Allosteric Targeting of the Fanconi Anemia Ubiquitin-Conjugating Enzyme Ube2T by Fragment Screening

Ube2T is the E2 ubiquitin-conjugating enzyme of the Fanconi anemia DNA repair pathway and it is overexpressed in several cancers, representing an attractive target for the development of inhibitors. Despite the extensive efforts in targeting the ubiquitin system, very few E2 binders have currently been discovered. Herein we report the identification of a new allosteric pocket on Ube2T through a fragment screening using biophysical methods. Several fragments binding to this site inhibit ubiquitin conjugation in vitro.


Differential scanning fluorimetry (DSF)
DSF experiments were carried out using a CFX96 real-time PCR detection system (Bio-Rad). The fragment library was provided by the Drug Discovery Unit at Dundee. 5 0.5 µl of each fragment stock of the library (200 mM in DMSO) were dispensed into 96-well plates. 39.5 µl of solution was added to each well to obtain a final concentration of 5 µM Ube2T, 2.5x SYPRO orange, 2.5 mM fragment (1.25% DMSO) in 100 mM Tris pH 8.0, 100 mM NaCl, 0.25 mM TCEP. The samples were heated from 25 °C to 95 °C with increments of 1 °C/minute, and fluorescence was measured at each step. Data analysis was performed as described by Niesen et al. 6 Fragments' effect on Ube2T melting temperature was measured as: T m = T m Ube2T+fragment -T m Ube2T Where T m Ube2T is Ube2T melting temperature in the assay buffer containing 1.25% DMSO and T m Ube2T+fragment is Ube2T melting temperature in the presence of 2.5 mM fragment in 1.25% DMSO.

Bio-layer interferometry (BLI)
BLI experiments were carried out using an Octet RED 384 instrument (ForteBio) at 25 °C. Ube2T was biotinylated using the Thermo Scientific TM EZ-Link TM NHS-PEG 4 -Biotin reagent, which enables biotin labeling of primary amine-containing macromolecules. Ube2T (170 µM) was incubated with NHS-PEG 4 -Biotin (1:1 molar ratio) for 30 minutes at room temperature in a buffer containing 20 mM HEPES pH 7.5, 100 mM NaCl, 0.3 mM TCEP. The excess of NHS-PEG 4 -Biotin was removed using a 2 ml Zeba desalt spin column (Thermo Scientific). Biotinylated Ube2T was loaded upon superstreptavidin (SSA) biosensors by incubation for 900 s at 15 µg/ml. Biocytin (Tocris) at 10 µg/ml was used as quenching agent to block free streptavidin sites. A second set of SSA-biosensors were quenched with biocytin and used as control sensors. The fragment library was screened at a concentration of 200 µM in a buffer containing 20 mM HEPES pH 7.5, 100 mM NaCl, 0.3 mM TCEP, 0.1% DMSO. For each fragment a baseline, association and dissociation step were acquired (60 s each) using both the Ube2T-loaded and the control set of biosensors. The response of each fragment was taken as the average response of the last 5 s of the association step. 90 fragments were selected as primary hits giving a response above a threshold value calculated as the sum of the robust standard deviation multiplied by three and the median. A 6-point concentration series was then prepared for each of the 90 fragments in threefold dilution steps from a top concentration of 500 µM (500 µM, 166.7 µM, 55.6 µM, 18.5 µM, 6.17 µM, 2.06 µM). For each point of the concentration series a baseline step of 60 s, an association step of 60 s and a dissociation step of 90 s were acquired using both the Ube2T-loaded and the control sensors. Data were processed using the ForteBio Data Analysis Software and adopting the doublereferencing method to remove drift and well-to-well artefacts.

Chemical shift perturbation (CSP) experiments
In order to further validate fragments' binding and identify their binding site, the 13 fragments validated as positive hits in ligand-observed NMR experiments were further tested in CSP experiments. [ 1 H-15 N]-HSQC spectra were recorded at 25 °C on 220 l samples (in 3 mm capillary tubes, Bruker) of 15 N-labeled Ube2T 1-154 at a concentration of 80 M in a buffer containing 1% DMSO, 50 mM Potassium phosphate pH 6.8, 85 mM NaCl, 1 mM DTT and 10% D 2 O with and without 2 mM fragment. Analysis of the superposed [ 1 H-15 N]-HSQC spectra recorded with and without fragment highlighted 6 fragments that displayed pronounced shifts (chemical names and vendors that provided the 6 fragments are shown in Table S1). These 6 fragments were taken forward for a concentration-dependent study where [ 1 H-15 N]-HSQC spectra were recorded, as described above, at increasing concentration of fragment: 0.5 mM (with 0.25% DMSO in the buffer), 1 mM (0.5% DMSO), 2 mM (1% DMSO) and 3 mM (1.5% DMSO). It is worth noting however that several fragments were not entirely soluble throughout the titration. CSPs were calculated at the highest fragment concentration reached (3 mM) using the following formula: where Δδ H and Δδ N are the measured change in 1 H and 15 N chemical shift upon ligand binding. 14, 15

Isothermal titration calorimetry (ITC)
All the experiments were carried out using the MicroCal PEAQ-ITC (Malvern). 5 mM ligand was titrated against 220 M protein at 25 °C, whilst stirring at 750 rpm in a buffer containing 100 mM Tris pH 8.0, 100 mM NaCl, 0.5 mM TCEP. A control titration of 5 mM ligand into buffer was performed in order to account for the heat of dilution. The data were fitted to a single binding site model fixing the stoichiometry to 1 (given the weak affinity) using the MicroCal PEAQ-ITC Analysis Software.
Crystallization and structure determination! Full-length Ube2T was crystallized using sitting drop vapor diffusion. 1.5 l of 21 mg/ml Ube2T in 100 mM Tris pH 8.0, 100 mM NaCl, 0.25 mM TCEP was mixed with an equal volume of crystallization buffer (2.8 M Potassium acetate, 0.1 M Tris-HCl pH 8.5) and equilibrated against 0.5 ml of reservoir solution at 20 °C. Well diffracting crystals were obtained by micro seeding of initial crystals into fresh drops. Soaking was performed by adding 4 l of crystallization buffer containing the fragment dissolved in 15% DMSO/dioxane (2:3) to the drop containing Ube2T crystals, obtaining an approximate fragment concentration of 12 mM. After soaking, crystals were flash frozen in liquid nitrogen. Data were collected at Diamond Light Source (i04-1 beamline) at 0.9282 Å wavelength and processed using XDS, 16 POINTLESS 17 and AIMLESS 18 from the CCP4 program suite 19 to a resolution limit of 2.4 Å. Refmac5 20 and Coot 21 were then used to refine the structure, using PDB entry 1YH2 13 as a template.!The quality of the model was checked using MolProbity. 22 Ligand structure and restraints were generated using the PRODRG server. 23 Ligand's occupancy was set to 0.8 during the structure refinement. Data collection and refinement statistics are shown in Table S2.

In vitro ubiquitination assays
A synthetic gene of human FANCL optimized for E. coli expression was purchased from GeneArt. The coding sequence for the central and RING domains (residues 109-375) was subcloned into a vector containing an N-terminal 6xHis-Smt3 tag, expressed and purified as previously described. 3,24 The human FANCD2-12xHis was a kind& gift from M. R. Hodskinson and K. J. Patel and was expressed and purified as previously described. 25 Recombinant ubiquitin (Ub) was expressed either as wild-type protein or bearing a SGCGSG overhang at the N-terminus. The cysteine residue in the overhang was targeted for site-specific incorporation of a DyLight 800-4xPEG Maleimide (Life Technologies) dye following manufacturers protocol. Labeled species was further subject to cation exchange chromatography and stored at -20 °C as single-use aliquots. All ubiquitin reactions were performed at 30 °C in a 100 mM Tris pH 7.5, 100 mM NaCl, 5 mM MgCl 2 , 1 mM TCEP hydrochloride, 0.5% (w/v) poly(ethylene glycol) 6000 buffer system. FANCD2 ubiquitination reactions (10 µl) contained 2 µM of labeled Ub, 25 nM of recombinant human E1, 0.1 µM of Ube2T, 0.1 µM of human FANCL (109-375), 1 µM of FANCD2, 2.5 mM of the indicated fragments in 5% DMSO (or just 5% DMSO for the control reaction) and 5 mM ATP. Reactions were carried out for 15 minutes and terminated by boiling with LDS loading buffer (Life Technologies). The samples were resolved by SDS-PAGE and analyzed by direct fluorescence monitoring using Li-COR Odyssey Infrared Imaging System. Integrated intensities of FANCD2-Ub from three independent experiments were obtained using Image Studio (Odyssey) imaging software, normalized to DMSO reactions and plotted as mean ± standard error of mean (SEM) or ± range using GraphdPad Prism7. Statistical analysis was carried out using GraphdPad Prism7. Ube2T~Ub thioester reactions (20 µl) contained 40 µM of Ub, 0.2 µM of recombinant human E1, 20 µM of Ube2T, 2.5 mM of the indicated fragments in 5% DMSO (or just 5% DMSO for the control reaction) and 5 mM ATP. Reactions were carried out for 5 minutes and terminated with non-reducing LDS loading buffer or by boiling with LDS loading buffer containing reducing agent. The samples were resolved by SDS-PAGE, stained and analyzed using Li-COR Odyssey Infrared Imaging System. Integrated intensities of Ube2T species from three or four independent experiments were obtained using Image Studio (Odyssey) imaging software. Intensity levels of ubiquitinated Ube2T in the reduced samples were subtracted from total levels of ubiquitin bound Ube2T species in the non-reduced samples to obtain relative measure of the Ube2T~Ub thioester species. Percentage of ubiquitin loading was calculated using Ube2T levels in the no ATP reactions. Values were plotted as mean ± SEM using Excel ( Figure S9). The 6xHis-MBP-RNF4 RING fusion and UbcH5c were purified as described by Plechanovová et al., 26 except the MBP tag was retained for the RNF4 fusion. Ubiquitination reactions using RNF4 and UbcH5c were carried out as described above expect using 0.05 µM RNF4 and limiting the total reaction time to 10 minutes.