A. castellanii strain 30234 was acquired from the American Type Culture Collection (ATCC). This strain was maintained in peptone-yeast extract-glucose (PYG) broth (ATCC medium 354) at 28°C in the dark as detailed in (Steenbergen et al., 2004). For experimental use and routine maintenance, A. castellanii was cultured as adherent cells in PYG medium at 28°C for 5-7 days in 75-cm² culture flasks (Bozue and Johnson, 1996). Amoebae were collected by tapping the flasks, centrifuged at 3,000 rpm for 10 minutes, and suspended in sterile phosphate buffer saline (1× PBS, pH 7.2). The cells were counted using a modified Fuchs-Rosenthal chamber, and viability was evaluated by trypan blue staining (Pruksaphon et al., 2022). The initial viability was always greater than 98% (data not shown).

C. albicans strain ATCC 90028 and environmental isolates of C. albicans were obtained and cultivated according to the directions from Thailand Bioresource Research Center (TBRC), BIOTECH Culture Collection (BCC), Pathum Thani, Thailand. The environmental isolates were collected from different natural ecological niches in Thailand and listed in Table 1 . The C. albicans isolates were stored at -80 °C. Prior to use in experiments, the isolates were grown on Sabouraud dextrose agar (SDA; Difco Laboratories., Detroit, MI) for 24 hours at 28°C. Then, yeast cells were harvested by centrifugation at 5,000 rpm for 15 minutes and washed three times with PBS. The yeasts were then cultivated on peptone yeast extract glucose (PYG) agar at 28°C for 48 hours, collected by centrifugation at 5,000 rpm for 15 minutes, and washed three times with PBS. Cell counting was performed using a hemocytometer prior to their use in all experiments.

To study yeast-amoebae interactions, three hundred C. albicans yeast cells were spread on PYG agar with a disposable spreader. A. castellanii cells (5×10³ cells/mL in 5 mL) were poured onto the agar surface containing the yeast cells. The mixture was swirled to ensure complete coverage of the plate by amoebae. Plates were sealed with parafilm and incubated at 28°C for 4 weeks to assess for surviving colonies of C. albicans and photographed using a stereomicroscope (Drawell, China). After 4 weeks of interaction, 4 surviving colonies from each strain were randomly picked and transferred into fresh PYG agar without amoeba. The plates were incubated at 28°C for 48 hours, and the macroscopic morphologies of each strain were visualized. The microscopic morphologies were determined under light microscopy (Nikon Eclipse 50i, Tokyo, Japan). Furthermore, colonies of C. albicans that interacted with A. castellanii for 4 weeks (exposed) were maintained on PYG medium and subsequently investigated for various characteristics by comparison to the control group (unexposed).

To investigate the survival of C. albicans (unexposed or exposed isolate) after phagocytosis by amoeba, 1 mL of A. castellanii, 10⁶ cells/mL in PBS, was added to a 12-well plate. C. albicans yeast cells were then added to the acclimated cultures of A. castellanii at a 1:10 effector-to-target ratio and incubated at 28°C for 24 hours. After that, non-phagocytosed C. albicans were collected with PBS into a new tube for indirect quantification of yeast cells. The plates were placed on ice for 10 minutes to loosen the amoeba cells from the bottoms of the plates. Then, the inoculated A. castellanii was lysed by pulling the suspension through a 27-gauge needle five times or more. The yeast cells were then subjected to a 10-fold dilution. The viability of yeast cells was assessed by spreading them on SDA and incubating at 28°C for 24 hours, determined by the colony-forming units (CFU)/mL count and calculated as direct determination of the cell viability (Pruksaphon et al., 2022).

Yeast cell suspensions were subjected to serial dilutions, and 3 μL droplets of each dilution (10⁶–10 cells/mL) were deposited onto PYG agar supplemented with various stressors at different concentrations. Menadione (MD; Sigma-Aldrich, St. Louis, MO) and hydrogen peroxide (H₂O_2; Merck KGaA, Germany) were utilized to assess resistance to oxidative stressors, while sodium dodecyl sulfate (SDS; Calbiochem, Japan) and Congo red (CR; Sigma-Aldrich, St. Louis, MO) were employed to examine resistance to cell wall stress. The stressors were administered at the following concentrations: 0.025 mM, 0.05 mM, and 0.1 mM for menadione; 1 mM, 2.5 mM, and 5 mM for hydrogen peroxide (Cuéllar-Cruz et al., 2009); 10 μg, 50 μg, and 100 μg for sodium dodecyl sulfate; and 100 μg and 150 μg for Congo red (Schroeder and Ikui, 2019). The azole antifungal drug fluconazole (at 20 and 40 μg/mL) was utilized for antifungal sensitivity testing (Sun et al., 2023). Plates with the various stressors were incubated at 28°C for 24 hours, and the number of colonies were enumerated at a dilution of 10³ to determine and compare the colony forming units (CFU)/mL for each strain (Wang et al., 2017). After an incubation period of 48 hours, photographs of plates were taken. To study heat stress, plates were placed at 42 °C for 24 hours, and CFU/mL were then determined.

To assess yeast germination, 1×10³ cells/mL of unexposed C. albicans isolates or AC-exposed isolates were inoculated into pooled healthy human serum (Sigma-Aldrich, St. Louis, MO). The inoculated samples were incubated at 37°C for 2 hours, and germlings were counted using a light microscope (Nikon Eclipse 50i, Tokyo, Japan). The percentage of germinated cells was determined by examining 30 fields with each isolate.

The A549 cell line (human lung adenocarcinoma cell line, CCL-185™) was obtained from ATCC and cultured in Dulbecco’s minimal essential medium (DMEM; Gibco, Thermo Fisher Scientific, Inc., Waltham, MA) supplemented with 10% fetal bovine serum (FBS; Gibco, Thermo Fisher Scientific, Inc., Waltham, MA) in an incubator at 37°C under 5% CO₂. The impact of C. albicans on A549 cell damage was assessed by measuring LDH release into supernatant. Briefly, yeast cells (control unexposed cells or amoebae-exposed cells) were suspended in Hank’s balanced salt solution (HBSS; Gibco, Thermo Fisher Scientific, Inc., Waltham, MA) supplemented with 1% healthy human serum. Then, confluent monolayers (5×10⁵ cells/mL in a 12-well plate) were inoculated with each isolate of C. albicans at an optimized MOI of 5. After 12 hours of infection, the culture supernatants were centrifuged to remove yeast cells and debris, and LDH enzymatic activity was measured using a Cobas 8000 automatic analyze (Roche Diagnostics, Mannheim, Germany) (Owen et al., 2015). The LDH positive control was established by lysing the uninoculated A549 cells with 0.5% Triton X-100 in HBSS. The enzymatic activity of supernatant from uninoculated control A549 cells was normalized as the background level. The percentage of A549 cytotoxicity was calculated following formula:

Single colonies from unexposed (UN) and exposed (EX) isolates of C. albicans TBRC-BCC 59444 were isolated on PYG medium. Three colonies from each isolate (UN and EX) were individually inoculated into 5 mL of YPG medium and allowed to grow overnight at 30°C. The resulting cultures were designated as EX1, EX2, EX3, UN1, UN2, and UN3 isolates. The DNA of each isolate was extracted as detailed in (Wangsanut et al., 2023).

DNA samples, each containing 6 μg or more, were shipped to the Novogene service for whole-genome DNA sequencing using the Illumina Sequencing Package. The DNA sequencing data were returned in FASTQ (.fq) format file, containing sequencing reads and corresponding base quality. The data was then converted to sorted BAM format using Samtools (Li et al., 2009). To analyze results, paired-end reads were mapped to the reference genome of C. albicans SC5314 using Yeast Mapping Analysis Pipeline (YMAP) (Abbey et al., 2014). Additionally, Single nucleotide polymorphisms (SNPs) and insertion-deletion (indel) variations were investigated by GATK (DePristo et al., 2011). Copy-number variation (CNV) and Structural variants (SVs) were detected by CNVnator (Abyzov et al., 2011) and BreakDancer (Chen et al., 2009), respectively. Identified mutations were individually assessed using the Integrative Genomics Viewer (IGV) (Ene et al., 2018).

To determine the formation of biofilm in amoeba-exposed compared to unexposed yeast, a biofilm formation assay was performed using a C. albicans isolate from a water-hot spring (WH) according to the protocol described (Gulati et al., 2018). Briefly, C. albicans isolates grown for 48 hours on PYG were diluted in RPMI-1640 to obtain 5×10⁴ cells/mL. Subsequently, 100 μL of C. albicans (5×10³ cells/well) were seeded into a 96-well plate and incubated at 37 °C for 24 hours. Afterward, the plate was washed three times with PBS and fixed with 99% (v/v) methanol. The plate was then air-dried at room temperature for 45 minutes. After that, the biofilm was stained with 0.05% crystal violet (CV) for 30 minutes and washed three times with PBS. To measure absorbance, 200 μL of 33% (v/v) acetic acid in PBS was added to de-stain for 45 minutes, and one-hundred microliter of each well was transferred to a new plate and measured at OD₅₉₅ nm.

To investigate the impact of the FTH2 gene mutation on the growth of exposed isolates compared to unexposed isolates under iron-free, iron-depleted, and iron-replete conditions (Hsu et al., 2011). To generate each iron condition, PYG medium was supplemented with 100 μM bathophenanthroline (BPS) to chelate iron ions (Fe²⁺ or Fe³⁺) in the media which is an iron-free condition. For the iron-depleted condition, PYG medium was supplemented with 100 μM BPS and 0.01 mM ferric ammonium citrate (FAC). In the iron-repletion, 0.1 mM FAC was added to the medium. In brief, yeast cells were cultured on PYG medium at 28 °C for 48 hours. After that, the cells were counted to obtain 10⁶ cells/mL and subjected to a 10-fold dilution, ranging from 10⁶ to 10¹. Candida cells were spotted on the medium conditions and incubated at 28 °C for 24 hours. The number of colonies were counted at a dilution of 10³ to determine and compare the colony forming units (CFU)/mL for each isolate.

To evaluate the growth of an adenine auxotrophic C. albicans strain with an ADE4 gene mutation, a synthetic complete (SC) medium with or without adenine was used. SC medium was prepared by supplementing with 2% dextrose, 6.7% yeast nitrogen base (YNB) plus ammonium sulfate, and with 0.2 mM adenine (SC+Ade) or without adenine (SC-Ade) (Wangsanut et al., 2017). Briefly, the yeast cells were cultured on PYG medium at 28 °C for 48 hours. Subsequently, the cells were counted using a hemocytometer. The suspended cells were then subjected to a 10-fold dilution ranging from 10⁶ to 10¹. To observe the growth of Candida colonies, these cells were spotted onto SC+Ade or SC-Ade medium and incubated at 28 °C for 24 hours.

The data underwent statistical analysis utilizing a range of tests. The normality of the dataset was assessed using Shapiro-Wilk test. Virulence-related phenotypic characteristics between unexposed and exposed-Candida were compared using Kruskal-Wallis non-parametric test. Growth at 42 °C was determined by one-way ANOVA. Two-way ANOVA with Tukey’s multiple comparison test was used to compare other experiments. A significance threshold of P < 0.05, P < 0.001, and P < 0.0001 was adopted for all statistical analyses. The data analysis was conducted using GraphPad Prism 9.0 software (version 9.0) and IBM SPSS Statistics (version 29).

We investigated the interaction between five environmental isolates of C. albicans (from flower; F, sediment; S, mushroom; MH, water-hot spring; WH, moss; MS) and A. castellanii. The experimental approach involved co-culturing these organisms on PYG agar and is detailed in Figure 1A . The co-cultures were maintained for 4 weeks, at which time small colonies of yeast were macroscopically visible, predominantly situated beneath the amoebae mat ( Figure 1B ). Microscopic examination of C. albicans surviving colonies revealed a mixed population comprised by yeast, pseudohyphal, and hyphal forms in all isolates of environmentally derived C. albicans, with varying proportions among each individual isolate ( Figure 1C ).

To ensure a representative sampling, four single colonies were randomly selected from each isolate to study their phenotypic characteristics and transferred onto fresh PYG agar plates devoid of amoebae. Unexposed isolates were used as controls. Within a span of 24 hours, discernible colonies composed exclusively of yeast cells appeared on the agar surface. Importantly, after a period of 3 days of agar growth, colonies from C. albicans MS, WH, S, and F were smooth while colonies of S were rough ( Figure 1D ). Moreover, we found that four isolates (MS1-4, WH1-4, S1-4 and MH1-4) generated hyphae following their interaction with amoebae ( Figures 1D–F ). Notably, unexposed Candida cells from each isolate exhibited a normal yeast morphology ( Figures 1G, H ). Even though isolates MH1-4 presented yeast colonies, they also generated segments with hyphal morphology around their colonies ( Figures 1D–F ). However, there was a single MS isolate that spontaneously germinated to form hyphae, even though it had not been exposed to amoebae ( Figures 1G, H ). The experiment was subsequently replicated utilizing C. albicans ATCC 90028. However, no surviving colonies were observed from the control strain on three co-incubated plates with amoebae even after an additional week (4-weeks) of incubation. These findings provide compelling evidence that, following interaction with amoebae, the environmental strains were capable of displaying a diverse array of cellular and colony morphologies, even in the absence of amoebae in the growth medium.

Our prior findings revealed that interaction with amoeba results in the alteration of C. albicans morphologies. We speculated that this change might lead to an increase in the pathogenicity of the fungus. Thus, we next investigated the impact of amoebae interaction on the pathogenicity of C. albicans, particularly focusing on hyphal formation and invasion of the lung human A549 cell line. To achieve this, we analyzed various virulence-related phenotypic characteristics of isolates derived from five C. albicans isolates. We observed that exposed isolates WH, S, and MH exhibited significantly faster and greater amounts of hyphal formation compared to their respective unexposed strains when cultured in healthy human serum within 2 hours after incubation. In contrast, exposed isolates MS and F did not show differences in hyphal formation when compared to their unexposed isolates ( Figure 2A ). Next, to assess fungal-associated cytotoxicity, we measured the release of LDH by the A549 cells (Saiprom et al., 2023). We investigated whether this cytotoxicity was present during interactions with host-exposed C. albicans isolates. After inoculating the exposed C. albicans isolates and their unexposed isolates with A549 cells, we measured LDH enzyme release at 12 hours post-incubation. The result showed that the unexposed strain resulted in 5% cytotoxicity. Exposed MS and MH exhibited cytotoxicity percentages similar to their unexposed, ranging from 2% to 5%. On the other hand, a high level of LDH was detected from A549 cells inoculated with exposed WH, S, and F, with levels up to 99% ( Figure 2B ). Moreover, we assessed the percentage survival of exposed isolates compared to that of their unexposed strains after being phagocytosed by amoebae. The results showed that the unexposed strains had a low survival rate, whereas the exposed isolates exhibited a higher percentage of survival ( Figure 2C ). Our findings suggest that the WH and S evolved a heightened ability to generate hyphae and invade and destroy host cells. Although the exposed F had a percentage of hyphal formation similar to its unexposed counterpart, it displayed a remarkable capacity to cause damage to A549 cells. This interesting characteristic suggests that this particular isolate might be able to produce elevated levels of Candida exotoxins or enzymes compared to its unexposed. However, further studies are necessary to elucidate the specific factors responsible for the virulence of the F isolate.

To investigate whether cultivation with amoebae affected the growth of the C. albicans isolates via stressor responses, the fungal isolates were subjected to various individual stress conditions. These conditions included exposure to differing concentrations of oxidative stresses (H₂O₂ and menadione), cell wall stresses (Congo Red and SDS), and thermal stress at 42°C. Growth was assessed on plates at a concentration of 10³ cells, and the average growth from six experiments (two independent technical replicates for each of three biological repeats) was used as a quantitative measure to compare the effects of stress on growth of each isolate compared to its growth in standard media.

Isolates were tested in various concentrations of H₂O₂ and menadione, the latter being a cytotoxic quinone that generates superoxide (Frohner et al., 2009). Results showed that the growth of WH1-4 and other isolates was similar to control conditions at concentrations of 1 and 2 mM of H₂O₂, but not at 5 mM ( Figures 3C, D , Supplementary Figure 1 ). Moreover, the amoebae-exposed isolates survived menadione up to a concentration of 10 μM ( Figures 3A, B , Supplementary Figure 1 ), marking them as the most resistant to superoxide ion. When the number of colonies for each sample was calculated, the results show that amoebae-exposed isolates exhibited significantly enhanced growth under both oxidative stressors compared to their unexposed isolates. Moreover, all amoebae-exposed isolates showed increased growth when exposed to thermal stress at 42°C, especially S4, WH4, MH1-4, F1, F3, and F4 ( Figures 3I, J , Supplementary Figure 1 ). Most of the amoebae-exposed C. albicans strains demonstrated an increased resistance to SDS ( Figures 3E, F ). In contrast, amoebae-exposed several isolates were sensitive to Congo Red and could not grow under this stressor but isolates WH1-4 and MH2 grew in the presence of Congo Red ( Figures 3G, H , Supplement Figure 2 ). Furthermore, these amoebae-exposed isolates exhibited a halo zone around their colonies on Congo Red, a trait absent in the unexposed isolate. These results demonstrated the substantial influence of interactions between amoebae and C. albicans. Such interactions significantly impact the adaptive resistance of C. albicans to oxidative agents, cell wall stressors, and high-temperature challenges.

We further investigated the impact of amoebae on the growth of the amoebae-exposed yeast in the presence of an antifungal agent. We assessed changes in drug susceptibility by performing spot assays, comparing the amoebae-exposed isolates to their unexposed counterparts for growth in the presence of fluconazole. All unexposed isolates were susceptible to 20 μg/mL and 40 μg/mL of the antifungal drug. Interestingly, although many of the exposed isolates remained susceptible to fluconazole, isolates WH1, S1, S2, and S4 exhibited a prominent increase in resistance to fluconazole compared to their unexposed counterpart ( Figures 3K, L , Supplementary Figure 1 ). These findings demonstrate that interactions with amoebae may induce adaptations in C. albicans that alter their susceptibility to fluconazole.

In a previous study on C. neoformans host adaptation, genotypic variations were shown to play a role in altering morphology and virulence (Magditch et al., 2012). Thus, we investigated whether amoeba interactions could have an impact on genotypic variation in an environmental C. albicans strain. To achieve this, we conducted a comprehensive analysis of the C. albicans genome for an amoebae-exposed WH isolate and its unexposed counterpart. The total length of the SC5314 assembled genome sequence was 14,282,666 base pairs, while the mapping rate of unexposed and exposed isolates range from 92.15% to 94.61%. The average depths were between 52.55 and 88.51, and coverages range were from 99.68% to 99.74%. These results were in the qualified normal range. Thus, we examined single nucleotide polymorphisms (SNPs), insertion-deletion (Indels), copy number variation (CNVs), and structural variation (SVs), and utilized the reference genome of C. albicans SC5314 for comparisons. The resulting genome-wide scale changes were analyzed by YMAP plots ( Supplementary Figure 2 ).