In this study, we utilized the C. auris CAU-09 (South Asian, clade I) and C. albicans SC5314 strain (25). For inoculum preparation, both the yeasts were cultured overnight in Yeast Extract Peptone Dextrose (YPD) broth at 30°C with continuous shaking at 200 rpm. Following incubation, the yeast cells were washed three times with 1x phosphate-buffered saline (PBS, Cat: 10010023, Thermo Fisher Scientific). The blastoconidia were then counted using a hemocytometer and adjusted to the desired cell density (25).

We used outbred ICR CD-1 male mice in this study, aged 4 to 6 weeks and weighing 25–30 grams. The mice were purchased from Envigo Harlan in Seattle, WA, USA. To induce immunosuppression, the mice received intraperitoneal (I.P.) injections of 200 mg cyclophosphamide and subcutaneous (S.C.) injections of 250 mg cortisone 21-acetate (Cat: C3130, Millipore Sigma) per kilogram (kg) of body weight, administered two days before infection. To prevent bacterial superinfection, the antibiotic, enrofloxacin (Baytril^®, Elanco, Germany) was added to the drinking water at 50 µg/ml, starting the same day and continuing until day 10 post-infection (25–27).

The immunosuppressed mice were then intravenously (I.V.) infected with a lethal dose of 5 × 10⁷ (for survival studies) or a sublethal dose of 1 × 10⁷ (for fungal burden studies) of C. auris (CAU-09) cells/mouse. For C. albicans, immunosuppressed mice were infected with a lethal I.V. dose of 5 × 10⁴ yeast cells/mouse. The inoculum size for each Candida species was determined based on previously published research and preliminary experiments to achieve comparable disease severity (25, 27). Because C. albicans is lethal at an inoculum ~2 logs lower than that of C. auris, even in immunocompetent mice, its inoculum was adjusted to produce a survival curve similar to that of untreated mice infected with C. auris, ensuring clinical relevance and consistency across models (25, 28).

One day post-infection, the mice were randomly assigned to the following non-blinded treatment groups: 0 (PBS diluent or placebo, n=10 + 9 mice, from two experiments), 0.2 (n=10 mice), 2.0 µg/mouse GM-CSF (Cat: Z03300, GenScript) (n=10 + 9 mice, from two experiments) or 0.5 mg/kg/day of micafungin (n=10 mice) or 2.0 µg/mouse GM-CSF + 0.5 mg/kg/day of micafungin (n=9 mice). Uninfected immunosuppressed mice served as a control for any unintended mortality caused by opportunistic bacterial infections. GM-CSF was administered I.P. once daily from day 1 through day 4. A subtherapeutic daily dose of micafungin was initiated on day 1 post-infection and continued until day 7. The mice were monitored for survival over 21 days post-infection (25, 27, 29).

For fungal burden, mice were euthanized on day 4 post-infection to collect kidneys, hearts, and brains. The organs were weighed, homogenized, and quantitatively cultured using 10-fold serial dilutions on YPD plates. The plates were incubated at 37 °C for 48 hours before colony-forming units (CFUs) per gram of tissue were enumerated (25, 27).

For the histopathological examination, kidneys and hearts from representative mice were harvested on days 4 and 7 post-infection. The organs were fixed in 10% zinc-buffered formalin, embedded in paraffin, sectioned, and stained with Hematoxylin and Eosin (25–27, 29).

Immunocompetent ICR CD-1 mice were injected with 2.0 µg/mouse I.P. of GM-CSF or PBS once daily for up to 4 days. Post euthanization on day 4, peritoneal exudates and bone marrow cells were harvested from each mouse. Bone marrow neutrophils were purified using the MojoSort Mouse Neutrophil Isolation Kit (Cat: 480058, BioLegend) according to the manufacturer’s instructions. The purity of neutrophils was assessed using flow cytometry as described below. Bone marrow neutrophils and cells from peritoneal exudates were then evaluated using the assays described below.

Approximately 1x10⁶ bone marrow neutrophils and peritoneal exudates were plated on 24-well plates in complete RPMI-1640 media (Cat: R8758, Millipore Sigma) supplemented with 10% FBS (Cat: A5670801, Thermo Fisher Scientific) and 1% Pen-Strep antibiotic cocktail (Cat: 15140122, Thermo Fisher Scientific). The cells were allowed to settle and adhere for two hours before infecting them with CAU-09 (MOI 1) for four hours. To determine the percentage killing, CAU-09 was grown for four hours in complete RPMI media without host cells. Host neutrophils and peritoneal exudate cells were lysed using 0.05% Triton X-100. We enumerated fungal cells after incubation with and without host cells by plating the fungus on YPD agar plates and counting CFU the following day (30). The percentage killing of C. auris was calculated and compared.

Approximately 2.5 x 10⁵ RAW264.7 macrophages (ATCC) were plated on a 24-well plate in complete RPMI media and treated with 20 ng/ml GM-CSF overnight. The following day, the macrophages were infected with CAU-09 (MOI 1) for 4 hours. Percentage killing was determined and compared as described above.

Approximately 1 × 10⁶ bone marrow neutrophils were infected with CAU-09 (MOI 1) for four hours. Post-infection, the neutrophils were centrifuged at 500 g for 5 minutes, and the supernatant was collected. Neutrophil elastase activity was determined using a mouse neutrophil elastase ELISA kit (Cat: ab252356, Abcam) according to the manufacturer’s instructions.

Cytotoxicity of RAW264.7 cells, bone marrow neutrophils, and cells from peritoneal exudates upon infection with CAU-09 (MOI 1, 4 hours) was assessed by measuring lactate dehydrogenase in the cell culture supernatant using the CytoTox 96 Non-Radioactive Cytotoxicity Assay kit (Cat: G1780, Promega), following the manufacturer’s instructions.

Mouse spleens and kidneys from infected and GM-CSF-treated mice were collected on days 4 and 7 and processed individually by homogenization through a 100 μm cell strainer. Red blood cells (RBCs) were lysed using 1x RBC lysis buffer (Cat: SC-296258, Santa Cruz Biotech, Dallas) and filtered through 100 μm sterile filters (25). Splenocytes and Kidney cells were stained with antibodies against CD3, CD4, CD19, F4/80, CD11b, and Ly6G. To determine the purity of neutrophils and peritoneal macrophages, the cells were stained using Ly6G, CD11b, and F4/80. The details of the antibodies used are provided in Supplementary Table S4. The stained cells were acquired using a BD LSR II flow cytometer, and the data were analyzed with FlowJo V10.

An overnight culture of CAU-09 was washed twice with PBS, then manually counted on a hemocytometer to adjust the MOI to 1. Fungal cells were stained with 10 μM CellTrace CFSE Cell Proliferation Kit (Cat: C34554, Thermo Fisher Scientific) for 45 minutes at 37 °C. Post-staining, the fungal cells were washed twice with PBS to remove any additional dye before infecting host cells (RAW264.7 macrophages (pre-treated overnight with 20 ng/ml GM-CSF), bone marrow neutrophils, and cells from peritoneal exudates) for 1.5 hours. The host cells are washed twice with PBS to remove any planktonic fungus, then lifted using a lift buffer (PBS containing 2% FBS, 0.5 mM EDTA, and 0.1% sodium azide). Fungal uptake by host cells was then assessed using flow cytometry (31).

An overnight culture of CAU-09 was washed twice with PBS and then manually counted to adjust the MOI to 1. RAW264.7 macrophages were pre-treated with 20 ng/ml GM-CSF overnight. Host cells (RAW264.7 macrophages, bone marrow neutrophils, and cells from peritoneal exudates) were pre-treated with 10 μM CM-H₂DCFDA (Cat: C6827, Thermo Fisher Scientific) for 30 minutes at 37 °C. Post-staining, the host cells were washed twice with PBS to remove excess dye, then infected with CAU-09 for 1.5 hours. The cells are washed twice with PBS to remove any planktonic fungus, then lifted using the Lift buffer. ROS generation upon fungal infection was then assessed using flow cytometry (32).

For cytokine analysis, the spleens were collected from euthanized mice on day 4 post-infection and homogenized in PBS containing a Protease Inhibitor cocktail (Cat: 78425, Thermo Fisher Scientific). Cytokine concentrations in splenic lysates were assessed using the Mouse Magnetic Luminex Assay Kit (Cat: LXSAMSM, R&D Systems) according to the manufacturer’s instructions, and analyzed using the Luminex system (Invitrogen).

The mouse groups were compared by the log-rank (Mantel-Cox) test for survival studies. In these studies, at least nine mice per group were used based on prior experience with the C. auris infection model, which results in 100% mortality when untreated, and on a power analysis for a two-arm survival comparison (placebo vs. treatment) using the log-rank (Mantel-Cox) test. This sample size provided 90% power by assuming a 50% survival benefit in the treatment group (Hazard Ratio = 0.5), one-sided α = 0.05, equal allocation, and complete event occurrence. Selected treatment groups were repeated, and data from replicates were pooled for analysis. No confounders were considered in the study, since all immunosuppressed uninfected mice appeared healthy.

Sample size calculations for fungal burden experiments assumed α = 0.05, 90% power, and an effect size of at least 1.2 log CFU, resulting in an estimated sample size of approximately 10 mice per group. We used 11 mice per group to ensure we have sufficient surviving mice at the time of fungal enumeration on day 4 post-infection.

Percent weight loss of mice (Supplementary Figure S1B), tissue fungal burden, immune cell functional activity, and cytokine analysis (Supplementary Figure S6) data are graphically presented as median ± interquartile range (IQR). Statistical significance between the two treatment groups was assessed using the Mann-Whitney test. For multiple comparisons among groups, we used the Kruskal-Wallis test followed by Dunnett’s correction for cytokine analysis, and a 2-way ANOVA (Tukey’s test) for elastase activity. These statistical tests were used to nullify the effects of outliers, making them suitable for small sample sizes.

Power analysis was not conducted for experiments involving the determination of immune cell frequency. The data for immune cell phenotyping (Supplementary Figures S2, S4, S5) were presented as bar graphs with each data point representing an independent biological replicate (3 per group), ensuring independence between groups. Because comparisons involved two independent groups without a directional hypothesis, we used an unpaired, two−tailed Student’s t−test to assess differences in means. Data are shown as mean ± SEM to indicate precision, and the t−test is appropriate under the assumptions of approximate normality and equal variances, which are reasonable for these aggregated phenotyping measures.

Graphical representations and statistical analyses of numerical data were performed using GraphPad Prism (v.10.4). Detailed descriptive statistics are presented as part of the main figure or separately (Supplementary Table S1-3) under the supplementary file. Histopathological images are representative of >3 biological replicates.

Mice were housed in the Lundquist Institute (TLI) pathogen-free animal facility until infection, and all animal procedures were approved by and in compliance with the TLI IACUC guidelines (Protocol # 2024-33082-01).

The study design, data presentation, statistical analysis, and reporting were conducted in accordance with the ARRIVE (Animal Research: Reporting of In Vivo Experiments) guidelines to ensure transparency, reproducibility, and ethical standards in animal research (33).

Since C. auris causes lethal bloodstream infections in immunosuppressed patients, we utilized a clinically relevant cyclophosphamide and cortisone acetate-induced immunosuppressed mouse model in an outbred ICR CD-1 background. We hematogenously infected these mice via tail vein injection with a lethal inoculum (5 x 10⁷ yeast/mouse) of C. auris (CAU-09) or (5 x 10⁴ cells/mouse) of C. albicans (SC5314) after 2 days of immunosuppression. Mice were treated with intraperitoneal injection of GM-CSF once daily for up to four days, starting 24 hours post-infection. All infected mice were randomly divided into different treatment groups. Treatment efficacy and mouse health were evaluated by monitoring survival for 21 days and body weight loss at day 4 post-infection. No mice were excluded from the analysis (Figure 1A, Supplementary Figure S1A).

Twenty-one-day survivals of mice were analyzed for each treatment group using Kaplan-Meier analysis. All placebo (diluent vehicle)-treated mice (n=19, pooled from two independent experiments) succumbed to infection before day 21, with 0% survival and a median survival time (MST) of 10 days (95% CI: 4–15). A low dose of GM-CSF (0.2 µg/mouse, n=10) improved survival to 30% with a median survival of 11.5 days (95% CI: 6 – >21), showing a trend toward significance compared to placebo (HR = 0.485, 95% CI: 0.2232–1.054; p = 0.0577). Consistent with this trend, a high-dose of GM-CSF (2.0 µg/mouse, n=19 pooled from two independent experiments) further enhanced survival to 32% and extended median survival to 16 days (95% CI: 11 – >21), achieving statistical significance versus placebo (HR = 0.3901, 95% CI: 0.1898–0.8018; p = 0.0022) (Figures 1B, C).

We also evaluated whether the GM-CSF therapy synergizes with an approved antifungal drug. A sub-therapeutic dose (0.5 mg/kg body weight) of micafungin combined with GM-CSF (2 μg GM-CSF/mouse, n=9) was tested. Micafungin monotherapy (n=10) demonstrated limited benefit, with 21-day survival of 20% and an MST of 5.5 days (95% CI: 5–13), which was not significantly different from placebo (HR = 0.8776, 95% CI: 0.3921–1.965; p = 0.7288). Combination therapy with GM-CSF and micafungin demonstrated the most significant improvement, with 55.6% survival at day 21 and a median survival of >21 days. This regimen significantly reduced mortality risk compared to placebo (HR = 0.2784, 95% CI: 0.122–0.6356; p = 0.0064). The survival benefit of combination therapy did not reach statistical significance compared with GM-CSF (χ² = 0.8857, p = 0.3466) or micafungin monotherapy (χ² = 3.629, p = 0.0568) (Figures 1B, C).

C. auris-infected mice lose weight as the infection progresses, which reflects the overall health of the mice. Thus, we compared weight loss among different treatment groups after 4 days of infection. The placebo-treated mice (n=9) showed the highest weight loss (Median:18.42%, IQR: 16.72 – 23.52), followed by micafungin (n=9, Median: 18.42%, IQR: 9.925 – 18.42), GM-CSF (n= 8, one mouse died before day 4 weight measurement, Median: 11.62%, IQR: 9.078 – 17.57), and combination therapy (n=9, Median: 11.62%, IQR: 3.130 –18.42). Thus, compared to placebo, both GM-CSF monotherapy (U = 9, two-tailed exact p=0.0058) and antifungal combination therapy (U = 18.50, two-tailed exact p=0.0489) showed significantly less weight loss due to C. auris infection, demonstrating superior protective efficacy compared to the placebo-treated group (Figures 1D, E).

Interestingly, GM-CSF therapy failed to prevent mortality in C. albicans-infected mice (n=9/group), but lost significantly less weight (n=3, median: 11.11%, IQR: 5.56 – 13.89) vs placebo (n=5, median: 19.77%, IQR: 16.91– 21.2) in surviving mice at day 3 post infection (U = 0, two-tailed exact p=0.0179) (Supplementary Figures S1B, C).

We further investigated fungal control in target organ tissues by evaluating fungal burden after 4 days of GM-CSF therapy. Immunosuppressed ICR CD-1 mice infected with a sub-lethal 1 × 10⁷ yeast/mouse inoculum of C. auris were randomly assigned to two treatment groups: placebo or GM-CSF therapy. In this model, mice were treated with 2 μg/mouse of GM-CSF once daily, starting 24 hours post-infection and continuing for up to 96 hours, after which the target organs of kidneys, heart, and brain were harvested and processed for fungal enumeration and histopathological examination (Figure 2A).

The GM-CSF treatment significantly reduced median C. auris fungal burden in the mice’s kidneys by 1.42 log (Hodges-Lehmann difference, U = 4, two-tailed exact p<0.0001), hearts by 0.75 log (U = 18.5, two-tailed exact p=0.0043), and brains by 0.435 log (U = 20.5, two-tailed exact p=0.0134) per gram tissue when compared to placebo, respectively (Figures 2B, C). The histopathological examination corroborated these data on tissue fungal burden. Specifically, mice treated with GM-CSF had significantly fewer and smaller fungal abscesses in the kidney at day 4 and in both the kidney and the heart at day 7 post-infection. Furthermore, compared to day 4, the fungal lesions in kidney tissues were reduced even further by day 7. Additionally, GM-CSF-treated mice showed less necrosis and tissue damage than the placebo-treated mice, as indicated by visually more intact kidney and heart tissue architecture (Figures 2D, E).

We also studied the effect of GM-CSF (2 μg/mouse) on the spleen immune cell population in uninfected and infected mice. Treatment of uninfected mice with GM-CSF for 4 days significantly increased the populations of CD3⁺ T cells (mean difference: 35.80, 95% CI: 15.88 – 55.72, two-tailed t-test p=0.0075) and CD4⁺ T helper cells (mean difference: 19.87, 95% CI: 2.649 – 37.08, two-tailed t-test p=0.0328) vs placebo treated mice (Figures 3A, B; Supplementary Figures S2A, B, two-tailed p<0.05). We also found an increase in CD11b⁺ granulocytes (mean difference: 11.82, 95% CI: 0.207 – 23.43, two-tailed t-test p=0.0475) in GM-CSF-treated mice (Figure 3D, Supplementary Figure S2D), which was generally accompanied by a strong trend toward increased neutrophils (mean difference: 13.85, 95% CI: -2.555 – 30.26, two-tailed t-test p=0.079) vs placebo-treated mice (Figure 3E, Supplementary Figure S2E). The CD19⁺ B cells (mean difference: -16, 95% CI: -32.26 – 0.2577, two-tailed t-test p=0.0523) and macrophage population (mean difference: 2.867, 95% CI: -9.956 – 15.69, two-tailed t-test p=0.5684) in the spleen were not changed significantly by GM-CSF treatment (Figures 3C, F; Supplementary Figures S2C, F). Interestingly, four days of GM-CSF treatment also increased the neutrophil population in the bone marrow (Supplementary Figure S3A) and both the neutrophil and macrophage populations in the intraperitoneal cavity in immunocompetent mice (Supplementary Figure S3B).

Next, we investigated the effect of GM-CSF treatment on the frequency of different immune cell subsets in the spleen on days 4 and 7 after C. auris infection using flow cytometry. GM-CSF treatment resulted in a strong trend of elevated CD4⁺ T helper cell frequency on day 4 post-infection vs placebo treatment (mean difference: 6.733, 95% CI: -0.01190 – 13.48, two-tailed t-test p= 0.0503), but this trend was diminished by day 7 (mean difference: 12.79, 95% CI: -12.84 – 38.41, two-tailed t-test p = 0.2382, Figures 4A, Supplementary Figure S4A). B Cell frequencies were similar in both the placebo and GM-CSF-treated mice after days 4 and 7 post-infections (Figure 4B, Supplementary Figure S4B). The CD11b⁺ granulocytes have a significantly elevated frequency in GM-CSF-treated mice vs placebo-treated mice (mean difference: 4.433, 95% CI: 0.3644 – 8.502, two-tailed t-test p = 0.039) and a strong trend of elevated neutrophils (Ly6G⁺) frequency (mean difference: 4.763, 95% CI: -0.01623 – 9.543, two-tailed t-test p = 0.0505) on day 4 post-infection, but not on day 7 (Figures 4C, D; Supplementary Figures S4C, D). The macrophages (F4/80⁺) were not altered by GM-CSF treatment compared to the placebo on either day 4 or 7 post-infection (Figures 4E, Supplementary Figure S4E). Notably, T helper cells, B cells, granulocytes, and neutrophils showed a decrease in the spleen by day 7 in both treatment groups, while macrophage populations were significantly increased.

We also determined the frequencies of immune cells in the kidneys of immunosuppressed mice after 7 days of C. auris infection, the primary target organ in this model. GM-CSF-treated mice showed a significant increase in the B cell (mean difference: 0.8967, 95% CI: 0.07003 – 1.723, two-tailed t-test p = 0.0395) and macrophage (mean difference: 6.067, 95% CI: 0.6957 – 11.44, two-tailed t-test p = 0.035) populations vs placebo-treated mice. In contrast, the frequencies of T helper cells, granulocytes, and neutrophils were not significantly different from those in the placebo group (Figures 4F–K; Supplementary Figures S5A-E). We did not evaluate the immune cell populations in the kidneys on day 4 post-infection. The detailed statistical description of the frequencies of immune cell phenotype is provided in Supplementary Table S1 under the Supplementary file.

We also assessed the levels of proinflammatory cytokines (IFN-γ, TNF-α, IL-2, IL-4, IL-6, IL-12p70, IL-17A, and IL-10) in the spleen of placebo or GM-CSF-treated C. auris-infected or uninfected mice. GM-CSF-treated C. auris-infected mice showed significantly increased levels of IL -6 (Median: 11.64 pg/ml, IQR: 9.133 – 37.74) vs uninfected mice (Median: 7.748 pg/ml, IQR: 7.666 – 7.817, adjusted p=0.0043, Supplementary Figure S6A). Similarly, TNF-α levels (Median: 1.604 pg/ml, IQR: 1.497 – 1.646) were also significantly increased vs uninfected mice (Median: 1.256 pg/ml, IQR: 7.666 – 7.817, adjusted p=0.0128, Supplementary Figure S6B). Notably, TNF-α was detected at very low levels (<2 pg/ml), while other cytokines were undetectable.

We further investigated the immunomodulatory effects of GM-CSF treatment on ex vivo functional activities of immune cells and profiled immune cells from the bone marrow and peritoneal exudates of GM-CSF- or placebo-treated mice. Bone marrow neutrophils from GM-CSF-treated mice (3 out of 5) demonstrated increased phagocytic uptake of labelled C. auris (Figure 5A), along with significantly increased clearance of the fungus at four hours (U = 2.5, 2-tailed exact p=0.0397, Figure 5B). Phagocytic uptake of pathogens by innate immune cells is accompanied by the generation of reactive oxygen species (ROS) as an early antimicrobial response (34). Thus, we next investigated how GM-CSF treatment modulated neutrophil ROS responses during C. auris infection. Remarkably, C. auris-induced ROS was significantly higher in neutrophils from GM-CSF-treated mice compared to the placebo group (U = 0, 2-tailed exact p=0.0079, Figure 5C). Activated neutrophils often utilize ROS-dependent neutrophil extracellular trap formation (NETosis) for efficient clearance of pathogens (35). As a surrogate for NETosis, we measured neutrophil elastase release into the media upon C. auris infection. As expected, an increased C. auris-induced ROS in neutrophils from GM-CSF-treated mice also showed significantly increased secreted neutrophil elastase (Mean difference ± SE: 1214 ± 138.1, 2-way ANOVA with Tukey’s adjusted p=0.0032). NETosis is usually associated with increased neutrophil cell death during infection (36). However, we did not observe a significant increase in neutrophil cell death in C. auris-infected neutrophils derived from the bone marrow of GM-CSF-treated mice compared to placebo-treated mice (Figures 5D, E). A similar enhanced antifungal phenotype was also observed for cells from the peritoneal cavity, with these cells from GM-CSF-treated mice showing a strong trend in the enhancement of phagocytic uptake of fluorescently labelled C. auris (U = 3, 2-tailed exact p= 0.0556, Figure 5F), clearance of fungus (U = 0, 2-tailed exact p= 0.0079, Figure 5G), and generation of C. auris-induced ROS (U = 0, 2-tailed exact p= 0.0079, Figure 5H). However, unlike bone marrow neutrophils, these cells were significantly more resistant to C. auris-induced cell death than cells from the placebo group (U = 0, 2-tailed exact p=0.0079, Figure 5I).

In the absence of GM-CSF, there was negligible macrophage infiltration in the peritoneal cavity (Supplementary Figure S3B). Therefore, to investigate the effect of GM-CSF on macrophage antifungal responses, we used RAW264.7 macrophages. GM-CSF-treated RAW macrophages exhibited a phenotype similar to that of ex vivo neutrophils. An overnight treatment of GM-CSF resulted in a significantly enhanced uptake (U = 33, 2-tailed exact p=0.0242, Figure 5J) and clearance of C. auris by the GM-CSF-treated macrophages (U = 37, 2-tailed exact p=0.0011, Figure 5K). GM-CSF treatment also resulted in a significantly enhanced C. auris-induced ROS (U = 66.5, 2-tailed exact p=0.0035, Figure 5L). Additionally, as previously observed with cells from the peritoneal cavity, GM-CSF-treated RAW macrophages were significantly more resistant to C. auris-dependent cell death (U = 36, 2-tailed exact p=0.001, Figure 5M). The detailed statistical descriptions of these in vitro anti-C. auris immune cell functional activities are available in Supplementary Tables S2-S3 under the Supplementary file.

Finally, no adverse events (e.g., weight loss, respiratory distress, behavioral changes, moribundity) were observed in uninfected immunocompetent (n=5) or immunosuppressed (n=3) mice during or after four days of GM-CSF administration (2 μg/mouse), indicating that the treatment was well tolerated under these conditions.