Scientific Rationale

The various metabolic pathways of medically relevant fungi have recently come under investigation for their roles in fungal pathogenicity and the acquisition of drug resistance (8, 9). Specifically, the mitochondrion and its numerous components and functions are emerging as factors in determining the effectiveness of current antimycotic therapies in controlling human fungal infections. In vitro experiments have shown that loss of mitochondrial DNA (mtDNA) in C. glabrata correlates to increased resistance to azole antifungals (6). Paradoxically, clinical isolates of azole-resistant C. glabrata rarely show impairment of mtDNA function. The significance of the seemingly contradictory in vitro assays remains to be understood (10, 11).

The importance of mitochondrial function to fungal virulence is no better understood. C. glabrata with dysfunctional or missing mtDNA, also known as petite strains, are documented as being less pathogenic than non-petite strains (6). However, this observation is only reproducible when the genetic damage is artificially induced with chemical reagents such as ethidium bromide. The work of Ferrari et al. showed the C. glabrata petite mutant clinical isolate BPY41 was actually more virulent than its non-petite parent BPY40 in murine infection models (12). The 15-day survival rate of mice infected with BPY41 was 25% compared to 65% when infected with BPY40. Similarly, murine fungal loads of BPY41 after 7-day infection were 10- to 100-fold greater on average. Interestingly, when BPY40 was treated with ethidium bromide to generate a petite mutant, this newly generated petite strain was noticeably less pathogenic than BPY41, with a 15-day survival rate of 85% but similar fungal loads to its parent strain BPY40 (13) These conflicting observations clearly underscore the need to better understand the role mitochondria play in pathogenic fungi.

Similar work with Candida albicans has proven even more challenging because C. albicans petite mutants are more difficult to obtain (13, 14). In one exceptional case, Cheng et al. were able to produce a viable petite mutant by passing C. albicans SC5314 through murine spleens by intravenous inoculation (15, 16). After five serial passages, a mutant strain named P5 was isolated and characterized. P5 exhibited several phenotypes characteristic of respiratory mutants, including the inability to proliferate on glucose-deficient media. Subsequent oxygen consumption measurements suggested that while the electron transport chain was intact and functional, it was no longer coupled to ATP synthesis (16). With regards to clinical relevance, P5 was determined to be 10-fold less susceptible to fluconazole and voriconazole, more resistant to phagocytosis and neutrophils, and less sensitive to superoxide generators. In addition, P5 sustained a completely non-lethal infection in mice for up to 60 days. In contrast, SC5314 infection resulted in 100% mortality within 9 days (15).

The reported work with C. glabrata and C. albicans petite mutants suggests that modifying fungal mtDNA in laboratory settings with chemical agents cannot be relied upon to investigate the clinical behavior of these invasive fungi. Similarly, clinically isolated Candida petite mutants may not be amenable to further experimentation. The isolation of petite mutants from patients rarely occurs, and for the few strains harvested thus far, the occurrence of nuclear DNA mutations cannot be ruled out. Identifying specific mitochondrial contributions to Candida clinical behavior is thus complicated because the selective pressures of the host environment cannot be reasonably and reliably controlled to allow modification of only fungal mtDNA.

An alternative, untested approach would be to modulate fungal mitochondria function during an active infection. However, the current toolbox of mitochondria modulators (Figure 1) targets the oxidative phosphorylation pathway and does not discriminate between fungal or mammalian mitochondria (17–20). Disruption of this process is typically lethal for both the invading fungi and its mammalian host. A possible fungal-selective mitochondria inhibitor is the antibacterial agent ilicicolin H (Figure 1, compound 1) (21). Studies by Trumpower et al. have shown that ilicicolin H binds to purified S. cerevisiae cytochrome bc₁ complex with ~100-fold selectivity over isolated bovine bc₁ complex (IC₅₀ = 0.003 μM and 0.200 μM, respectively) (22, 23). However, 1 displays considerable toxicity against HeLa cells (ED₅₀ = 2 μg/mL or 4.6 μM) (21). Follow-up studies with ilicicolin H have been slow to emerge, possibly because there are no longer any commercial sources of this natural product.

Figure 1

Some known inhibitors of fungal mitochondria oxidative phosphorylation. Ilicicolin H (1) has demonstrated 100-fold selectivity for fungal mitochondria in biochemical assays, but displays moderate toxicity against HeLa cells. Antimycin (2) and funiculosin (more...)

Our project seeks to identify compounds capable of modulating cellular respiration in fungi, preferably by affecting mitochondrial function, without triggering a similar response in mammalian cells. It has been reported that fungal mitochondria possess several proteins lacking human orthologs (8, 24), and we have adopted a phenotypic assay tree in the hope of identifying modulators of these unique targets. Probes successfully developed within this project will be of great value for interrogating the metabolic requirements of fungal virulence. In addition, discovered probes may provide essential leads for the development of new antifungal drugs that operate in a completely unexploited target space.

Materials and Reagents

• Alamar Blue was purchased from Invitrogen (Catalog No. DAL1100).
• JC-1 dye was obtained from Invitrogen (Catalog No. T-3168).
• Amphotericin B was purchased from Sigma-Aldrich (Catalog No. A2411).
• Geldanamycin was obtained from AG Scientific (Catalog No. G-1047).
• Ilicicolin H was previously purchased from Analyticon Discovery (Catalog No. NP-005728) and was purified by column chromatography prior to use.
• Carbonyl cyanide m-chlorophenyl hydrazone (CCCP) was purchased from Sigma-Aldrich (Catalog No. C2759).

Cell Lines

The following cell lines were used in this study:

• Candida albicans CaCi-2; a drug-resistant clinical isolate was provided by the Whitehead Institute (25). This strain was used for the primary assay.
• Candida albicans CaCi-17; a strongly drug-resistant clinical isolate was provided by the Whitehead Institute (25). This strain was used for the primary assay
• Saccharyomyces cerevisiae W303-1a (ATCC 208352); a well-characterized Saccharomyces strain used for determination of glycolytic vs. respiratory metabolism, as well as in the mitochondrial disruption assay.
• Candida glabrata (ATCC 2001); used for determination of glycolytic vs. respiratory metabolism.
• NIH-3T3 mammalian fibroblasts (ATCC; CRL no.1658); used for mammalian toxicity assays.

2.1. Assays

A summary listing completed assays and corresponding PubChem AID numbers is provided in Appendix A. Refer to Appendix B for the detailed assay protocols.

2.2. Probe Chemical Characterization

After preparation as described in Section 2.3, the probe (ML316) was analyzed by UPLC, ¹H, ¹⁹F and ¹³C NMR spectroscopy, and high-resolution mass spectrometry. The data obtained from NMR and mass spectroscopy are consistent with the structure of the probe, and UPLC indicates an isolated purity of greater than 95%. The complete synthetic protocol is provided in Appendix C, and the associated spectral data are provided in Appendix E.

The solubility of the probe (ML316) was experimentally determined to be 1.1 μM in phosphate buffered saline with 1% (v/v) DMSO. The probe is reasonably stable in PBS solution (>60% remaining after a 48-hour incubation). The data from the PBS stability assay is provided in Figure 2. The probe is unstable in human and murine plasma with less than 1% remaining after incubation at 37°C for 5 hours. The ethyl carbamate appears to be the primary source of instability since replacing the carbamate with a methyl group (CID 56604835) significantly increases plasma stability (Table 1). However, the stable analog CID 56604835 shows no activity against the C. albicans test strains (see Section 3.4 for full details).

Figure 2

Stability of the Probe (ML316) and CID 56604835 in PBS Buffer (pH 7.4, 23°C).

Table 1

Plasma Stability of Probe ML316 and analogs CID3889161 and CID56604835.

Experimental procedures for all analytical assays are provided in Appendix D. The physical properties of the probe ML316 are summarized in Table 2.

Table 2

Summary of Probe Properties Computed from Structure ML316.

2.3. Probe Preparation

Probe 1 (ML316) and most of the analogs prepared for structure-activity relationship analysis were prepared by the four-step sequence outlined in Scheme 1. Cyclohexanone was subjected to a Strecker reaction with 4-fluoroaniline and trimethylsilyl cyanide. The resulting adduct 4 was cyclized to the hydantoin 5 with sodium cyanate, then converted to the thiohydanatoin 6, and acylated with ethyl chloroformate to provide the probe 7. Full experimental details are provided in Appendix C.

Scheme 1

Synthesis of the Probe (ML316).

Probe attributes

• Inhibits growth in primary screen cell line C. albicans CaCi-2 at an IC₅₀ ≤ 10 μM.
• Inhibits growth in resistant cell line C. albicans CaCi-17 at an IC₅₀ ≤ 10 μM.
• Inhibits growth in S. cerevisiae grown on glycerol at an IC₅₀ ≤ 10 μM.
• Inhibits growth in C. glabrata grown on glycerol at an IC₅₀ ≤ 10 μM.
• Shows at least 10-fold selectivity between the CaCi-2 test strain and murine 3T3 cells.
• Shows at least 10-fold selectivity between S. cerevisiae grown on glycerol and S. cerevisiae grown on glucose.
• Show at least 10-fold selectivity between C. glabrata grown on glycerol and C. glabrata grown on glucose.

The current project emerges from an earlier MLPCN effort to identify compounds capable of reversing antifungal drug-resistance, thereby providing probes to interrogate the mechanisms responsible for drug-resistance in medically relevant fungi (PubChem summary AID 2007). Compounds that directly inhibit fungal growth were outside the scope of the original project, and a secondary assay integrated into that project measured a compound’s intrinsic antifungal activity (AID 2387), rather than its ability to reverse resistance to the widely used antimycotic fluconazole. This secondary assay identified a number of compounds that possessed potent single agent activity against a range of opportunistic human fungal pathogens, but only under culture conditions that require mitochondrial respiratory metabolism for growth and survival.

We mined our data reported in PubChem for growth inhibition of Candida albicans as well as mammalian fibroblasts (AID 2387 and 2327, respectively). Analysis of this data identified 43 compounds that inhibited C. albicans growth with IC₅₀ ≤ 10 μM. Approximately 75% (30 of 43) of these single-agent antimycotics showed modest cytotoxicity against fibroblasts (IC₅₀ ≥ 20 μM). Based on this data mining, seventeen dry powders were obtained and evaluated for C. albicans and fibroblast growth inhibition. After examining the re-test data, a spirocyclic thiohydantoin (CID3889161) was prioritized for further development.

The initial hit (CID 3889161, Figure 3) is a potent inhibitor of C. albicans growth (IC₅₀ = 0.3 μM), with insignificant cytotoxicity against murine 3T3 fibroblasts (IC₅₀ ≥ 26 μM). Analogs of CID3889161 were designed and prepared accordingly, resulting in the identification of the more potent probe ML316 (see Section 3.4).

Figure 3

Dose Response Curves for Initial Hit (CID 3889161). CID 3889161 was used over a range of concentrations up to 26 μM in the primary assay and secondary assays. Dose curves were generated with Genedata Condeseo and show normalized percent activity (more...)

3.1. Summary of Screening Results

As described previously, an earlier high-throughput screen evaluated 302,509 compounds for their ability to chemosensitize the fluconazole-resistant Candida albicans strain CaCi-2 to a sublethal concentration of fluconazole (PubChem AID 1979). A secondary screen of the primary assay hits was included to identify any intrinsic antimycotic activity, and 342 compounds were evaluated in this assay. 43 compounds displayed inhibitory activity against CaCi-2 with IC₅₀ values below 10 μM (AID 2387). While their antifungal activity disqualified these compounds from the prior project, this set of 43 compounds was re-purposed for the current campaign to identify potent single-agent antimycotics.

DMSO stocks of only 39 compounds were available for re-test against the C. albicans strain CaCi-2 (PubChem AID 588529) and the more highly drug-resistant CaCi-17 (PubChem AID 588530). Of the 39 compounds, 23 surpassed the less than 10 μM IC₅₀ screening cutoff for both assays. Unfortunately, the cytotoxicity of many compounds appeared to be non-specific as determined by a murine 3T3 fibroblast counter screen (AID 588634). Of the 23 antifungal compounds, only two compounds (CID 26662188 and CID 3889161) displayed greater than 10-fold selectivity towards fungal cells. One compound (CID 26662188) was deemed chemically intractable and subsequently discarded.

In order to identify additional candidates, the PubChem data for AID 2387 was revisited. The CaCi-2 inhibitory activity threshold was raised from 10 μM to 15 μM to produce a list of 67 compounds. This hit list was cross-referenced against the 39 compounds previously evaluated as described above and also checked for fibroblast toxicity using available PubChem data (AID 2327). After applying these filters, an additional 17 compounds were obtained.

Dry powders for these 17 compounds, as well as for CID 3889161, were procured for re-testing. In addition, several analogs of promising compounds were obtained to provide preliminary SAR data and validate these scaffolds. After purity analysis and structural confirmation by NMR spectroscopy, a total 24 substances was screened according to the workflow outlined in Figure 4. The screens in this pathway consist of CaCi-2 and CaCi-17 inhibitory activity, mammalian toxicity, and growth media-dependent activity against C. glabrata and S. cerevisiae. The associated PubChem AIDs for these assays are provided in Figure 4. CID 3889161 emerged as the most promising candidate and was subsequently prioritized for further investigation. A number of analogs related to this scaffold were prepared and assayed for fungal selective respiratory inhibition. The results of the SAR investigation are presented in detail in Section 3.4.

Figure 4

Critical Path for Probe Development.

3.2. Dose Response Curves for Probe

Figure 5ML316 Selectively Inhibits Growth of C. albicans (A and B) over 3T3 Fibroblasts (C)

C. albicans CaCi-2 growth inhibition (AID 623899), IC₅₀ = 0.04 μM (A); C. albicans CaCi-17 growth inhibition (AID 623897), IC₅₀ = 1.0 μM (B); Murine 3T3 fibroblasts growth inhibition (AID 602394), IC₅₀ > 26 μM (C); Dose curves were generated with Genedata Condeseo and show normalized percent activity for the individual doses, performed in duplicate.

Figure 6ML316 Inhibits Growth of S. cerevisiae and C. glabrata When Fed Glycerol but Not Glucose

S. cerevisiae on glycerol growth inhibition (AID 623969), IC₅₀ = 0.004 μM (A); S. cerevisiae on glucose growth inhibition (AID 623971), IC₅₀ > 3.0 μM (B); C. glabrata on glycerol growth inhibition (AID 623965), IC₅₀ = 0.011 μM (C); C. glabrata on glucose growth inhibition (AID 623967), IC₅₀ > 3.0 μM (D). Dose curves were generated with Genedata Condeseo and show normalized percent activity for the individual doses, performed in duplicate.

3.3. Scaffold/Moiety Chemical Liabilities

A potential chemical liability associated with the probe is the possible hydrolysis of the ethyl carbamate. As discussed in Section 2.2, slow hydrolysis in PBS was observed with noticeable loss of the carbamate only occurring after 24 hours. This instability was addressed with the preparation of N-alkyl derivatives and larger, more stable carbamates. However, these analogs displayed diminished activity against the C. albicans test strains, and the complete SAR study results are presented in Section 3.4.

A search of PubChem for the probe (ML316) indicated that the probe has not been previously evaluated in any other assay. A structure-based search in SciFinder and Reaxys did not identify any publications or patents in which the probe appeared.

3.4. SAR Tables

In order to investigate the activity of the hit (CID 3889161), a collection of 51 structurally related analogs were synthesized and evaluated for their ability to inhibit growth in the C. albicans test strains. The biological assay data of these analogs are presented in Tables 3–8 (PubChem AID 623899, AID 623897, AID 602394). Characterization data (¹H NMR spectra and UPLC chromatograms) of these analogs are provided in Appendix G.

Table 3

SAR Analysis of Ethyl Carbamate.

Table 8

Growth media-dependent activity of C. albicans active analogs.

The hit compound (CID 3889161, Table 3, Entry 1) displayed significant activity against the C. albicans CaCi-2 strain (IC₅₀ = 0.3 μM), but was considerably less effective at inhibiting the growth of the more resistant CaCi-17 strain (IC₅₀ = 8.5 μM). There was also no noticeable toxicity to fibroblasts (IC₅₀ > 26 μM). The weak activity towards CaCi-17 was the primary attribute targeted for improvement during subsequent SAR studies.

The importance of the ethyl carbamate was examined first, and the results are presented in Table 3. Alternative alkyl carbamates displayed moderate activity against CaCi-2 (Table 3, entries 2–5). Potency against this strain correlated with the size of the carbamate, and smaller alkyl chains were more effective. Except for the methyl carbamate derivative which only showed marginal activity, these analogs had no effect on CaCi-17 growth. Phenyl carbamate (Table 3, entry 6) showed no activity against either C. albicans strain. Replacing the nitrogen cap with other functionalities such as methyl sulfonamide, butyl amide, and N-alkyl chains did not confer additional antifungal activity (Table 3, entries 7–10). Similarly, the unsubstituted nitrogen failed to inhibit C. albicans growth (Table 3, entry 11).

While the substituent of the imide nitrogen seemed resistant to modification, the neighboring N-tolyl group was considerably more pliant (Table 4). Substitution at the phenyl ring’s 4-position was not required for activity, and removing the methyl group appeared to improve potency against the more resistant CaCi-17 line (Table 4, entry 2). Conversely, incorporating methyl ether at that position attenuated the compound’s effect on both C. albicans strains (Table 4, entry 3). Progressing to electron-deficient phenyl rings provided the most significant gains in potency (Table 4, entries 4–7). While the trifluoromethyl was the worst performer among the electron-withdrawing substituents examined, this analog was still only 2-fold weaker than the original hit. In comparison, the 4-fluoro, chloro, and cyano replacements were all 4 to 8-fold more effective against both CaCi-2 and CaCi-17. Comparison of positional effects showed that the location of the substituent has only a marginal influence on the compound’s potency (Table 4, entries 8–9), although minor gains in PBS solubility were observed. The 4-ethyl and 4-propyl derivatives retain potency against CaCi-2 (Table 4, entries 10–11), but potency diminishes rapidly with larger substituents at this position (Table 4, entries 12–14). However, significantly smaller substituents are not effective growth inhibitors either (Table 4, entries 15–16). The N-benzyl analog showed a 10-fold drop in activity (Table 4, entry 17). The activity trends of this subset suggest the N-tolyl cap may lie within a sizeable pocket. While excessively large substituents attenuate activity, smaller functionalities (e.g. isopropyl or N-H) seem unable to occupy this space effectively and therefore lower cellular activity.

Table 4

SAR Analysis of Eastern Arene Ring.

Modification of the spirocyclic cyclohexane led to drastic reductions in activity (Table 5). Incorporating oxygen into the ring system provided a large increase in solubility, but this was accompanied by a 100-fold drop in potency (Table 5, entry 2). Substituting smaller aliphatic rings reiterated this trend; gains in solubility were offset by the inability to affect C. albicans growth (Table 5, entries 3–5).

Table 5

SAR Analysis of Northern Cyclohexane Ring.

Table 6 summarizes attempts to alter the underlying thiohydantoin core. Exchanging any of the chalcogens for other elements (O → N or S → O) eliminated cellular activity (Table 6, entries 2–3). Similarly, switching the nitrogen substituents produced an inactive derivative (Table 6, entry 4). N-alkyl hydantoins showed no ability to inhibit C. albicans growth (Table 6, entries 5–6) and neither did N-alkylated 4-thiohydantoins (Table 6, entries 7–8). Many intermediates prepared were also evaluated for antifungal activity, but none of these compounds were able to affect C. albicans (Table 7). The results summarized in Table 7 clearly demonstrate the importance of the carbamate (cf. Table 7, entries 7–8 & 10).

Table 6

Modifications of the Thiohydantoin Core.

Table 7

Bioactivity of Various Intermediates.

Six of the most potent inhibitors of C. albicans, including hit CID 3889161, were then evaluated against C. glabrata and S. cerevisiae under different growth conditions to investigate their ability to perturb cellular respiration pathways in fungi (Table 8). All of the compounds tested were able to deter growth in both fungal strains at nanomolar concentrations (IC₅₀’s 0.006 – 0.058 μM) when the yeast were cultured in non-fermentable media. Conversely, yeast grown on a glucose-derived media showed considerable resistance to the test compounds (IC₅₀ > 3 μM). Amphotericin B significantly inhibited C. glabrata and S. cerevisiae growth regardless of the growth media (Table 8, entry 7).

Of the analogs evaluated, the 4-fluoro derivative (CID 56604860) demonstrated the most activity against the C. albicans test strains CaCi-2 (IC₅₀ = 0.04 μM) and CaCi-17 (IC₅₀ = 1.0 μM) (Table 4, entry 4). In addition, CID 56604860 appears to disrupt cellular respiration processes of the fermenting yeast C. glabrata and S. cerevisiae. CID 56604860 is a potent growth inhibitor of these species when only non-fermentable carbon sources are available (IC₅₀ = 0.011 and 0.004 μM, respectively), but its antifungal properties can be mitigated by permitting the yeast to ferment glucose (IC₅₀ > 3.0 μM) (Table 8, entry 3). Because of its impressive performance in these cellular assays, CID 56604860 was nominated as the probe (ML316) for the respiration-selective growth inhibition of pathogenic fungi.

3.5. Cellular Activity

Both primary assays were performed with whole cells. One secondary screen evaluating toxicity utilized murine 3T3 fibroblast cells, and growth inhibition assays of whole S. cerevisiae and C. glabrata on different media were included as additional secondary assays. An overview of the assays is provided in Section 2.1, and full experimental details can be found in Appendix B. The probe (ML316) clearly met the established cellular activity criteria specified for this project (refer to Section 4, Table 9).

Table 9

Comparison of Probe ML316 to Project Criteria.

3.6. Profiling Assays

The probe was evaluated for binding to a broad panel of receptors, ion channels and enzymes. ML316 exhibits a remarkable selectivity profile against the 67 targets screened. When tested at 10 μM, the dopamine transporter showed approximately 80% inhibition with an estimated IC₅₀ of 1.7 μM. The remaining 66 targets showed less than 50% inhibition when treated with 10 μM ML316 (Figure 7).

Figure 7

Selectivity profiling of ML316.

4.1. Comparison to existing art and how the new probe is an improvement

A search of the available literature identified several compounds capable of disrupting fungal respiration (17–20). Details of this prior art search are provided in Appendix F. With the exception of ilicicolin H, these compounds do not show appreciable selectivity towards fungi over mammalian targets or cells. Hence, we decided to compare the probe ML316 with ilicicolin H.

Previous work with the antifungal antibacterial ilicicolin H has shown that this natural product is a potent inhibitor of the mitochondrial cytochrome bc₁ complex (22, 23). Using purified proteins, Trumpower et al. determined that ilicicolin H binds to the S. cerevisiae enzyme complex with almost 100-fold greater selectivity than the corresponding bovine homolog. To our knowledge, this selectivity has not been reproduced within a cell-based assay.

A commercial sample of ilicicolin H was procured from Analyticon Discovery and determined to be approximately 85% pure by HPLC. The commercial sample was purified by column chromatography over silica gel prior to testing in cellular assays. We recently attempted to obtain additional quantities of ilicicolin H but discovered that this compound is no longer commercially available.

Probe ML316 was tested alongside ilicicolin H in a direct comparison of biological activity. The results are summarized in Table 10. ML316 clearly shows a stronger inhibitory effect upon Candida albicans growth than ilicicolin H. While both compounds show differential activity relative to growth media, the distinction is far less pronounced with ilicicolin H regardless of the fungal test strain.

Table 10

Comparison of Probe ML316 to Prior Art.

Given the complex structure of ilicicolin H, optimization of biological and physical properties through synthetic methods will be difficult. Despite having been isolated 40 years ago (21), the first and only total synthesis of the ilicicolin H was reported in 1985 (26). That this pioneering synthesis produced racemic material further underscores the inherent difficulties of preparing analogs of this compound. While semi-synthetic analogs are possible (27, 28), the lack of a commercial feedstock renders this option untenable at the present time.

As a small molecule probe, ML316 is the superior compound because of its cellular potency and selectivity, and its simple preparation and chemical tractability facilitate further optimization.

4.2. Mechanism of Action Studies

A series of secondary assays were integrated into the current project to interrogate the possible mode of action. ML316 appears to disrupt cellular respiration of fungal organisms as demonstrated by the distinctly different growth patterns of C. glabrata and S. cerevisiae when they are fed with fermentable and non-fermentable carbon sources. In both species, ML316 is most effective when the yeast are grown in glycerol-containing media and displays a greater than 250-fold reduction in potency when the yeast are grown in glucose-containing media.

Additionally, direct disruption of mitochondrial membrane potential in C. glabrata and S. cerevisiae was assessed using a fluorescence probe JC-1. JC-1 is a lipophilic, cationic dye that can selectively enter mitochondria and reversibly changes from red to green when homeostatic mitochondrial membrane potential is lost. Application of JC-1 dye to living yeast treated with 1 μM ML316 determined that depolarization of the mitochondrial membrane does not occur (Figures 9 and 10) (PubChem AIDs 624062, 624068). Comparison of the DMSO controls (Figures 9 and 10, A) with ML316 (Figures 9 and 10, B) suggests the yeast can still maintain their normal mitochondrial membrane potential; it appears that the molecular target of ML316 may be integrated within an alternative metabolic pathway that fungi utilize for respiration (10).

Figure 9

Evaluating Loss of S. cerevisiae Mitochondrial Membrane Potential with JC-1. Δψ_m depolarization was monitored by FACS analysis with JC-1 mitochondrial staining. DMSO negative control in S. cerevisiae (A); S. cerevisiae treated with 1.0 (more...)

Figure 10

Evaluating Loss of C. glabrata Mitochondrial Membrane Potential with JC-1. Δψ_m depolarization was monitored by FACS analysis with JC-1 mitochondrial staining. DMSO negative control in C. glabrata (A); C. glabrata treated with 1.0 μM (more...)

4.3. Planned Future Studies

Target identification of ML316 will be the main priority. Current evidence suggests ML316 disrupts a critical respiratory pathway independent of mitochondrial membrane potential. ML316 will now be evaluated in genetic studies using genome-wide over-expression and deletion libraries to define its mode(s) of action in S. cerevisiae cultured under fermentative versus respiratory growth conditions. To complement genetic approaches, affinity precipitation and proteomic approaches based on Stable Isotope Labeling with Amino acids in Cell culture (SILAC) technology may also be used for target identification.

The probe will be submitted to the Center for Medical Mycology, an affiliate of the University Hospitals of Cleveland and Case Western Reserve University, to evaluate its antifungal activity against a panel of pathogenic fungi.

Further chemical optimization of ML316 will focus on optimization of aqueous solubility and plasma stability. The N-tolyl substituent appears to be the most tractable position, and the introduction of heteroarenes like pyridines at this location may afford the desirable solubility. Analogously, embedding the ethyl carbamate into an oxazole or similar ring systems might address the observed plasma instability. By addressing the solubility and plasma stability issues, we seek to prepare the probe for application in well-established mouse models of human disease in order to determine whether mitochondrial function is required for infection by drug-resistant C. glabrata and C. albicans.

CaCi-2 Dose-Response Primary (AID 558529, AID 602190, AID 623899, AID 623981, AID 624006)
CaCi-17 Dose-Response Primary (AID 588530, AID 602189, AID 623897, AID 623982)

Counter screen Mammalian Cell Toxicity Dose Response (AID 588634, AID 602394)

Secondary Glycerol Assay Protocol (AID 623969, AID 623965, AID 624014, AID 624012)

Secondary Glucose Assay Protocol (AID 623971, AID 623967, AID 624013, AID 624025)

Secondary Mitochondrial Disruption Assay (AID 624062, AID 624068)

Data Analysis

For the primary screen and other assays, negative-control (NC) wells and positive-control (PC) wells were included on every plate. The raw signals of the plate wells were normalized using the ‘Stimulators Minus Neutral Controls’ method in GeneData Assay Analyzer (v7.0.3). The median raw signal of the intra-plate NC wells was set to a normalized activity value of 0, while the median raw signal of the intra-plate PC wells was set to a normalized activity value of 100. Experimental wells were scaled to this range, resulting in an activity score representing the percent change in signal relative to the intra-plate controls. The mean of the replicate percent activities were presented as the final ‘PubChem Activity Score’. The ‘PubChem Activity Outcome’ class was assigned as described below, based on an activity threshold of 70%:

• Activity_Outcome = 1 (inactive), less than half of the replicates fell outside the threshold.
• Activity_Outcome = 2 (active), all of the replicates fell outside the threshold, OR at least half of the replicates fell outside the threshold AND the ‘PubChem Activity Score’ fell outside the threshold.
• Activity_Outcome = 3 (inconclusive), at least half of the replicates fell outside the threshold AND the ‘PubChem Activity Score did not fall outside the threshold.