phagocytosis. These findings suggest how Cryptococcus might tune its cell surface to induce but survive repeated phagocytosis during early pathogenesis in the lung.

difficult to detect before it is severe and hard to treat.A better understanding of the early stages of pathogenesis is needed to improve detection and prophylactic strategies.
Cryptococcal pathogenesis is thought to begin when spores or small/desiccated yeast are inhaled into the lower reaches of the lungs (2).Here they encounter alveolar macrophages, which can sometimes clear the infection.Otherwise, Cn may take up long-term, asymptomatic residence in the lung, presumably inside granulomas.If Cn is not cleared, it appears that the extent to which the host can mount a granulomatous response (via adaptive immune function) dictates the ability to control fungal growth (3).Therefore, a susceptible host may develop fulminant disease either upon initial infection or after loss of control over a latent one.In either scenario, Cn growth in the presence of macrophages is required.
In the mouse model of cryptococcosis, a large inoculum of Cn delivered intravenously is capable of causing rapid CNS infection, with yeast cells entering the brain directly from the bloodstream (4)(5)(6).When mice are infected via the airway (as it is thought to occur in humans), Cn takes hold and grows in the lungs for several days prior to introducing disseminating particles (bare yeast, yeast inside macrophages, or both) into the blood (7).These disseminating particles are more efficient at brain dissemination than cultured Cn since an almost undetectable blood burden (8) seeds the CNS effectively (7).Thus, time spent in the lung produces a Cn phenotype better suited to CNS dissemination than that of culture-grown Cn.Similarly, in the zebrafish model, we have observed that after IV inoculation, yeast cells are universally phagocytosed and spend time inside macrophages before returning to the bloodstream with enhanced capacity for virulence (9).We, therefore, hypothesize that phagocytosis and intracellular survival are not just barriers to cryptococcal pathogenesis but necessary steps that enhance the ability of Cn to disseminate in the blood.
Cn encounters multiple macrophage types during its stay in the lung.Initial phagocytosis is most likely by alveolar macrophages, but other resident macrophages, and recruited monocyte-derived macrophages are also present.Based on transcription profiling, alveolar macrophages remain a single subtype, while resident interstitial subsets fall into two types, CD14 + and CD14 - (10).Of note, there is evidence that alveolar macrophages are more likely to kill Cn than either of the interstitial types (11).The fate of Cn in macrophages also depends on activation and polarization.Most host receptormediated interactions with Cn tend to induce TH1 and/or TH17 responses, which are host protective against fungi, although there is evidence that Dectin-2 can induce an anti-inflammatory M2 phenotype in macrophages (12).
Multiple elements of the cryptococcal cell wall can serve as microbe-associated molecular patterns (MAMPs), molecular triggers of phagocytosis that can induce both pro-and anti-inflammatory responses depending on their context and the receptors they bind.Cryptococcal MAMPs include chitin, β-glucan, and mannose, among others.The most prominent of these is β-glucan, which is detected by macrophages via Dectin-1, complement receptor 3 (CR3), and ephrin A2 (EphA2) and tends to induce inflammation (13).In Candida, β-glucan is thought to be "masked" from these receptors by display of mannose, and strains unable to do so have decreased virulence (14,15).Cryptococcus also expresses mannose at the cell wall where it could have a similar masking role (16).On the other hand, mannose itself is detected by phagocytes by way of Dectin-2, DC-SIGN, mannose receptor, and others (17), which can induce both inflammatory and anti-inflammatory responses.All of these cell wall substances are thought to be "masked" by the polysaccharide capsule, which impedes phagocytosis by multiple mechanisms (18).

RESULTS
Macrophage-experienced Cryptococcus cells exhibit enhanced replication in the zebrafish larva and the mouse.In prior experiments using the zebrafish model of cryptococcosis, we observed that Cn inoculated into the bloodstream were completely phagocytosed by macrophages and neutrophils within ~2 hours (9).After a variable amount of time (1-3 days), surviving yeast cells could be seen circulating in the bloodstream in small numbers.This "secondary fungemia" (since initial inoculation produced a primary fungemia) set the stage for dissemination to specific organs (including the brain) and sometimes overwhelming fungemia.The cells of secondary fungemia had essentially become more virulent.We hypothesized that time inside phagocytes had altered the virulence potential of these cells, making them capable of new interactions with the host, including dissemination and enhanced growth.To test this, we generated macrophageexperienced Cn (MECs) in vitro.We co-cultured murine J774 cells with an EGFP-express ing KN99 strain (JMD163) (19) for 3 hours to allow adhesion and phagocytosis, then washed off any non-adherent Cn.The infected J774 cells were then incubated overnight and finally lysed so that the Cn cells could be isolated.These MECs were then inoculated into the vasculature of zebrafish larvae at our usual inoculum of 30-70 fluorescent cells per larva.MECs showed more rapid and extensive replication in the larva than either our usual YPD-cultured Cn or tissue culture-conditioned cells (control yeast treated exactly as MECs except with no J774 cells present) (Fig. 1A).Host mortality was also increased, as survival of MEC-infected larvae dropped off at 4-5 dpi (Fig. 1A and data not shown).To determine whether this was an artifact of the J774 cell line, we repeated this experiment using human THP-1 macrophage-like cells, with the same result (Fig. S1).
To determine whether pre-phagocytosis had the same effect in a mammalian infection, we performed a competition assay using EGFP-expressing MECs and TC-experi enced Cn expressing mRuby3 (20, 21) (JMD225, produced from the same parental strain as JMD163).These were inoculated together into A/J mice via the intranasal route.A total mixed inoculum of 1 × 10 4 CFU was instilled and lungs were harvested at days 8 and 30 for fungal burden measurement.At each timepoint, the proportion of EGFP-expressing cells (MECs) increased over the initial inoculum (Fig. 1B), consistent with enhanced replication by MECs in the lung.Therefore, macrophage-experienced Cn demonstrated enhanced replication inside both zebrafish and murine hosts.

MECs have altered cell diameter and capsule production
We reasoned that the change in virulence seen in MECs would involve the most conspicuous cryptococcal virulence factor, the polysaccharide capsule.We compared cell diameter and capsule thickness of YPD-grown, TC-experienced, and J774-experienced cells using India ink staining, with results shown in Fig. 2A and B. While the cell bodies of YPD-grown yeast were modestly smaller than in the other two conditions, there were wide variations in capsule production in tissue culture-conditioned and macrophageexperienced yeast.Not unsurprisingly, MECs made a very thick capsule.

MECs in the zebrafish undergo phagocytosis despite their encapsulation
It is known that capsule production is a primary virulence factor that prevents phagocy tosis of Cn.It would follow that the enhanced growth of MECs in vivo could be due simply to extracellular survival and life.To test this hypothesis, we infected Tg(mpeg:egfp) larvae with our red fluorescent JMD225 strain and fixed these larvae at 2.5 hours after infection to quantify rates of phagocytosis for YPD-grown, TC-experienced, and macrophage-experienced Cn.While MECs were phagocytosed significantly less than controls (Fig. 2C), phagocytosis was not abolished.Even cells with relatively large capsules were found completely and partially engulfed by macrophages (Fig. 2D) (more detailed views of the scene in Fig. 2D are shown in Fig. S2 and Video S1).

MECs have increased mannose and chitin exposure despite their thick capsule
One mechanism by which the capsule is thought to prevent phagocytosis is by masking MAMPs on the cell wall, including mannose and chitin (16,22).To explain the phagocy tosis of even heavily encapsulated MECs, we examined the relative exposure of these elements using fluorescent-labeled concanavalin A (ConA-which binds mannose) and wheat germ agglutinin (WGA-which binds oligomeric chitin).We analyzed micrographs of stained yeast grown in YPD along with TC-experienced and J774-experienced yeast, measuring relative fluorescence.Exposure of both was greater in MECs than in TC-expe rienced Cn and Cn grown in YPD (Fig. 3A and B).Furthermore, in MECs, the ConA signal was at the periphery of the cell wall and not just the bud scars, as it was in the other conditions (Fig. 3C).A similar peripheral pattern was seen with WGA staining (Fig. 3D).This finding was reminiscent of those by Denham et al. (23), who reported similar staining in a subset of small, dissemination-prone cryptococcal cells isolated after several days in the mouse lung.The authors proposed that extra exposure of ConA was due to the relatively thin capsules of this subset.MECs, in contrast, have variable but overall rather thick capsules (Fig. 2B).If capsules were the primary means of masking these substances, we would expect staining intensity to vary inversely with capsule thickness.This turned out not to be the case (Fig. 4A and B).Finally, to determine whether these findings were an artifact of our fixation and microscopy methods, we analyzed the same cryptococcal strain incubated overnight in RPMI-a frequently used method for inducing Widefield images were taken with a 20×, 0.85NA objective.
capsules.Induction of the capsule was as expected in RPMI (Fig. 4C) but ConA and WGA staining were markedly reduced compared to MECs (Fig. 4D and E).

MECs may show enhanced brain dissemination but TC experience accounts for most of this effect
In our prior investigations of brain dissemination in zebrafish, we have used cryptococcal cells grown overnight in YPD (19).In observing zebrafish infections with MECs, we noticed that brain lesions were more frequent and grew subjectively faster than in experiments with YPD-grown cells.Subjectively, TC-experienced Cn growth in the brain was intermediate (Fig. 5A).To quantify brain dissemination in this model, we have previously reported 43 parenchymal lesions in a set of 77 fish observed over 4 days (0.56 lesions per larva) (19).TC-experienced yeast and MECs averaged more brain lesions overall (Fig. 5B), although this difference did not meet 95% significance.Contrary to our subjective impression (Fig. 1A), we found that the brain lesion rate was indistinguishable between TC-experienced yeast and MECs (Fig. 5B).To quantify brain dissemination of TC-experienced Cn and MECs in greater detail, we infected Tg6(kdrl:mCherry) zebrafish (in which endothelial cytoplasm is red fluorescent) and followed Cn arrival and fates in the brain.In all, we followed 78 lesions in 44 larvae infected with TC-experienced yeast and 77 lesions in 43 larvae infected with MECs.Although there were trends toward better brain dissemination in MECs, no statistical difference was found in terms of brain lesion clearance (Fig. 5C) or day-to-day replication in brain lesions (Fig. 5D and E).Thus, while macrophage experience enhances the overall virulence of Cryptococcus, its likely advantage in brain dissemination over YPD-grown cells can be induced by TC exposure alone.MECs out-replicate TC-experienced cells overall likely due to enhanced growth in the caudal hematopoietic tissue and the general circulation.

DISCUSSION
The Cn-macrophage interaction is a key step in cryptococcal pathogenesis, and here we show that in vitro experience in a macrophage enhances the yeast's ability to survive and proliferate in both the zebrafish and murine models.Evasion of host immunity is a central theme of Cn pathogenesis (24,25), but initial survival in macrophages has implications for subsequent survival and virulence.Our data suggest that the initial interaction between yeast and macrophage licenses the yeast cells to adopt a hyperviru lent phenotype.This licensed phenotype includes increased cell body and capsule size but also a particular alteration in MAMP exposure.Although licensed cells are heavily encapsulated, they are still phagocytosed readily in zebrafish.Using this model, we have previously reported yeast escaping from macrophages in early pathogenesis, only to be phagocytosed again (9).In the mouse lung, Cn spends 1 to 2 weeks before it is capable of bloodborne dissemination.Knowing that Cn is capable of lytic and non-lytic exit from macrophages and even direct transfer in between (26-29), our findings support the notion that individual yeast cells are phagocytosed and released multiple times during this period (30).Cn could be using selective exposure of MAMPs to sustain a sequence of alternating intracellular and extracellular growth periods in the lung.
How might such a sequence benefit cryptococcal growth?Two possibilities are as follows: (a) by taking advantage of more permissive macrophage types which arrive later in infection and (b) by altering the cell surface to manipulate the nature of subsequent intracellular visits.As noted above, alveolar macrophages, likely the first encountered in the lung by Cn, are shown to be more fungicidal than the interstitial macrophages which might perform subsequent phagocytosis (11).Surviving and escaping alveolar macro phages might also provide Cn the opportunity to take up residence in a more permissive cell.Such a scheme for reaching permissive macrophages has been shown in mycobacte rial infection (31), and a similar interaction has been suggested for Cn as well (32).As for manipulating the outcome of subsequent phagocytosis, specific host receptors that initiate recognition and phagocytosis of a microbe have distinct implications for the fate of the ingested organism and the inflammatory state produced.DC-SIGN, for example, one receptor that binds mannose, has a mixed reputation for inducing pro-or antiinflammatory signaling after phagocytosis (33)(34)(35), in particular leading to expression of IL-10, an inflammation-dampening signal long associated with Cn pathogenesis.Also as noted above Dectin-2, another receptor that recognizes mannose, can induce T H 2 immunity in certain circumstances (12).Therefore, by increasing mannose exposure, initial phagocytosis could well be inducing changes that alter the outcome of subse quent macrophage encounters.
A recent mouse study has revealed that average cryptococcal cell size gradually decreases in the 10-14 days between lung infection and bloodborne dissemination (23).The authors identified in the lung a distinct population of the smallest Cn cells most capable of dissemination to the brain.What process drives this differentiation?Our data suggest that a single sojourn inside of a macrophage is not enough to account for it.We show that MECs, with their single night of macrophage experience, are better at brain dissemination than YPD-grown Cn but their advantage over TC-experienced cells is minimal at best.The best characterization of the MECc virulence advantage is sheer in vivo replication, whether in the bloodstream and brain of the fish or the lungs of the mouse.
Phagocytosis is a vital step in cryptococcal pathogenesis.Any yeast that ultimately reach the CNS to cause severe disease have spent time inside a macrophage, or are daughters of cells that have.Intuitively, one might assume that survival in macrophages would induce a phenotype that avoids future phagocytosis, for the sake of enhanced extracellular replication.Our data reinforce the theory that Cryptococcus emerges from macrophages primed for growth in vivo but also permissive to re-phagocytosis, providing a molecular mechanism for the latter.While the production of the capsule is known to inhibit phagocytosis, this inhibition may be balanced by the exposure of select MAMPs, providing another mechanism by which Cn may exchange a stressful intracellular environment for a more permissive one (32).These findings add detail and support the long-suggested hypothesis that macrophages are both an opposition to cryptococcal pathogenesis and a necessary collaborator (30,(36)(37)(38).

Zebrafish care and maintenance
Adult zebrafish were kept under a light/dark cycle of 14 hours and 10 hours, respectively.Larval zebrafish were incubated at 28.5°C in E3 buffer (39), switching larvae to E3-MB containing 0.2 nM 1-phenyl-2-thiourea (PTU) (Sigma-Aldrich) at 18-24 hours post-fertili zation to inhibit pigment formation.All larvae were manually dechorionated between 24 and 30 hpf.Prior to microinjection or imaging, larvae were anesthetized in E3-MB containing 0.2 mg/mL Tricaine (ethyl 3-aminobenzoate; Sigma-Aldrich).For prolonged time-lapse imaging, larvae were mounted in 1% low-melting point agarose, 0.2 mg/mL Tricaine, and 0.2 nM PTU (final concentrations) on a coverglass-bottom dish.All adult and larval zebrafish procedures were in full compliance with NIH guidelines and approved by the University of Iowa Institutional Animal Care and Use Committee (Protocol #0102075-002).

Transgenic zebrafish lines
Tg(kdrl:mCherry-CAAX) (40) is designated y171Tg at the Zebrafish International Resource Center (ZFIN) and was kindly provided by Daniel Castranova.Tg(mpeg:egfp) (41) is designated gl22Tg and was kindly provided by Dr. Anna Huttenlocher.Transgenic lines are maintained by outcrossing to AB wildtype fish every other generation.ABs are obtained from ZFIN regularly to maintain hybrid vigor.

Cryptococcal strains and growth conditions
KN99α was kindly provided by Dr. Kirsten Nielsen.Cultures were handled using standard techniques and media as described previously (42,43).

J774A.1
The mouse macrophage cell line J774A.1 (ATCC TIB-67) was kindly provided by Dr. Melanie Wellington.A confluent T-75 flask of cells exhibiting normal morphology and cell adhesion was split 1:2 in 15 mL Dulbecco's modified Eagle medium supplemented with 10% fetal bovine serum (FBS), 1% Glutamax, and 1% penicillin-streptomycin and allowed to grow overnight.Cryptococcus strains were cultured in 50 mL liquid YPD overnight.The Cryptococcus cells were washed three times in sterile Dulbecco's phosphate-buffered saline (DPBS). 2 × 10 9 Cryptococcus cells/mL were opsonized with the 18B7 monoclo nal antibody (provided by Dr. Damian Krysan) for 30 minutes prior to co-incubation with macrophages.J774A.1 medium was replaced with 15 mL fresh medium and 5 × 10 8 Cryptococcus cells (MOI 20), followed by a 3-hour incubation at 37°C with 5% CO 2 .Simultaneously, 5 × 10 8 Cryptococcus cells were added to 15 mL J774A.1 culture medium in a T-75 culture flask and incubated at 37°C with 5% CO 2 as a control.The macrophage/Cryptococcus co-culture was then washed three times with warm DPBS to remove non-adherent yeast, and then replenished with 15 mL fresh J774A.1 medium and allowed to incubate overnight at 37°C with 5% CO 2 .To lyse the macrophages, 5 mL 10% Triton X-100 in DPBS (final concentration of 2.5%) was added to the co-culture and the control flask then incubated for 20 minutes at room temperature while rocking.Cryptococcus cells were then centrifuged and washed twice with sterile DPBS before use.

THP-1
The human monocyte cell line THP-1 (ATCC TIB-202) was also kindly provided by Dr. Melanie Wellington.Cells were harvested from confluent flasks, then seeded in a T-75 flask at a density of 3 × 10 5 cells/mL in 15 mL RPMI supplemented with 10% FBS, 1% Glutamax, 1% penicillin streptomycin, 10 mM 4-(2-hydroxyethyl)-1-piperazineethanesul fonic acid (HEPES), and 1 mM sodium pyruvate.Cells were primed with 100 ng/mL Phorbol 12-myristate 13-acetate (PMA) and allowed to incubate for 2 days at 37°C with 5% CO 2 (46).Media was then replaced with culture medium without PMA and cells were incubated for an additional day at 37°C with 5% CO 2 .Cryptococcus strains were cultured in 5 mL liquid YPD overnight.The Cryptococcus cells were washed three times in sterile DPBS. 1 × 10 8 Cryptococcus cells/mL were opsonized in pooled human serum (MP Biomedicals) for 30 minutes and washed three times with DPBS prior to co-incubation with macrophages.THP-1 medium was replaced with 15 mL fresh medium and 9 × 10 6 Cryptococcus cells (MOI 20), followed by a 3-hour incubation at 37°C with 5% CO 2 .Simultaneously, 9 × 10 6 Cryptococcus cells were added to 15 mL THP-1 culture medium in a T-75 culture flask and incubated at 37°C with 5% CO 2 as a control.The co-culture was then washed three times with warm DPBS to remove non-adherent Cryptococcus cells, then replenished with 15 mL fresh THP-1 medium and allowed to incubate overnight at 37°C with 5% CO 2 .To lyse the macrophages, 5 mL 10% Triton X-100 in DPBS (final concentration of 2.5%) was added to the co-culture and the control flask then incuba ted for 20 minutes at room temperature while rocking.Cryptococcus cells were then centrifuged and washed twice with sterile DPBS before use.

Cryptococcus lectin staining
Macrophage-experienced, culture medium-experienced, and overnight YPD-cultured Cryptococcus cells were washed twice in sterile DPBS and resuspended in DPBS at a concentration of 1 × 10 8 cells/mL.Cryptococcus cells were then fixed in 2% paraformal dehyde in DPBS for 12 minutes at room temperature and washed three times with sterile DPBS.Cells were stained with concanavalin A (Con A) tagged with Alexa Fluor 633 (Fisher #C21402) and wheat-germ agglutinin (WGA) tagged with Alexa Fluor 555 (Fisher #W32464) as previously described (23) in addition to calcofluor white.Briefly, cells were incubated in WGA for minutes in the dark at room temperature while rotating.Following incubation, 50 ug/mL Con A was added to the cells and incubated for an additional 3 minutes.Cells were then centrifuged and washed twice in DPBS, then combined 1:1 with India ink for imaging.

Zebrafish infection by microinjection
Single colonies from 4°C stocks were inoculated into YPD medium and incubated shaking overnight at 30°C.In addition, Cryptococcus cells were macrophage experienced or culture medium experienced as described above.YPD cultures were washed twice with sterile DPBS before use.All cultures were diluted in DPBS containing 10% glycerol and 2% PVP-40 (polyvinylpyrrolidone, Sigma Aldrich) (47) to an OD 600 of 5.0 in a 1:10 dilution of phenol red.After manual dechorionation of embryos at ~28 hpf, IV inocula tions were performed as previously described (48), with the alteration that larvae were positioned on a 3% agarose plate formed with holding grooves as described in reference (39).Initial inoculum was documented by direct microscopic observation and only larvae with initial inoculum between ~30 and 70 fluorescent yeast cells were used.

Live zebrafish microscopy
Confocal imaging was performed on a Zeiss Axio Observer.Z1/7 body equipped with a Zeiss Airyscan detector.All channels were collected using Airyscan Multiplex settings with default processing.Cryptococcal India ink and lectin staining images (Fig. 3; Fig. S3) were captured using a Zeiss LD C-Apochromat 60×/1.1 Oil Korr UV VIS IR objective.Correlative DIC images in Fig. S2 were captured shortly after the confocal collection in widefield mode using a Zeiss 40×/1.1 Water corrected Plan Apochromat objective and re-scaled manually to overlay with confocal images taken with the same objective.All other images were captured using a Zeiss Plan-Apochromat 20×/0.8objective.Live imaging was performed with larvae anesthetized in tricaine as previously described (48) and simply resting on the bottom of a glass-bottom dish or immobilized in 1% low melt agarose.For the collection of large numbers of events, a combination of widefield and confocal imaging was used to create scout and detail images for later analysis.Widefield imaging was performed using the same Zeiss optical setup, with image capture using a Hamamatsu Flash4.0V3 sCMOS camera.Widefield fluorescence excitation was generated with a Colibri 7 type RGB-UV fluorescence light source.The filter set was Zeiss set 90 LED.

Fixed zebrafish microscopy
Larvae were inoculated with cryptococcal cells prepared as described elsewhere and euthanized in 250 mg/L Tricaine-S at 2.5hpi.Larvae were then fixed with 4% paraformal dehyde overnight at 4°C.Prior to imaging, they were washed repeatedly in 1× PBS and stored at 4°C until viewing.Images for analysis of phagocytosis were collected in widefield mode using a Zeiss Plan-Apochromat 20×/0.8objective.The confocal image in Fig. 2D and Video S1 was collected using Airyscan multiplex settings and default Airyscan processing.

Image processing
Images were processed with Zeiss Zen software applying default Airyscan processing followed by z-stack alignment if needed.Single slices, orthogonal projections, and 3D renderings were produced in Zen software and exported as TIF files for assembly using Adobe Photoshop and Illustrator.3D renderings were done with default maximum settings in Zeiss Zen.

Murine virulence studies
Cryptococcus cells were incubated with J774A.1 macrophages or culture medium and collected as described above.To differentiate cells from each condition, GFP-expressing Cryptococcus cells (JMD163) were incubated with macrophages and mRuby3-expressing Cryptococcus cells (JMD225) were incubated in a culture medium.Following collection, both cultures were diluted in DPBS containing 10% glycerol and 2% PVP-40 (polyvinyl pyrrolidone, Sigma Aldrich) (47).8-to 10-week-old A/J female mice (JAX) were inoculated with a 1:1 mix of each Cryptococcus strain as described in reference (49).Briefly, mice were inoculated with 5 × 10 3 CFU of each strain via intranasal route (1 × 10 4 CFU total).Each group contained three mice.Brains and lungs were collected 8 days post-inocula tion and homogenized in sterile PBS.Homogenates were diluted in a 10-fold dilution series, plated on YPD, and incubated at 30°C for 48 h.Colonies were counted using a Zeiss SteREO V8 microscope with Zeiss achromat S 0.63× Reo objective and Lumencor SOLA light engine, to differentiate GFP-and mRuby3-expressing cells.

FIG 2
FIG 2 Tissue culture conditions and macrophage exposure induce increased cell body size and capsule production, with differing effects on phagocytosis.(A) Cell body diameters for YPD-grown, TC-experi enced, and J774-experienced (MEC) Cn.Diameters measured via widefield microscopy.Data from three biological replicates of at least 20 cells per condition per replicate.Statistical comparisons represent the results of two-tailed unpaired t-tests.(B) Capsule diameters for the same cells analyzed in panel A. Statistical comparisons represent results of two-tailed unpaired t-tests.(C) The proportion of fluorescent yeast cells inside EGFP + cells (macrophages) at 2.5 hpi.Statistical comparisons represent results of one-way ANOVA with multiple comparisons.(D) EGFP + cells (mpeg + macrophages) phagocytosing J774-experienced Cn at ~2.5 hpi.Yeast 1 and 2 are fully engulfed, while #3 is partially engulfed.Live confocal image taken with 40×, 1.1NA water objective.3D rendering shown.Grid lines are 20 µm apart.Animation of this data in Video S1.

FIG 3
FIG 3 Yeast were incubated overnight in the indicated conditions, then isolated and fixed for microscopy at 63×/1.4NA.MECs have increased exposure to mannose and chitin despite the production of a thick capsule.Concanavalin (A) and wheat germ agglutinin (B) mean fluorescence intensities for Cn cultured in YPD, TC-experienced, or J774-experienced.Statistical comparisons represent the results of two-tailed, unpaired t-tests.(C) India ink and fluorescence images of YPD-cultured, TC-experienced, and J774-experienced Cn, three examples each.Staining in the first two conditions is punctate, at bud scars, or absent, while staining of J774-experienced cells is both punctate and circumferential.ConA is tagged with Alexa Fluor 633 (Fisher #C21402).Widefield images were taken with a 20×, 0.85NA objective.Exposures and intensity adjustments are uniform for all panels.(D) Example of ConA and WGA staining that overlaps on a different J774-experienced cell.Note intense staining at the bud scar with overlapping circumferential staining.WGA is tagged with Alexa Fluor 555 (Fisher #W32464).

FIG 4
FIG 4 Yeast were incubated overnight in the indicated conditions, then isolated and fixed for staining and subsequent microscopy.Exposure of ConA and WGA on macrophage-experienced Cn.ConA (A) or WGA (B) mean fluorescence intensity (y-axis) plotted versus capsule thickness (x-axis) in J774-experienced Cn.We find no linear correlation between PAMP exposure and capsule thickness, demonstrating that the capsule itself is not a proxy for the degree of masking.R 2 plotted via simple linear regression.C-E.Capsule thickness (C) is similar in J774-experienced Cn and Cn with capsule induced with RPMI but exposure of ConA (D) and WGA (E) is less.J774 media and MEC data are the same as in Fig. 2B, re-displayed in this context for clarity.Statistical comparisons represent the results of two-tailed unpaired t-tests.

FIG 5
FIG 5 J774-experienced Cn are more effective than YPD grown at colonizing the CNS in zebrafish but the difference is mostly accounted for by tissue culture conditions.(A) Yeast (green) outside the CNS vasculature (magenta) at 1 and 2 dpi after infection with YPD-grown Cn (left), TC-experienced Cn (middle) and J774-experienced Cn (right).Growth is subjectively greater in TC-experienced and J774-experienced Cn.Airyscan confocal images were collected with 40×, 1.1NA water objective.(B) Comparison of brain lesions per fish between historical data with YPD-grown Cn (ref ), TC-experienced, and J774-experienced Cn.There is a strong trend toward fewer brain lesions per fish with YPD-grown Cn but this does not reach statistical significance with the data available.Statistical comparison for this and all other panels of this figure represents results of Mann-Whitney tests of log-transformed ratios.(C) Percent of brain lesions cleared between 1 and 3 dpi, between TC-experienced and J774-experienced Cn.The trend is toward more clearance of TC-experienced cells, but not statistically significant.(D and E) Fold change of number of yeast per brain lesion in lesions not cleared from 1 to 2 dpi (D) and 2 to 3 dpi (E).