Identification of Small Molecule Modulators of Diguanylate Cyclase
Total Page:16
File Type:pdf, Size:1020Kb
bioRxiv preprint doi: https://doi.org/10.1101/402909; this version posted August 28, 2018. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license. Identification of Small Molecule Modulators of Diguanylate Cyclase by FRET-based High-Throughput-Screening Matthias Christen1, , Cassandra Kamischke2, Hemantha D. Kulasekara2, Kathleen C. Olivas3, Bridget R. Kulasekara4, Beat Christen1, Toni Kline5, and Samuel I. Miller2,4,6, 1Institute of Molecular Systems Biology, Eidgenössische Technische Hochschule (ETH) Zürich, CH-8093 Zürich, Switzerland 2Department of Microbiology, University of Washington, Seattle, United States 3Seattle Genetics, Inc., 21823 30th Drive SE, Bothell, Washington 98021 4Department of Genome Sciences, University of Washington, Seattle 5Sutro Biopharma, 310 Utah Avenue, South San Francisco, CA 94080 6Department of Medicine, University of Washington, Seattle The bacterial second messenger cyclic diguanosine monophos- parent role of c-di-GMP in the cell cycle and the presence of phate (c-di-GMP) is a key regulator of cellular motility, the cell many paralogous DGC enzymes controlling diverse cellular cycle, and biofilm formation with its resultant antibiotic tol- functions indicate that there is likely tight spatial and tem- erance, which may make chronic infections difficult to treat. poral regulation of c-di-GMP (10–12). Bacterial genomes Therefore, diguanylate cyclases, which regulate the spatiotem- encode multiple GGDEF domains in proteins with signal- poral production of c-di-GMP, may be attractive drug tar- sensing domains (13). However, the presence of multiple gets to control biofilm formation that is part of chronic infec- paralogs makes it difficult to study the signaling process- tions. In this paper, we present a FRET-based biochemical high- throughput screening approach coupled with detailed structure- ing properties of c-di-GMP signaling networks using con- activity studies to identify synthetic small molecule modulators ventional genetic techniques. Therefore, chemical genetic of the diguanylate cyclase, DgcA, from Caulobacter crescentus. approaches to inactivate each segment of the c-di-GMP sig- We identified a set of 7 small molecules that in the low µM range naling network may be an attractive approach to study the regulate DgcA enzymatic activity. Subsequent structure activity overall biological function of c-di-GMP. In pathogenic bac- studies on selected scaffolds revealed a remarkable diversity of teria, cellular production of c-di-GMP is essential to main- modulatory behaviors, including slight chemical substitutions tain biofilm formation, especially under stressed conditions that revert the effects from allosteric enzyme inhibition to acti- such as induction by aminoglycoside antibiotics (1). Small vation. The compounds identified represent novel chemotypes molecules that effectively inhibit DGC activity have the po- and are potentially developable into chemical genetic tools for tential to prevent biofilm formation, thus, making DGCs the dissection of c-di-GMP signaling networks and alteration of interesting targets to develop new classes of antimicrobial c-di-GMP associated phenotypes. In sum, our studies under- line the importance for detailed mechanism of action studies for agents. inhibitors of c-di-GMP signaling and demonstrate the complex We developed a sensitive and robust FRET based DGC activ- interplay between synthetic small molecules and the regulatory ity assay and performed high-throughput (HT) screening on mechanisms that control the activity of diguanylate cyclases. a comprehensive compound library to evaluate 27,502 small molecules for inhibition of the Caulobacter crescentus DGC c-di-GMP | FRET | High-throughput screening | diguanylate cyclase inhibitor DgcA (CC3285). DgcA has been extensively characterized Correspondence: [email protected]; [email protected] and therefore serves as a model enzyme to study c-di-GMP related phenotypic effects (9, 11, 12, 14). Similar to the ma- Introduction jority of DGC enzymes, DgcA is subjected to high affinity binding of c-di-GMP to an allosteric site (I-site), which effi- The second messenger cyclic dimeric guanosine monophos- ciently blocks enzymatic activity in a non-competitive man- phate (c-di-GMP) mediates diverse bacterial cellular pro- ner and resides distant from the catalytic pocket (A-site). Mu- cesses including; antibiotic resistance, biofilm formation, tational analysis of DgcA has provided convincing evidence extracellular carbohydrate and adhesin production, pilus- that c-di-GMP binding to several conserved charged amino and flagellum-based motility, and cell cycle progression acids at the I-site is a key mechanism for allosteric regulation (1–5). Signal integration into c-di-GMP networks is, in of DGCs (9). part, controlled by diguanylate cyclases (DGCs) that convert two molecules of GTP to c-di-GMP. The enzymatic activ- ity of DGCs resides within a conserved domain comprised Results of the amino acids, glycine-glycine-aspartate-glutamate- Development of a FRET-based biochemical high- phenylalanine (GGDEF) that forms the enzymatic active site. -throughput screen to monitor c-di-GMP production. The GGDEF domain shares similarity to the PALM 4 domain To identify small molecule inhibitors of DgcA, we estab- found in other classes of nucleotide cyclase (6–10). The ap- lished a sensitive FRET-based activity assay. We pre- Christen et al. | bioR‰iv | August 28, 2018 | 1–19 bioRxiv preprint doi: https://doi.org/10.1101/402909; this version posted August 28, 2018. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license. (Fig 1A, closed circles). Based on the FRET ratio for the A B free and c-di-GMP saturated biosensor, we determined the 1200 corresponding increase in c-di-GMP (Fig. 1B). Our sensitive 1.0 800 and robust FRET assay detects c-di-GMP production in the 0.9 nM range and permits determination of initial rate kinetics at 0.8 73.2 +/- 1.6 nM 400 c-di-GMP min-1 levels below occurrence of allosteric c-di-GMP inhibition. 0.7 FRET/CFP c-di-GMP [nM] 0.6 0 FRET-based HT-screening and discovery of primary 0 5 10 15 20 0 5 10 15 20 hits for diguanylate cyclase inhibition.. A major chal- Time [min] Time [min] lenge of biochemical HT screening is to gain significance on C D large-scale experimental datasets for quality control and hit selection. Through prior screening, we determined plate uni- 14 n=25‘966 Compounds 50µg/ml 12 1.1 c-di-GMP formity, signal variability and repeatability assessment assays Negative Control 10 100 for CV and Z-factor calculation. The average Z-factor for 3 1000] Positive Control 1.0 200 * 8 300 n=25‘966 0.9 plates read on two consecutive days was 0.71, and the CV 6 400 500 FRET/CFP for high, medium, and low signal were well below 5%. The 4 0.8 700 1000 chemical library screened encompasses 27,505 commercially 2 0.7 nM available compounds derived from small-molecule chemi- Frequency [ 0 0.7 0.8 0.9 1.0 1.1 0 5 10 15 20 25 cal libraries maintained at the Institute of Chemical Biology 3 FRET/CFP Compound ID [ *10 ] (ICCB) at Harvard University, Boston, MA, funded by the Regional Centers of Excellence in BioDefense and Emerging Fig. 1. FRET assay for c-di-GMP. (A) Kinetics of fluorescence emission ration Infectious Disease (NSRB/NERCE). During screening, we change (527/480nm) measured in a 384 well plate format. Affinity purified YcgR first preincubated the DgcA enzyme with 200 nL compound FRET biosensor was added in presence of 20 nM DgcA enzyme and 20µM GTP (5 mg/mL) in DMSO and measured fluorescence emission substrate (closed circles) or in absence of GTP (open circles). (B) Correspond- ing increase in c-di-GMP concentration derived from the change in fluorescence ratio before and after addition of the biosensor to monitor emission ration (535/470nm). Above 800 nM c-di-GMP, allosteric product inhibition fluorescence of chemical compounds and FRET at 470 and decreases DGC activity of DgcA. Each graph shows the average of three indepen- 535 nm. This second measurement will also detect com- dent measurements. The reaction volume per well was 20 µl in a 384 well Corning low volume flat bottom plate. (C) Histogram of the fluorescence emission ratio af- pounds that mimic c-di-GMP agonists. After addition of 20 ter 3 hours incubation of 20 nM DgcA with 20 µM GTP in presence of 50µg/ml µM GTP, FRET was measured 3 h after incubation as an end- compounds and 1% DMSO. Wells without GTP substrate added were used as pos- point measurement (Fig. 1C,D)) (materials and methods). itive controls (red), wells with exogenous c-di-GMP (5 µM) were used as negative controls (blue). (D) Plate uniformity of the HT screening. For every well, back- For a compound to be considered a hit, the following crite- ground fluorescence signal of the compound has been measured prior addition ria had to be met: (i) Compounds with c-di-GMP production of the FRET-biosensor. Compounds with auto fluorescence exceeding 7% of the below 300 nM after 3h incubation with DgcA and 20 µM FRET biosensor signal (535/470nm) were discarded. GTP were considered strong hits; (ii) duplicates were consis- tent with a difference in final c-di-GMP levels <100 nM; and viously reported a fluorescence resonance energy transfer (iii) background fluorescence prior to addition of the FRET (FRET)–based c-di-GMP biosensor, which we refer to herein biosensor must not exceed 7% of the FRET signal in CFP as biosensor (10).