Investigating a potential new anti-fungal drug target: Structure, function and inhibition of fungal BET bromodomains – FungiBET

The strategy used to study the bromodomains of Bdf1 involves a combination of yeast genetics, molecular biology, biochemistry, structural biology and synthetic organic chemistry. We use yeast genetics to study the effect of specific mutations in Bdf1 on the growth rate of Candida and on its virulence in a mouse model of invasive candidiasis. We use biochemical methods to isolate and purify the bromodomains of Bdf1 and use X-ray crystallography to determine their atomic structure, alone and in complex with inhibitors. We perform high-throughput screening of chemical compounds to identify specific inhibitors of the bromodomains, and characterize the hits in various biochemical and biophysical assays. We use synthetic organic chemistry to generate analogs of bromodomain inhibitors in an effort to develop more potent and selective inhibitors.

Our primary results include the following:
(1) We generated mutant C. albicans strains which fail to express Bdf1 or which express a mutant inactivated in one or both bromodomains (BDs). Using these strains we showed that Bdf1 BD functionality is essential for C. albicans viability in vitro.
(2) Using similar mutant strains in which Bdf1 is expressed under the control of a Tet promoter, we showed that Bdf1 BD functionality is required for virulence in a mouse model of invasive candidiasis.
(3) We developed a homogeneous time-resolved fluorescence (HTRF) assay to measure the binding of Bdf1 BDs to an acetylated histone peptide. Using this assay we showed that Bdf1 BDs are resistant to human BET inhibitors, implying structural differences in the ligand binding pockets of human BET and Bdf1 BDs.
(4) Using our HTRF assay to perform a high-throughput screen of 80,000 chemical compounds, we identified small molecules which inhibit Bdf1 BD1 or BD2 with low micromolar affinity and with 25- to 50-fold selectivity over the human BET orthologs.
(5) We determined high-resolution crystal structures of these compounds bound to their target Bdf1 BDs, revealing the stereochemical basis for inhibitor selectivity.
(6) We identified a dibenzothiazepinone compound which phenocopies the effect of a Bdf1 BD mutation on the growth of C. albicans, demonstrating that Bdf1 is sensitive to small-molecule inhibition in fungal cells. Taken together, this work establishes Bdf1 as a promising antifungal drug target in C. albicans that can be selectively inhibited without antagonizing human BET function. The above work is described in a recent publication (Mietton et al., Nature Comm. 2017; PMID: 28516956).
Our current efforts are focused on validating Bdf1 as an antifungal target in C. glabrata.

Using mutant strains of C. albicans, we have demonstrated that Bdf1 bromodomains are a potential anti-fungal target in this fungal pathogen. Moreover, we identified small-molecule compounds which selectively inhibit C. albicans Bdf1 bromodomains without antagonizing human bromodomain protein function. These findings raise the exciting prospect of developing a novel antifungal therapy by chemically inhibiting Bdf1 bromodomains.

1. Garnaud C, Champleboux M, Maubon D, Cornet M, Govin J. (2016) Histone Deacetylases and Their Inhibition in Candida Species. Front Microbiol 7:1238.

2. Ferri E, Petosa C, McKenna CE. (2016) Bromodomains: structure, function and pharmacology of inhibition. Biochem Pharmacol 106 :1-18.

3. Mietton F, Ferri E, Champleboux M, Zala N, Maubon D, Zhou Y, Harbut M, Spittler D, Garnaud C, Chauvel M, d’Enfert C, Kashemirov BA, Hull M, Cornet M, McKenna CE, Govin J, Petosa C. (2017) Selective BET bromodomain inhibition as an antifungal therapeutic strategy. Nature Comm., 8:15482

4. Selective bromodomain inhibition of fungal Bdf1. U.S. provisional patent application No. 62/366973 filed on July 26, 2016.

Overview: The central hypothesis of this project is that inhibition of the fungal BET family protein Bdf1 is a valid therapeutic strategy to combat invasive fungal infections. The project aims to verify this hypothesis for Candida albicans and Candida glabrata, the two most common pathogenic yeasts of humans.

Biomedical Relevance: Invasive fungal infections are a major cause of morbidity and mortality, with over 800,000 deaths per year worldwide. Candida species are the most common nosocomial pathogens. Of these, C. albicans and C. glabrata rank first and second in isolation frequency and cause approximately 70% of all candidemia. Currently, only four drug classes are available to treat candidemia and other systemic fungal infections. These drugs act by targeting the fungal cell wall (echinocandins) or cell membrane (polyenes, azoles) or by inhibiting nucleic acid synthesis (flucytosine). The limited number of anti-fungal strategies, the high cost and toxicity of available drugs and a rise in drug-resistant strains poses an urgent need for the development of novel therapeutic agents.

Rationale: We propose to investigate a potentially new antifungal strategy, inspired by recently discovered anti-cancer compounds that target a specific family of transcriptional regulators – the BET (“Bromodomain and Extra-Terminal”) proteins. BET proteins bind chromatin through their two bromodomains (BD1 and BD2), small helical domains which specifically recognize histones acetylated on lysine residues. Several recently discovered BET bromodomain inhibitors have been used to validate human BET bromodomain inhibition as a therapeutic strategy for numerous cancers and other pathologies. Genetic evidence in S. cerevisiae and C. albicans implicates the fungal BET protein, Bdf1, as a potentially new antifungal drug target.

Preliminary data: We have determined the crystal structures of Bdf1 bromodomains BD1 and BD2 from S. cerevisiae and from C. albicans. The ligand binding pockets of these structures are distinct from those of human BET proteins, raising the prospect of identifying a small molecule inhibitor selective for the fungal bromodomains. We developed AlphaScreen and FRET-based assays which will allow the high-throughput screening of chemical libraries to identify Bdf1 bromodomain inhibitors. In parallel, we established an in silico protocol to perform virtual screening on our Bdf1 crystal structures, and in a preliminary screen identified a candidate inhibitor of C. albicans Bdf1 BD2.

Objectives: The goal of this project is to establish proof-of-concept that chemical inhibition of fungal Bdf1 bromodomains is a valid anti-fungal therapeutic strategy. The specific aims are to: (i) investigate the structure and function of Bdf1 in C. albicans and C. glabrata; (ii) verify Bdf1 bromodomain inhibition as a valid antifungal therapeutic strategy; (iii) to develop chemical compounds that selectively inhibit Bdf1 bromodomains and demonstrate their translational potential into a novel class of antifungal drug.

Consortium: This project is a collaboration of research teams in Grenoble (led by Carlo Petosa, Jerome Govin and Muriel Cornet) and Los Angeles (led by Charles McKenna). The work envisaged is a comprehensive effort that combines expertise in structural biology (Petosa), yeast genetics and epigenetics (Govin), medical mycology (Cornet), and synthetic organic and medicinal chemistry (McKenna).

Expected results: This project should advance our understanding of Bdf1 function, yield selective small-molecule inhibitors of Bdf1 bromodomains and verify whether Bdf1 bromodomain inhibition is a valid antifungal therapy. A successful outcome is expected to generate patentable intellectual property and technology transfer opportunites for industry, paving the way for subsequent lead optimization and for a phase I clinical trial.