Validation and Application of a Novel Target-Based Whole-Cell Screen to Identify Antifungal Compounds

Traditional approaches to drug discovery are inefficient and have several key limitations that constrain our capacity to rapidly identify and develop novel experimental therapeutics. To address this, we have devised a second-generation target-based whole-cell screening assay based on the principles of competitive fitness, which can rapidly identify target-specific and physiologically-active compounds. Briefly, strains expressing high, intermediate, and low levels of a preselected target protein were constructed, tagged with spectrally distinct fluorescent proteins (FPs), and mixed together. The pooled strains were then grown in the presence of various small molecules, and the relative growth of each strain within the mixed culture was compared by measuring the intensity of the corresponding FP tags. Chemically-induced population shifts indicated that the bioactivity of a small molecule was dependent upon the target protein’s abundance and thus established a functional interaction. Here, we described the molecular tools required to apply this technique in the prevalent human fungal pathogen Candida albicans and validated this approach using dihydrofolate reductase. However, our approach, which we have termed target abundance-based fitness screening (TAFiS), should be applicable to an extensive array of molecular targets and in any genetically tractable microbe. While the folate biosynthetic pathway has provided a rich source of antibacterial, antiprotozoal, and anticancer therapies, it has not yet been exploited to develop uniquely antifungal agents. Although there have been attempts to develop fungal-specific inhibitors of dihydrofolate reductase, the protein itself has not been unequivocally validated as essential for fungal growth or virulence. The purpose of this study was to establish dihydrofolate reductase as a valid antifungal target. Using a strain with doxycycline-repressible transcription of DFR1 (PTETO-DFR1), we were able to demonstrate that Dfr1p is essential for growth in vitro. Furthermore, nutritional supplements of most forms of folate are not sufficient to restore growth when DFR1 expression is suppressed, or when its activity is directly inhibited by methotrexate, indicating that Candida albicans has a limited capacity to acquire or utilize exogenous sources of folate. Finally, the PTETO-DFR1 strain was rendered avirulent in a mouse model of disseminated candidiasis upon doxycycline treatment. Collectively, these results confirm the validity of targeting dihydrofolate reductase, and by inference other enzymes in the folate biosynthetic pathway as a strategy to devise new and efficacious therapies to combat life-threatening invasive fungal infections. Both the fungal fatty acid synthase and desaturase have been established as essential for fungal growth, virulence in a systemic infection, and are structurally distinct from their mammalian counterparts. However, there has been little success in developing clinically relevant inhibitors that target the fungal fatty acid biosynthetic pathway, even though there are agents that inhibit the bacterial orthologs. The purpose of this study was to identify physiologically-active compounds that interact with the fungal fatty acid biosynthetic pathway. Given the challenging nature of targeting these enzymes using traditional approaches, we attempted to apply the newly validated TAFiS method, a target-based whole-cell screen. To accomplish this, Candida albicans strains with altered expression of the fungal fatty acid synthase or the desaturase were used to screen a total of 41,261 small molecules. The primary hits were confirmed through dose responses, and 16 compounds had differential activity on the strains with altered desaturase abundance, indicating a functional chemical-target interaction. Additionally, the growth inhibition exerted by the most potent hits was restored in the presence of exogenous fatty acids, further supporting their inhibition of the fatty acid desaturase. Of the 16 confirmed hits, 11 were structurally related, containing a hydrazide core that was essential for interaction with the desaturase. Collectively, these results have identified a novel scaffold that interacts with the fungal fatty acid desaturase and can be further developed into experimental therapeutics. It also confirms that the TAFiS methodology is highly effective in identifying small molecules that functionally interact with targets that are poorly suited to traditional chemical screening approaches.