Lichen secondary metabolites were mainly obtained as gifts from the Huneck’ chemical library (Botanical Museum in Berlin), stored in the lab of Pharmacognosy of the Faculty of Pharmacy in Rennes (Institute of Chemical Sciences). For most compounds, information concerning isolation of these compounds from lichens or obtention as lichen compound derivatives can be retrieved (Huneck and Yoshimura, 1996; Desmarets et al., 2023). (+)-Lichesterinic acid (LIA) was obtained by synthesis according to Sweidan et al. (Sweidan et al., 2016). Note that diterpene compounds are not usually reported from lichen; however, the diterpene compounds 1, 32 and 35 in this screen have been recently described to occur in lichens (De Angelis et al., 2003; Kerboua et al., 2021; Koopaie et al., 2023).

Crystallized LDCs were dissolved in dimethyl sulfoxide (DMSO) to a concentration of 25 mM directly before first usage and stored at -20°C for potential subsequent validation analyses. In some cases, the compounds were not completely soluble at this concentration and were further diluted with DMSO to reach total solubilization. The highest tested concentration for each LDC can be found in Supplementary Table 1. All assays were conducted with freshly dissolved compounds except for the low-throughput BFI assays in N. glabratus of LDCs 48, 51, 55, 56, 59 and 60 for screen hit validation. Fresh stocks were no longer available, and frozen stocks were used instead.

C. albicans SC5314 wild type (WT) was used for all antifungal experiments. N. glabratus BG2 (WT) was only used for the screen and validation of LDCs as well as for biofilm eradication experiments. C. dubliniensis IHEM 14280 (WT) and Pichia kudriavzevii IHEM 6104 (WT) were only used for complementary experiments with LIA. All growth media were prepared with ddH₂O. YPD (10 g/l yeast extract, 20 g/l peptone, 20 g/l D-glucose) and YPD agar plates (YPD + 20 g/l agar) were subsequently autoclaved (25 min, 121°C, 210 kPa). RPMI media (10.4 g/l RPMI-1640 powder (Sigma), 34.53 g/l MOPS, pH adjusted to 7.0 with NaOH) was sterile filtered and stored at 4°C.

C. albicans and N. glabratus were streaked on YPD agar plates and incubated at 37°C for 24 h. For each experiment overnight cultures (ONCs) of C. albicans and N. glabratus were inoculated with material from YPD agar plates (not older than 2 weeks) in 3 ml YPD in glass eprouvettes and incubated at 30°C at 170 rpm shaking for 14 h – 18 h.

For antifungal susceptibility testing, C. albicans and N. glabratus were grown in RPMI media at 37°C to simulate human blood conditions. For experiments on planktonic cells, they were grown in 96-well flat bottom plates (Greiner Bio-One, Austria) which were sealed with gas-permeable adhesive membranes (Excel Scientific) and closed with a lid. Biofilms were grown in 96-well tissue culture test plates (U-bottom, TPP, Switzerland) which were closed with the supplied lids. The rim wells of either plate were not used because they are prone to drying out. They were filled with 100 µl of sterile ddH₂O instead to avoid evaporation.

The minimal inhibitory concentration (MIC) of LDCs in C. albicans and N. glabratus was determined following the standard Clinical and Laboratory Standards Institute (CLSI) protocol M27 for yeast strains (CLSI, 2017) with some modifications. C. albicans and N. glabratus, were both treated with two-fold serial dilutions of LDCs. For 25 mM stocks of LDCs the tested final concentrations were 250 µM, 125 µM, 62.5 µM, 31.3 µM, 15.6 µM, 7.8 µM, 3.9 µM and 2.0 µM. Serial dilutions were done with DMSO and then further diluted 1:10 with RPMI. 10 µl of those dilutions were then added to the wells of 96-well flat bottom plates. In some cases, this initial dilution resulted in visible precipitation at the 1–2 highest concentrations; in these instances, particular care was taken to thoroughly resuspend the sample and to accurately apply 10 µl. ONCs were prepared in 3 ml YPD and incubated at 30°C and 170 rpm for 14-18 h. Fresh RPMI was inoculated with ONCs to an OD₆₀₀ of 0.001 (appr. 1x10⁴ cells/ml) and vortexed thoroughly. 90 µl of the prepared inoculum were then added to each well of 96-well flat bottom plates already containing the diluted compounds. After mixing, the plates were sealed with gas-permeable adhesive membranes, covered with a lid and incubated without shaking at 37°C for 48 h. Each plate contained a two-fold serial dilution of amphotericin B (5 µM - 0.31 µM) as antifungal control and six wells with 1% DMSO as growth controls. OD₄₉₀ was measured after 24 h and 48 ± 1 h with a plate reader (Glomax multi+ detection system, Promega). To analyze growth capacity measured OD₄₉₀ values were then normalized to the mean of corresponding growth controls. Dose-response curves were generated and, if possible, MIC₅₀ values (= concentration at which growth was inhibited by 50% compared to untreated control cells) were calculated (GraphPad Prism 8).

To assess biofilm inhibitory concentrations (BIC) in C. albicans and N. glabratus, both were grown under biofilm-inducing conditions and treated with two-fold serial dilutions of LDCs. Assays were performed as described by Delattin et al. (Delattin et al., 2014) with some modifications. Serial dilutions of the compounds were prepared as previously described (see above) and 10 µl from each of these conditions were then loaded onto the wells of 96-well tissue culture test plates. Concentration of compounds and controls were the same as in MIC experiments. RPMI was inoculated to an OD₆₀₀ of 0.1 with ONCs and 90 µl of the culture were then added to each well of the U-bottom plates. Plates were then incubated for 1 h at 37°C to allow adherence of the cells. The supernatants with non-adherent cells were then removed and the adherent cells were gently washed with 100 µl PBS pH 7.4 (80 g/l NaCl, 2 g/l KCl, 14.4 g/l Na₂HPO₄, 2.4 g/l KH₂PO₄). 100 µl of fresh RPMI with compounds were added and the plates were incubated at 37°C.

After 24 ± 1 h of incubation, inhibition of the biofilms was assessed through measurement of metabolic activity. Biofilms of C. albicans were stained with the resazurin-based redox dye cell titer blue^® (CTB; Promega, Germany) and biofilms of N. glabratus were stained with the tetrazolium salt XTT (ABCR, Germany). For both staining procedures, supernatants with non-adherent cells were removed and biofilms were gently washed with 100 µl PBS pH 7.4. Biofilms of C. albicans were then stained with 100 µl CTB staining solution (CTB diluted 1:100 in PBS pH 7.4) and biofilms of N. glabratus were stained with 100 µL XTT staining solution (0.25 mg/mL XTT + 0.1 µM menadione in PBS pH 7.4). The plates were then incubated at 37°C for 1 h in the dark. For XTT staining in the low throughput validation round as well as for biofilm eradication, biofilms were stained with 120 µl of XTT staining solution instead and after incubation, 100 µl of the supernatant were transferred into fresh 96-well flat bottom plates for measurements. For biofilms stained with CTB, fluorescence was measured by excitation with 525 nm and emission at 580 nm – 640 nm. For biofilms stained with XTT, OD₄₉₀ was measured. All measurements were done with a plate reader (Glomax multi+ detection system, Promega). To assess viability of cells in biofilms measured values were normalized to the mean of corresponding growth controls. Dose-response curves were generated and, if possible, BIC₅₀ values (= concentration at which metabolic activity in biofilms was inhibited by 50% compared to untreated controls) were calculated (GraphPad Prism 8).

To assess effects of compounds against pre-grown biofilms of C. albicans and N. glabratus, biofilm eradication (BFE) measurements were conducted. Assays were performed as described by Vriens et al. (Vriens et al., 2015) with some modifications. RPMI was inoculated to an OD₆₀₀ of 0.1 with an ONC and 100 µl aliquots of the culture were then added to wells of 96-well tissue culture test plates. The plates were then incubated for 1 h at 37°C to allow adherence of the cells. The supernatants were removed and adherent cells were gently washed with 100 µl PBS pH 7.4 to remove non-adherent/planktonic cells. 100 µl of fresh RPMI, preheated to 37°C, were added to the wells and plates were incubated at 37°C. After 24 h the supernatants were removed, the biofilms were gently washed with 100 µl PBS to remove non-adherent/planktonic cells and two-fold serial dilutions of compounds mixed with RPMI were added. The plates were then incubated at 37°C. After 24 h supernatants were removed from the incubated U-bottom plates and the wells were gently washed with 100 µl PBS pH 7.4 to remove non-adherent/planktonic cells. Biofilms were stained, incubated and measured as described for the assessment of BIC. To assess viability of cells in biofilms measured values were normalized to the mean of corresponding growth controls. Dose-response curves were generated and, if possible, BEC₅₀ values (= concentration at which metabolic activity in pre-grown biofilms was inhibited by 50% compared to untreated controls) were calculated (GraphPad Prism 8).

To assess toxicity of LDCs against human cells, viability assays were conducted with Normal Human Dermal Fibroblasts (NHDF) from an adult donor (PromoCell). Assays were conducted in Fibroblast Growth Medium 2 (PromoCell). DMSO content was 0.5% for both LDC-treated cells and untreated controls. To determine the viability of LDC-treated cells, CellTiter-Glo^® 3D Cell Viability Assay (Promega G9682, Madison, WI, USA) was used. The assays were performed according to the manufacturer’s guidelines. Cells were harvested at confluency of approximately 90% and diluted to a concentration of 10⁵ cells/100 µl. Subsequently, 100 µl aliquots were dispensed into 96-well white, clear bottom 96-multiwell plates (Fisher Scientific GmbH, Austria). The measurements were performed with the GloMax^® Discover Microplate Reader (Promega, Madison, WI, USA) at room temperature.

Cells were grown overnight and two-fold serial dilutions of LDCs were added to a final concentration of 250 µM, 125 µM, 62.5 µM, 31.3 µM, 15.6 µM, 7.8 µM and 3.9 µM. After LDC addition, they were incubated for 24 h at 37°C and 5% CO2. After 24 h equal volume of the CellTiter-Glo-Reagent (100 µl) was added to each well of the 96-well plate. The plate was inserted into the measurement device where the plate was shaken vigorously for 5 min, incubated for 25 min and the luminescence was recorded.

CellTiter-Glo^® 3D Cell Viability Assay (Promega G9682, Madison, WI, USA) was used to determine the cell viability in 2D in vitro cell culture. The luminescence measurements were performed with the assay reagent (Ultra-Glo™ rLuciferase) to determine ATP, which indicates viability of the cells and generates a detectable luminescence readout. For 100% toxicity level the cells were treated with 2 µl Triton-X-100 (10%) representing 0% of cell viability (V₀). Cells in the absence of LDCs and Triton were a negative control (V₁₀₀).

The viability of each well was calculated from the percentage of viable cells without LDCs and in the presence of LDCs (V_x) with the following equation:

Filamentation assays on solid media were performed as described by Böttcher et al. (Böttcher et al., 2020). Cells from an ONC of C. albicans were harvested and washed with an equal volume of PBS. Cells were then diluted in PBS to an OD₆₀₀ of 0.5 and 5 µl were spotted onto Spider (1% nutrient broth, 1% mannitol, 0.2% dipotassium phosphate and 1.35% agar), Spider D (1% nutrient broth, 0.5% glucose, 0.2% dipotassium phosphate and 1.35% agar) and RPMI (RPMI media with 2% agar) agar plates. Plates were incubated at 37°C and colony morphology was documented on days 3, 5 and 7.

Potential fungicidal effects against C. albicans were investigated by time-kill and antifungal carryover assays. Both assays were performed as described by Klepser et al. with minor modifications (Klepser et al., 1997). Flasks with 10 ml of RPMI media were inoculated to an OD₆₀₀ of 0.001 with an ONC of C. albicans and treated with either 160 µM, 80 µM or 40 µM of compound. DMSO was used for untreated controls and DMSO content was 1% for all flasks. Three replicate flasks were done for each treatment and controls. Flasks were incubated at 35°C at 150 rpm shaking for 24 h. Samples were taken, serially diluted in PBS and plated on YPD agar plates after 0, 2, 4, 6, 8, 12 and 24 h. The 0 h timepoint was sampled before addition of treatments. Plates were incubated at 30°C for 48 h for colony count determination. Data was plotted as log CFU/ml for each time point.

Antifungal carryover was assayed for all treatment concentrations of the time-kill assay in three replicates by plating 3-5x10² freshly diluted cells from an ONC of C. albicans onto YPD agar plates immediately after treatment. Plates were incubated at 30°C for 48 h for colony count determination. Antifungal carryover was defined as negative when the difference between control and treatment was ≤ 25%.

YPD was inoculated to an OD₆₀₀ of 0.1 and incubated at 30°C at 150 rpm shaking until an OD₆₀₀ of 0.3-0.5 was reached. Three samples were taken for timepoint 0 and the culture was then split into 10 ml aliquots for treatments and controls. Flasks were treated with either 80 µM, 60 µM or 40 µM of compound. DMSO was used for untreated controls and DMSO content was 1% for all flasks. Three replicate flasks were done for each treatment and controls. Flasks were incubated at 30°C at 150 rpm shaking for 2 h. For plating, samples were taken, serially diluted in PBS and plated on YPD agar plates after 0, 30 and 60 min and incubated at 30°C for 2 days for colony count determination. Data was plotted as log CFU/ml for each time point.

Cell death was assessed by propidium iodide (PI) staining. 1.25 ml of culture were sampled at the same time points as for plating. Samples were centrifuged at 10,000 g for 1 min and supernatant was removed. Cells were then resuspended in 500 µl PI staining solution containing 100 ng/ml of PI (Sigma, Germany) in PBS pH 7.4 and incubated for 15 min in the dark. Cells were then centrifuged again, the supernatant was removed and cells were resuspended in 250 µl of PBS. 200 µl of cell suspensions were transferred into 96-well U-bottom plates (Greiner Bio-One, Austria) and analyzed by flow cytometry (LSRII Fortessa, BD). Fluorescence (excitation: 561 nm, emission: 670/30 nm) was quantified at the single cell level, for each sample, 30,000 cells were analyzed with BD FACSDiva software.

In vivo testing of LIA in the C. elegans infection model was done as described by Breger et al. (Breger et al., 2007) with some modifications. To synchronize nematodes, eggs of the C. elegans Δglp-4 Δsek-1 strain were collected by bleaching as described by Porta-de-la-Riva et al. (Porta-de-la-Riva et al., 2012), placed on NGM (nematode growth medium) plates (2.5 g/L Bacto Peptone, 3 g/L NaCl, 17 g/L agar, supplemented with 5 mg/L cholesterol, 1 mM CaCl₂, 1 mM MgSO₄ and 25 mM KPO₄ buffer pH 6.0) agar plates seeded with Escherichia coli OP50 and incubated. Nematodes were then grown to the L3/L4 larvae stage and infected by feeding for 2 h on YPD agar plates seeded with C. albicans SC5314. After infection, worms were washed several times with ddH₂O and were finally resuspended in growth medium (3 g/L KH₂PO₄, 6 g/L Na₂HPO₄, 5 g/L NaCl, supplemented with 1.25 mM MgSO₄ 10 µg/mL cholesterol, 100 µg/mL kanamycin and 75 µg/mL ampicillin) to a final concentration of approximately 40 worms/1.5 mL. In parallel, larvae of the non-infected control were fed on NGM plates seeded with E. coli OP50 and underwent the same procedure as infected nematodes. 1.5 mL of the worm suspension were transferred into each well of a 24-well plate (Greiner Bio One, Austria), nematodes were immediately counted (t0) and treated with either LIA, the solvent DMSO (0.2%) or remained untreated (uninfected control). Incubation and infection were done at 25°C. Infected and uninfected nematodes were monitored daily with a stereomicroscope and living worms were counted over a period of five days post infection. Number of living nematodes at t0 was set to 100% and survival at each time point was expressed compared to t0.

The number of independent experiments (n) is indicated in the figure legends of the corresponding graphs. Each independent experiment was performed with a separate clone in at least three technical replicates. The results of all technical replicates were averaged for each experiment. For human cell culture and C. elegans, (n) is defined as independent experimental runs.

Overall normal distribution of data was analyzed by Shapiro-Wilk test and/or visual inspection of QQ-plots of residuals. Homogeneity of variance was tested using the Brown-Forsythe test. For normally distributed data, comparisons between two or multiple groups were done via analysis of variance (ANOVA). Non-normally distributed data were tested by the non-parametric Kruskal-Wallis test. P-values for the human cell viability assay and the C. elegans infection model were calculated using a two-way ANOVA approach, followed by the Benjamini-Hochberg procedure for multiple comparisons. C. elegans data were matched by time point, sphericity was not assumed and the Geisser-Greenhouse correction was used. Non-parametric data from time-kill assays were analyzed using a Kruskal-Wallis test followed by the two-stage step-up method of Benjamini, Krieger and Yekutieli. Acute toxicity data were analyzed within each timepoint by one-way ANOVA including linear trends for normally distributed data. Non-normally distributed data were examined by Kruskal-Wallis test. All statistical analyses were performed using GraphPad Prism 8 software.

Given the understudied yet promising antifungal potential of lichen-derived compounds (LDCs), we conducted a screen of 68 secondary metabolites, either directly isolated from the native lichen or synthesized based on known structural motifs (Supplementary Table 1). The screen was carried out in two clinically relevant species that differ significantly in virulence traits and resistance profiles: C. albicans, the major source of candidiasis, and N. glabratus, as a representative of NAC species. To evaluate antifungal efficacy, we (i) analyzed possible growth-inhibitory effects of LDCs in planktonic cells by measuring OD₄₉₀ 24 h and 48 ± 1 h after LDC treatment. Biofilm formation is a strong determinant of fungal virulence and a driver of resistance. Thus, (ii) we parallelly assessed potential activity of the 68 LDCs against biofilm formation. For that purpose, we quantified metabolic activity under biofilm-inducing conditions 24 h after challenge with LDCs. All LDC treatments were performed in serial dilutions and compared with untreated controls. For these serial dilutions, the highest concentration was set at 250 µM; for compounds that were not completely soluble at the necessary stock concentration, the dilutions were started at lower concentrations (Figure 1, Supplementary Table 1).

The combined approach determining the impact on both growth and biofilm formation allowed for the evaluation of a broader antifungal potential. Accordingly, we defined our hits as such that would act against both planktonic cells and biofilms, yielding at least 50% inhibition in any of the tested concentrations in either C. albicans or N. glabratus (Figure 1). We set this cutoff, first, for comparability to published IC₅₀ values and, second, following signal-to-noise considerations, i.e. a threshold stringent enough to minimize false positives and sufficiently sensitive to capture compounds with moderate but genuine activity that could be optimized through medicinal chemistry. Of note, previous screening campaigns have also specifically used 50% inhibition as their hit cutoff (Vila and Lopez-Ribot, 2016). Applying these criteria, our study yielded a total of 12 different hit LDCs. Five compounds (nos. 14, 23, 34, 37, 48) exhibited double inhibitory effects against planktonic growth and biofilm formation in C. albicans and ten (nos. 14, 21, 22, 34, 37, 47, 51, 56, 59, 60) in N. glabratus.

To confirm and further investigate the findings from the initial screen (Figure 1), growth and biofilm inhibition assays were conducted for all hit compounds using a low-throughput approach to generate dose-response curves. From these assays, we determined the minimum concentration required to inhibit 50% of growth (MIC₅₀) and biofilm formation (BIC₅₀) for C. albicans and N. glabratus, respectively (Supplementary Table 2). One LDC (no. 56) did not meet the established inhibition threshold of 50% for any assay and could thus not be validated as a hit. For four LDCs (nos. 14, 34, 37 and 48), we observed inhibition of planktonic growth of C. albicans below 50% (Figure 2A) with MIC₅₀ values ranging between 65.1 µM and 239.0 µM after 24 h. One LDC (no. 23) reduced growth below 50% only after 48 h (Supplementary Figure 1). For nine LDCs (nos. 14, 21, 34, 37, 47, 48, 51, 59 and 60), we observed inhibition of planktonic growth of N. glabratus below 50% (Figure 2B) with MIC₅₀ values between 5.9 µM and 177.9 µM after 24 h. Inhibition of C. albicans biofilm formation (Figure 2C) below 50% was observed for six LDCs (nos. 14, 22, 34, 37, 48 and 59) with BIC₅₀ values between 5.7 µM and 115.6 µM, while formation of N. glabratus biofilms (Figure 2D) below 50% was observed upon treatment with four LDCs (nos. 14, 34, 37 and 59), obtaining BIC₅₀ values between 45.9 µM and 153.5 µM.

Notably, the dibenzofurane (-)-isousnic acid (no. 34) and the two tested paraconic acids, (+)-lichesterinic acid (LIA, no. 37) and (-)-allo-proto-lichesterinic acid (aPLA, no. 14), were the only compounds that inhibited both growth and biofilm formation in both C. albicans and N. glabratus. Owing to their broad-spectrum activity, we prioritized these compounds as our top hits. While isousnic acid has already been associated with antifungal activity in previous studies (Goel et al., 2011a), LIA and aPLA remain poorly characterized in this context. Thus, we decided to continue our analyses only with LIA and aPLA (Supplementary Figure 2).

In order to evaluate the potential of LIA and aPLA for therapeutic applications, we first investigated their possible toxicity against human cells. In order to more accurately mimic the behavior of primary human cells, we chose non-immortalized cells, specifically Normal Human Dermal Fibroblasts (NHDF). After treatment with LIA and aPLA for 24 h, we assessed viability using a 2D-cell viability assay, which quantifies ATP to indicate the presence of metabolically active cells (Figure 3A). LIA was more tolerable than aPLA with an IC₅₀ of 104.8 ± 7.5 µM (in comparison to 51.6 ± 9.9 µM for aPLA). Given the better overall tolerability of LIA, we focused our further efforts on this paraconic acid.

We next assessed whether the antifungal effects of LIA derive from growth inhibition (fungistatic effects) or death induction (fungicidal effects). This distinction is crucial from a therapeutic point of view, since especially immunocompromised patients may not effectively eliminate fungal pathogens with fungistatic therapy alone. For that purpose, we conducted time-kill-assays in C. albicans, thus exposing cells to different concentrations of LIA, and counting viable colony-forming units (CFUs) at multiple time points during 24 h to assess changes in fungal burden (Figures 3B, C). While 40 µM did not significantly reduce viable cells over time, treatments with 80 µM and 160 µM resulted in complete killing of C. albicans. We also controlled for antifungal carryover, i.e. the residual presence of LIA upon CFU plating, which could have artificially suppressed fungal growth. However, we could not detect any carryover for any of the concentrations tested, suggesting a bona fide fungicidal effect of LIA (Supplementary Figure 3).

In addition to the standard time-kill experiments, we assessed the onset of antifungal activity in a logarithmically growing C. albicans culture by performing acute toxicity assays. Cultures were treated with LIA at concentrations of 40 µM, 60 µM, or 80 µM (corresponding to LIA’s MIC₅₀). Higher concentrations were excluded as they eliminated all viable colonies in time-kill assays. Effects were assessed at 30- and 60 minutes post-treatment using two complementary methods: propidium iodide (PI) staining to detect dead cells (Figure 3D) and CFU counts normalized to DMSO-treated controls to evaluate viability (Figure 3E). Treatment with LIA resulted in a dose-dependent increase (P-value for trend = <0.0001 for 30 min and <0.0001 for 60 min) in PI-positive cells that was accompanied by a corresponding dose-dependent decrease in cell viability (P-value for trend = 0.0012 for 30 min; trend could not be assessed for 60 min as data was not normally distributed), suggesting strong effectivity in rapidly growing C. albicans cultures (Figure 3).

Biofilms remain a key challenge in antifungal therapy (Fan et al., 2022). As already assessed (Figure 2, Supplementary Table 2), LIA is able to inhibit biofilm formation in C. albicans (BIC₅₀ 22.3 ± 4.1 µM) and N. glabratus (BIC₅₀ 90.2 ± 26.0 µM), a crucial step for fungal colonization and persistence. Of note, this is also the case in two other non-albicans species, C. dubliniensis and P. kudriavzevii, in which LIA not only inhibits planktonic growth (Supplementary Figure 4) but also biofilm formation. However, the eradication of established biofilms represents a different major obstacle in effective antifungal treatment, since mature biofilms exhibit increased resistance to antifungal agents and protect embedded cells from host immune responses. Thus, we investigated whether LIA is able to eliminate pre-grown C. albicans, N. glabratus, C. dubliniensis and P. kudriavzevii biofilms by assessing mature biofilm metabolic activity. LIA was able to effectively eradicate biofilms of all tested pathogenic fungi with BEC₅₀ values of 86.3 ± 26.8 μM in C. albicans (Figure 4A), 149.2 ± 25.6 µM in N. glabratus (Figure 4B), 189.7 ± 13.7 µM in C. dubliniensis (Supplementary Figure 4C) and 188.9 ± 12.3 µM in P. kudriavzevii (Supplementary Figure 4F), respectively. Altogether, LIA is effective in both preventing biofilm formation and disrupting mature biofilms.

In C. albicans, filamentation processes involving the formation of true hyphae or robust pseudohyphae play a major role in biofilm formation and virulence (in contrast, N. glabratus does not undergo true filamentation under standard conditions) (Galocha et al., 2019). Thus, we investigated the effects of LIA on C. albicans filamentation using different media. Colonies were grown on standard RPMI or filamentation-inducing (Spider and Spider D) agar plates supplemented with either 25 µM or 50 µM LIA (corresponding to the 1x and 2x BIC₅₀ values observed in vitro), and filamentation was monitored over 7 days; DMSO-only plates served as controls. On RPMI plates, LIA demonstrated dose-dependent inhibition of filamentation with the strongest effects apparent at day 3 (Figure 4C; Supplementary Figure 5). Although filamentation increased by day 7 in treated samples, clear differences between treatments and controls persisted. Spider plates revealed more dramatic inhibitory effects, with both 25 µM and 50 µM LIA treatments completely preventing radial filamentation at day 3 and allowing only for minimal filamentation at later days (Figure 4C; Supplementary Figure 5). These inhibitory effects were even further exacerbated in Spider D plates (Figure 4C; Supplementary Figure 5), in which mannitol is replaced by a lower concentration of glucose. These results provide a qualitative demonstration that LIA effectively inhibits C. albicans filamentation in a dose- and media-dependent manner, with the strongest effects observed under conditions where filamentation is actively promoted.

Following our combined in vitro results, we finally decided to evaluate the antifungal efficacy of LIA in vivo. For that purpose, we employed the nematode Caenorhabditis elegans, which has been proven to be a valuable infection model, for instance, for the identification of virulence factors and the evaluation of antifungal agents (Breger et al., 2007; Delattin et al., 2014). Preliminary testing revealed that LIA concentrations of 50 µM and above were toxic to C. elegans (Supplementary Figure 6), thus we proceeded with lower concentrations. In brief, worms synchronized in late larval stage were infected with C. albicans and treated with LIA. Animals were monitored daily over five days, with live and dead animals scored under a stereomicroscope. After 5 days of incubation, uninfected control worms showed 87.9% mean survival, while infected, untreated worms exhibited only 12.3% survival. Treatment with 33.3 µM LIA significantly increased survival rates to 38% (Figure 5). These results demonstrate that LIA provides protective effects against C. albicans infection at concentrations tolerated by the host organism.