Effect of the Novel Antifungal Drug F901318 (Olorofim) on Growth and Viability of Aspergillus fumigatus

F901318 (olorofim) is a novel antifungal drug that is highly active against Aspergillus species. Belonging to a new class of antifungals called the orotomides, F901318 targets dihydroorotate dehydrogenase (DHODH) in the de novo pyrimidine biosynthesis pathway.

ABSTRACT

F901318 (olorofim) is a novel antifungal drug that is highly active against Aspergillus species. Belonging to a new class of antifungals called the orotomides, F901318 targets dihydroorotate dehydrogenase (DHODH) in the de novo pyrimidine biosynthesis pathway. In this study, the antifungal effects of F901318 against Aspergillus fumigatus were investigated. Live cell imaging revealed that, at a concentration of 0.1 μg/ml, F901318 completely inhibited germination, but conidia continued to expand by isotropic growth for >120 h. When this low F901318 concentration was applied to germlings or vegetative hyphae, their elongation was completely inhibited within 10 h. Staining with the fluorescent viability dye bis-(1,3-dibutylbarbituric acid) trimethine oxonol (DiBAC) showed that prolonged exposure to F901318 (>24 h) led to vegetative hyphal swelling and a decrease in hyphal viability through cell lysis. The time-dependent killing of F901318 was further confirmed by measuring the fungal biomass and growth rate in liquid culture. The ability of hyphal growth to recover in drug-free medium after 24 h of exposure to F901318 was strongly impaired compared to that of the untreated control. A longer treatment of 48 h further improved the antifungal effect of F901318. Together, the results of this study indicate that F901318 initially has a fungistatic effect on Aspergillus isolates by inhibiting germination and growth, but prolonged exposure is fungicidal through hyphal swelling followed by cell lysis.

INTRODUCTION

F901318 (olorofim) is a new class of antifungal drug that acts through a novel mechanism and is currently in phase II clinical development. It is highly active against all pathogenic Aspergillus species, with a MIC of <0.1 μg/ml (1–3). F901318 is the first drug candidate from the new orotomide class of antifungals. Its cellular target is dihydroorotate dehydrogenase (DHODH), the fourth enzyme in the de novo pyrimidine biosynthesis pathway (4). Although DHODH is also found in mammals, F901318 is not active against human DHODH (2). Pyrimidine biosynthesis is vital for many cellular processes, including DNA/RNA synthesis (the nucleobases cytosine, thymine, and uracil), the cell cycle (DNA), protein synthesis (RNA), cell wall synthesis (via UTP, which forms UDP-glucose), and phospholipid synthesis (via cytosine triphosphate [CTP]) (5).

The aim of this study was to investigate the antifungal effects of F901318 against the infective conidial form and the invasive hyphal form of Aspergillus fumigatus. Antifungal compounds may be described as being either fungistatic or fungicidal; however, it is frequently difficult to assign fungicidality to a particular drug, especially for fungi that produce hyphae. Over the years, many methods have been developed to determine the fungicidality of antifungal compounds (reviewed in reference 6). A common test is the determination of the minimal fungicidal concentration (MFC), where the clear microplate wells following antifungal susceptibility testing are subcultured and colonies are counted. The lowest concentration with a >99.9% growth reduction is the MFC. More recently, fluorescent viability staining has been used to determine the fungicidality of antifungal compounds (7–9).

The outcome of fungicidality testing can depend on the species of fungus tested, the developmental stage investigated, the concentration of drug, and the exposure time (10). Of the antifungal drug classes in clinical use, amphotericin B is usually considered to be a fungicidal drug, whereas azoles and echinocandins are either fungicidal or fungistatic, depending on the species. Voriconazole, for example, is fungistatic against Candida species but shows fungicidal activity against Aspergillus species, although this is time dependent (11, 12). The three marketed echinocandins have been widely considered fungicidal against Candida species but fungistatic against Aspergillus species. However, recently it has been shown that the effects of the echinocandin caspofungin on A. fumigatus is more complicated than previously appreciated (13, 14). Caspofungin causes hyphal hyperbranching coupled with repeated hyphal tip lysis followed by regenerative intrahyphal growth. This results in very compact colonies which continue to expand slowly. At high caspofungin concentrations after 40 h of exposure to the drug, A. fumigatus, intriguingly, undergoes paradoxical growth (reversal of growth inhibition) by some unknown mechanism (14).

MIC and MFC tests are typically carried out using conidia, even though it has been shown that antifungals can have a different effect against ungerminated conidia, germinated conidia with germ tubes (germlings), or hyphae (15). To examine the fungicidality of F901318 toward the three developmental stages of A. fumigatus, time-lapse microscopy, live cell imaging, and viability staining were employed. Additionally, key data were confirmed in a second species (A. flavus), and the biomass and growth rate of F901318-treated hyphae shifted to drug-free medium were assessed. Together, the results of this investigation demonstrated that F901318 has a time-dependent killing effect on Aspergillus spp., the extent of which is dependent on the developmental stage at the time of exposure.

RESULTS

F901318 inhibits germination, but not isotropic growth, of A. fumigatus conidia.

To analyze the effects of 0.1 μg/ml F901318 (approximately 2× MIC) on germination, A. fumigatus conidia were treated with the drug for up to 5 days and observed under a microscope every 24 h (Fig. 1A). Within 24 h, untreated conidia had grown isotropically from an initial diameter of 3 μm to a diameter of 8.7 μm with 100% germination. The F901318-treated conidia did not germinate but continued to grow in diameter at a linear rate of ∼1.5 μm/day. At day 5, germination was still not observed and the conidial diameter had increased to 10.5 μm, significantly larger than the 8.7-μm diameter of the germinated control conidia (P < 0.05; Fig. 1B). After exposure to F901318 for 8 days, there had been further isotropic growth of the conidia, without germination (data not shown).

FIG 1.

Conidia undergo isotropic growth but do not germinate when exposed to 0.1 μg/ml F901318. Conidia were imaged, and their diameters were measured 24, 48, 72, 96, and 120 h after addition of 0.1 μg/ml F901318. (A) DIC images of untreated and treated conidia over time. Bar = 5 μm. (B) Conidial diameters of treated versus untreated conidia over time. Error bars represent SD (n = 3). (C) TEM images of an untreated conidium and 24-h-treated conidium. Bar = 500 nm.

Transmission electron microscopy (TEM) of sections of freshly harvested ungerminated conidia and conidia treated for 24 h showed that conidia treated with F901318 were larger than fresh ungerminated conidia (Fig. 1C), consistent with previous measurements (Fig. 1A and B), and contained highly enlarged vacuoles.

F901318 inhibits polarized hyphal growth.

Time-lapse, live cell imaging was used to study the effects of 0.1 μg/ml F901318 applied for 16 h to germlings and mature vegetative hyphae of A. fumigatus (Fig. 2A). Elongation in both germ tubes and hyphae slowed down immediately after addition of F901318. Following this initial lag, growth partially recovered but the rate of elongation was much reduced compared to the growth rate in the absence of drug. Elongation of treated germ tubes was 6-fold slower than that of untreated germ tubes, and elongation of mature vegetative hyphae was reduced by 12-fold compared to that of untreated hyphae. Growth ceased completely 10 h after addition of F901318 in both growth forms (Fig. 2B).

FIG 2.

F901318 inhibits hyphal elongation in growing germ tubes and hyphae. Sixteen-hour time-lapse sequences of germlings and vegetative hyphae treated with 0.1 μg/ml F901318 were captured. (A) Images from representative time points from the 16-h time-lapse sequences of germlings and vegetative hyphae following treatment with F901318 are shown. Bar = 10 μm. (B) Hyphal elongation of germ tubes and vegetative hyphae after addition of F901318. Error bars represent SD (n = 3).

F901318 causes isotropic growth and lysis of A. fumigatus hyphae.

To investigate the effects of prolonged F901318 exposure, germlings exposed to 0.1 μg/ml F901318 were observed over a period of 46 h. The images in Fig. 3 represent different stages of a representative time-lapse sequence (see Movie S1 in the supplemental material). In the first 8 h after adding the drug, the germ tube width increased from approximately 4.5 μm to 5 to 7 μm. After 16 h, the first septum had formed and a second formed after 32 h, dividing the germling into three compartments (Fig. 3, black arrows). In the time-lapse sequence shown, one hyphal compartment continued to swell, finally lysing to release its cellular contents 34 h after the addition of the drug (Fig. 3, white arrows). Lysis of hyphal compartments occurred at various times after the initiation of treatment, with the earliest occurring at 24 h. The occurrence of lysis may indicate significant cell wall disorganization and a site of weakening in the cell wall within this hyphal compartment. In Fig. 3, the apical germ tube compartment had undergone considerable isotropic growth by the 46-h time point (Fig. 3, white arrow surrounded by black line). These observations suggest that prolonged F901318 exposure leads to the dysregulation of normal polarized growth and cell wall deposition or cross-linking, accompanied by an inability to maintain cell wall integrity. Inhibition of polarized growth, swelling, and lysis was also observed in A. fumigatus clinical isolate AF210 and an isolate of A. flavus, indicating that those effects are not specific to AF293 (Fig. S1).

FIG 3.

F901318 causes hyphal swelling and cell lysis. Images from representative time points of one of the 46-h time-lapse sequences of germlings exposed to 0.1 μg/ml F901318. Arrows highlight the formation of two septa (black arrows), a lysed compartment and the consequent release of cellular contents (white arrows), and an apical swollen compartment (white arrow surrounded by black line). Bar = 10 μm.

TEM imaging of hyphae treated with F901318 for 24 h showed hyphal swelling (Fig. 4A) and highly enlarged vacuoles (Fig. 4B). Ruptured hyphal compartments lacking normal components and possessing broken cell walls were evident following cell lysis in drug-treated samples (Fig. 4C).

FIG 4.

(A) TEM image of an untreated hypha and a hypha treated for 24 h with 0.1 μg/ml F901318. Images show the increased diameter of the treated hypha. Bar = 0.5 μm. (B) TEM image showing enlarged vacuoles in a hypha treated for 24 h with 0.1 μg/ml F901318. Bar = 2 μm. (C) TEM image showing ruptured cell walls in a hypha treated for 24 h with 0.1 μg/ml F901318. Bar = 2 μm.

F901318 kills A. fumigatus in a time-dependent manner.

As cell lysis was observed with prolonged treatment with 0.1 μg/ml F901318, the viability of A. fumigatus after exposure to F901318 was determined by staining with the fluorescent viability stain bis-(1,3-dibutylbarbituric acid) trimethine oxonol (DiBAC), which is an oxonol dye. Only dead cells are permeable to DiBAC, which fluoresces on binding to intracellular phospholipids (16). No staining was observed in untreated cells, which are impermeable to this dye.

Conidia, germlings, and hyphae were treated with F901318 for up to 120 h, and viability was measured every 24 h. Each of these three developmental stages lost viability at different rates (Fig. 5). Germlings and vegetative hyphae were found to be more sensitive to drug treatment than conidia. Reduced viability was observed in germlings and vegetative hyphae after 24 h and in conidia after 48 h. After 120 h, 95% of the germlings were dead, while only 9.4% of conidia had lost viability at this stage. In comparison, 100% of germlings treated with 4 μg/ml amphotericin B for 24 h were stained by DiBAC (Fig. S2), in contrast to the time-dependent mechanism of killing observed for F901318. Microscopic examination showed that exposure to F901318 caused cell lysis of apparently random individual apical and subapical compartments within a hypha. Due to the difficulty of direct quantification of lysed compartments, the total area of individual hyphae that exhibited DiBAC fluorescence in an image was determined. The area of dead hyphal compartments increased 25-fold from 77 μm² to 1,922 μm² between the 24-h and 120-h time points. Together, these results suggest that the disruption of hyphal growth and the homeostatic machinery by F901318 cannot be mitigated by A. fumigatus; it leads not only to a failure in growth but also to cell death.

FIG 5.

F901318 causes time-dependent cell lysis. (Top) The viability of conidia, germlings, and hyphae treated with F901318 was quantified using DiBAC staining. The images show a representative conidium, germling, and hypha stained with DiBAC. Bar = 10 μm. (Bottom) Histograms show the quantification of DiBAC staining (viability) of conidia, germlings, and hyphae treated with F901318 over time. For conidia and germlings, the percentage of DiBAC-stained conidia and germlings was quantified. For hyphae, the area of DiBAC fluorescence in the images was quantified. Error bars represent SD (n = 3).

A. fumigatus hyphae recover poorly following F901318 exposure.

We next investigated whether the remaining drug-treated hyphal compartments that had not lysed retained the ability to regrow once exposure to F901318 had ceased. After 24 h of treatment with 0.1 μg/ml or 1 μg/ml F901318, hyphae were washed and transferred into drug-free growth medium for 48 h and their dry weights were ascertained (Fig. 6A). Regrowth of hyphae exposed to 0.1 μg/ml F901318 was reduced 82-fold and was almost completely inhibited when treated with 1.0 μg/ml F901318. The results suggest that A. fumigatus cannot substantially recover from the morphological defects induced by F901318 and that the severity of these defects is dependent on the concentration of F901318.

FIG 6.

Prolonged exposure to F901318 improves its antifungal effect. (A) Postexposure recovery biomass assay of hyphae exposed to 0.1 μg/ml or 1.0 μg/ml F901318. Histograms show the reduction in dry weights of mycelia after being shifted to drug-free medium for 48 h. A higher concentration improved the antifungal effect of F901318. Error bars represent SD (n = 3). (B) Postexposure recovery growth rate assay of hyphae treated with F901318, voriconazole, caspofungin, and amphotericin B for 24 or 48 h before being shifted to drug-free medium. Graphs show the means from 3 determinations of growth recovery, which was measured by reading the OD at 405 nm on a plate reader every 10 min over 24 h. Prolonged exposure to F901318 improved its antifungal effect.

Prolonged exposure to F901318 improves its antifungal effect.

As staining with DiBAC had shown that longer treatment with F901318 leads to a greater decrease in viability, the growth recovery of hyphae treated with F901318 for 24 h and 48 h was compared with the ability of A. fumigatus to recover from exposure to other commonly used antifungal drugs. Hyphae were treated with 0.1 μg/ml F901318, 0.5 μg/ml caspofungin, 1 μg/ml voriconazole, or 4 μg/ml amphotericin B as comparators. After treatment for 24 h or 48 h, hyphae were washed to remove drug residues and transferred to Sabouraud (SAB) medium containing 10 μM uridine and 10 μM uracil (to mimic the pyrimidine concentrations in human blood). The optical density at 405 nm (OD₄₀₅) was measured every 10 min for 24 h on a plate reader.

As shown previously (17), regrowth following treatment with amphotericin B was almost undetectable, independent of the duration of treatment (Fig. 6B). After 24 h of treatment, the F901318-, caspofungin-, and voriconazole-treated hyphae produced 74%, 69%, and 89% of the biomass produced by the control, respectively (Fig. 6B, left). After 48 h of treatment, this was reduced to 24%, 63%, and 65%, respectively (Fig. 6B, right). These results show that prolonged treatment with these three antifungals increased their antifungal effect. The 48-h treatment led to a 1.06-fold increase in the antifungal effect of caspofungin and a 1.37-fold increase in the antifungal effect of voriconazole compared to the effects of the 24-h treatments. An even stronger effect was observed for the 48-h F901318 treatment, with a 3.09-fold increase of its antifungal effect compared to that of the 24-h treatment. This result demonstrates that prolonged exposure increased the detrimental effect of F901318 on A. fumigatus hyphae more so than caspofungin and voriconazole.

DISCUSSION

In this study, the inhibitory effects of the novel antifungal drug F901318 on the growth and morphogenesis of different developmental stages (conidia, germlings, and vegetative hyphae) were studied in the human-pathogenic fungus A. fumigatus. Additionally, the ability of hyphae to recover after treatment was investigated.

Conidia exposed to F901318 were not able to germinate, but they continued to grow isotropically (Fig. 1). This result is similar to that reported in a previous study by D'Enfert et al. (18), in which a pyrimidine biosynthesis mutant of A. fumigatus (pyrG⁻) was observed to swell isotropically, without germinating. This is an important observation because the hyphal morphology is the penetrative form in invasive fungal infections, including invasive aspergillosis (19). Furthermore, as both F901318-treated conidia and pyrG⁻ conidia are disrupted in the de novo pyrimidine biosynthesis pathway, this supports the view that pyrimidine synthesis is vital for germination. However, it also indicates that the de novo synthesis of pyrimidines is not required for conidial isotropic growth. This highlights a possible fundamental difference between this process and hyphal tip growth, suggesting a differential mode of cell wall synthesis.

During germination, several metabolic processes are upregulated to allow conidia to transition from a dormant to an active state, which is followed by the emergence of a germ tube and initiation of polarized hyphal growth (20–23). Proliferating cells require active de novo pyrimidine biosynthesis to meet the increased demand for pyrimidines for DNA/RNA synthesis, cell cycle regulation, protein synthesis, cell wall synthesis, and phospholipid synthesis (24–27). When actively growing germ tubes and vegetative hyphae were exposed to F901318 for more than 10 h, tip growth was inhibited (Fig. 2). Therefore, in conidia, germlings, and vegetative hyphae of A. fumigatus, it seems that polarized hyphal growth cannot be initiated or maintained when insufficient pyrimidines are present. In vivo infection studies with A. fumigatus demonstrated that F901318 significantly improves survival of the host, which indicates that hyphae cannot scavenge sufficient pyrimidines from the host (2).

Treatment of germlings or vegetative hyphae for less than 10 h resulted in a fungistatic effect of hyphal growth arrest. Prolonged exposure to F901318 caused excessive isotropic expansion of conidia (Fig. 1) and of individual hyphal compartments (Fig. 2A), which subsequently underwent cell lysis (Fig. 3). DiBAC staining demonstrated that extended exposure to F901318 led to a decreased viability of conidia, germlings, and vegetative hyphae, indicating a time-dependent killing effect of F901318 (Fig. 5). The extent of the killing effect was dependent on the developmental stage of A. fumigatus. Only 9.4% of conidia were killed after 5 days, showing that F901318 had mostly a fungistatic effect on these spores. This is in line with our unpublished data from MFC tests, also carried out with conidia, in which F901318 was concluded to have fungistatic effects. This could be due to the inhibition of germination and the consequent reduced requirement for pyrimidines, allowing the conidia to survive for longer in the presence of the drug. Nevertheless, the isotropic growth of conidia continued, which is interesting, as inhibition of DHODH would be expected to lead to a reduced availability of UDP-sugars, which are essential during hyphal cell wall synthesis. This suggests that there are sufficient stores of pyrimidines to allow this growth but insufficient stores to support germination or that the UDP-glucose requirements of the conidial cell wall are altered. The inhibition of conidial germination could imply a role for F901318 as a prophylactic drug by preventing conidia from germinating in the lungs of at-risk patient populations. The number of DiBAC-stained germlings and hyphae increased more dramatically with time, demonstrating that F901318 has a stronger effect on these developmental stages. It seems likely that germlings and vegetative hyphae have a store of pyrimidines, allowing them to survive until this store is exhausted, after which the effects of pyrimidine starvation from F901318 exposure impact cell viability. This could explain the time-dependent killing effect of F901318 on germlings and vegetative hyphae.

The DiBAC staining results showed that only a small decrease in viability was observed 24 h after F901318 exposure. A biomass assay was performed to study whether 24-h-treated hyphae could recover from F901318 treatment when placed in drug-free medium. The results also showed that growth recovery is strongly impaired after treatment with 0.1 μg/ml F901318 and even more so after treatment with 1.0 μg/ml (Fig. 6A). So, even though the viability was only slightly decreased after treatment for 24 h, as measured by DiBAC staining, the hyphae were not able to fully recover from F901318 treatment, suggesting that the antifungal effect of F901318 is mostly irreversible. A postexposure recovery growth rate assay showed that prolonged F901318 treatment increased the antifungal effect of F901318 (Fig. 6B). Regrowth of hyphae treated with F901318 for 48 h was reduced 3.09-fold compared to the regrowth of 24-h-treated hyphae. In line with the DiBAC staining results, this again demonstrates that F901318 exposure decreases the viability of A. fumigatus in a time-dependent manner.

F901318 causes A. fumigatus hyphal compartments to swell, eventually leading to cell lysis. Hyphal swelling was also observed in a study in which A. nidulans hyphae were exposed to the DNA synthesis inhibitor hydroxyurea (22), suggesting that swelling could be a direct result of inhibition of DNA synthesis and/or downstream global effects on cell metabolism and regulation. Hyphal swelling and lysis have also been reported in caspofungin-treated A. fumigatus hyphae. However, the inhibitory effects of caspofungin cause repeated hyphal tip lysis followed by regenerative intrahyphal growth (14), whereas the lysis caused by F901318 is random and not hyphal tip specific (Fig. 3) and intrahyphal growth was not observed. Interestingly, the continuous presence of caspofungin promoted excessive isotropic growth of ungerminated conidia and a delay in germination (S. D. Moreno-Velásquez and N. D. Read, unpublished data). Caspofungin belongs to the echinocandins, and drugs in this class inhibit 1,3-β-glucan synthase, which is required for the formation of β-glucans in the fungal cell wall. One of the products of the pyrimidine synthesis pathway is UTP, which is required for the formation of UDP-sugars, which are the substrates for the formation of the fungal cell wall polysaccharides chitin and 1,3-β-glucans (5, 28). Inhibition of pyrimidine synthesis could therefore have downstream effects on the structure of the cell wall, ultimately resulting in cell lysis. Ruptured cell walls were frequently observed under a confocal microscope (Fig. 5) and in TEM images of F901318-treated vegetative hyphae (Fig. 4D). TEM also revealed apparently enlarged vacuoles in treated conidia (Fig. 1C) and treated vegetative hyphae (Fig. 4C). Vacuolar size has been linked to the cell cycle in Candida albicans (29), so it is possible that the larger vacuoles are a result of the cell cycle disruption caused by inhibition of pyrimidine synthesis. It is likely that the formation of these large vacuoles plays a role in the swelling that was observed in conidia, germlings, and hyphae exposed to F901318 in combination with the isotropic growth that occurs. The vacuolar swelling may be due to an increased osmotic pressure that facilitates the lysis of the cell wall that has become weakened by F901318 treatment due to being starved of UDP-sugar precursors. Further work is required to investigate the effect of F901318 on the relationship between the formation of these large vacuoles, the osmotic homeostasis in the treated cells, and the lysis of their cell walls.

The results of this study have demonstrated that although F901318 has a mostly fungistatic effect on conidia, it has more dramatic effects on germlings and hyphae. Prolonged exposure to the drug causes hyphal swelling and isotropic growth, which lead to cell lysis, releasing cellular contents. It was therefore concluded that F901318 has a time-dependent killing effect against A. fumigatus. This is consistent with the effects of F901318 treatment seen in various animal models of fungal infection (2, 3, 30). Hope et al. investigated the pharmacokinetics/pharmacodynamics of F901318 in rodent models of aspergillosis and concluded that there was a time-dependent element to the antifungal activity (3). Recently, Wiederhold et al. observed sterilization of the brain in a virulent central nervous system model of Coccidioides infection (31). This study demonstrates that superficial observations of the effects of antifungal compounds do not always provide a straightforward distinction between whether the drug has fungistatic or fungicidal effects. Using classical methods, F901318 would usually be regarded as a fungistatic agent. However, we have shown in this study by using (time-lapse) microscopy that prolonged exposure to F901318 inhibits conidial germination while resulting in lysis of hyphal cells. These results indicate that F901318 exhibits fungicidal activity against the Aspergillus species (A. fumigatus and A. flavus) studied in our work to date.

MATERIALS AND METHODS

Strain and cultures.

Aspergillus fumigatus clinical isolate AF293 was used throughout this study. Where indicated, A. fumigatus clinical isolate AF210 and A. flavus clinical isolate AFl01 were compared to AF293. Conidia were incubated in standard Vogel's minimal medium (VMM) (32), Sabouraud (SAB) broth (BD Difco), or SAB dextrose agar (Oxoid). The inoculum concentration varied from 10³ to 10⁷ conidia/ml (see the relevant section). For microscopy, cultures were grown in VMM at 37°C in uncoated μ-Slide 8-well chambers (Ibidi).

Antifungal drugs.

F901318 (olorofim) was synthesized as previously described (2). Caspofungin, voriconazole, and amphotericin B were acquired from Sigma-Aldrich. Stock solutions were prepared in dimethyl sulfoxide (DMSO; Sigma-Aldrich) and diluted to the required concentration in the growth media.

Conidial diameter.

Freshly harvested conidia were exposed to 0.1 μg/ml F901318 (approximately 2× MIC) or 0.02% DMSO as a control. After 24, 48, 72, 96, and 120 h of exposure, bright-field and differential interference contrast (DIC) images of the treated and untreated conidia were acquired on a Nikon Eclipse TE2000-E wide-field microscope, using a 60× water immersion objective. A Hamamatsu Orca-ER C4742-80 camera (Hamamatsu Photonics UK Ltd.) with MetaMorph software (Molecular Devices) was used for image acquisition. The diameter of 50 conidia was measured at each time point. The experiments were performed in triplicate.

Transmission electron microscopy.

Conidia (10⁷ conidia/ml) and hyphae (10⁶ conidia/ml pregrown for 16 h) were treated with 0.1 μg/ml F901318 for 24 h in VMM at 37°C. Controls were untreated freshly harvested conidia and untreated hyphae grown for 40 h. Cultures were centrifuged and pellets were cryofixed by high-pressure freezing in a Leica EM PACT2 high-pressure freezing system (Leica Microsystems, UK) as described previously (33, 34). The pellets were subsequently embedded in Spurr's (epoxy) resin, and ultrathin sample sections were stained with lead citrate and uranyl acetate. Samples were examined with a JEM-1400 plus electron microscope, and images were captured using AMT Image Capture Engine software.

Sixteen-hour time-lapse, live cell imaging of growing germ tubes and vegetative hyphae.

Germlings were obtained by preincubating conidia for 8 h, and vegetative hyphae were obtained by preincubating conidia for 16 h. The inoculum size was 10⁵ conidia/ml, except for time-lapse imaging of hyphae, in which case the inoculum size was 10³ conidia/ml. Germlings and hyphae were treated with 0.1 μg/ml F901318 or 0.02% DMSO (controls).

The 16-h bright-field time-lapse image series was acquired on a Leica SP8X confocal microscope with a 40× (numerical aperture [NA], 0.85) dry objective, using an argon laser for illumination and a transmitted light detector. Time-lapse imaging was started immediately after the addition of F901318 or DMSO. z-stacks of 6 images 1 μm apart were acquired every 10 min for 16 h using Leica Application Suite software (LAS-AF). The elongation rate of 5 germ tubes per hypha was measured. The experiments were performed in triplicate.

Forty-six-hour time-lapse imaging of F901318-treated germlings.

Conidia were preincubated for 8 h at 37°C to obtain germlings (germinated conidia), which were treated with 0.1 μg/ml F901318. Bright-field time-lapse, live cell imaging was performed on a Nikon Eclipse Ti wide-field microscope with a 40× (NA, 0.6) dry objective using a 12-bit QI Cam cooled monocamera (QImaging, Canada) with Image Pro-Plus (version 7.0) software (Media Cybernetics). The first image in the series was acquired immediately after the addition of F901318, with z-stacks of 6 images 1 μm apart being captured every 10 min for 46 h.

Viability staining.

Conidia, germlings, and hyphae incubated in VMM were treated with 0.1 μg/ml F901318 at 37°C, and their viability was assessed after 24, 48, 72, 96, and 120 h of exposure to DiBAC (Molecular Probes) dissolved in DMSO and used at a final concentration of 2 μg/ml. Samples were incubated for 5 min at 37°C prior to imaging.

Images of conidia, germlings, and hyphae stained with DiBAC were acquired on a Leica SP8X confocal microscope with a 40× (NA, 0.85) dry objective using an argon or a white light laser for illumination. Excitation was set to 488 nm; emission was detected over the range of 500 to 590 nm. The percentage of at least 30 conidia and 50 germlings exhibiting DiBAC staining was quantified. For hyphae, the area of DiBAC-stained hyphae in 5 images was quantified using the Fiji distribution of ImageJ software (35), where images were segmented and the area of fluorescence in each image was selected and measured. Experiments were performed in triplicate.

Postexposure recovery biomass assay.

To obtain hyphae for the postexposure recovery biomass assay, 10⁶ conidia/ml were grown in VMM for 18 h at 37°C in flasks shaking at 200 rpm. Hyphae were treated with 0.1 μg/ml F901318, 1.0 μg/ml F901318, or 1% DMSO for 24 h, after which they were harvested and washed. A total of 0.3 g (wet weight) was then transferred into fresh, drug-free VMM and incubated for 48 h, after which the hyphae were dried and their weight was determined.

Postexposure recovery growth rate assay.

Hyphae were obtained for the postexposure recovery growth rate assay by pregrowing 10⁴ conidia/ml in SAB for 16 h at 35°C in 96-well plates. Hyphae were treated with phosphate-buffered saline (PBS) or PBS containing 0.1 μg/ml F901318, 0.5 μg/ml caspofungin, 1 μg/ml voriconazole, 4 μg/ml amphotericin B, or 1% DMSO (control). Drug concentrations represented 2× MIC. The plates were incubated for 24 h or 48 h at 35°C. Hyphae were washed 3 times with PBS-Tween and placed in fresh SAB containing 10 μM uridine and 10 μM uracil. The plates were sealed with a Breathe-Easy sealing membrane (Sigma). The OD₄₀₅ was determined every 10 min for 24 h.

Statistics.

Statistical analyses were performed using GraphPad Prism (version 7) software. For the analyses of conidial diameter, a two-way analysis of variance and Bonferroni's multiple-comparison tests with 99% confidence intervals were performed. For the postexposure effect biomass assay, a Student's t test was performed with 99% confidence intervals. All experiments were performed in triplicate (n = 3). Bars represent the standard deviation (SD).