Small-molecules that covalently react with a human prolyl hydroxylase – towards activity modulation and substrate capture† †Electronic supplementary information (ESI) available: Synthesis, mass spectrometry, inhibition assays, mutagenesis. See DOI: 10.1039/c8cc07706a

We describe covalently binding modulators of the activity of human prolyl hydroxylase domain 2 (PHD2) and studies towards a strategy for photocapture of PHD2 substrates.


Figures
Structures of glycine-coupled salicylic acids and analogues used in this work.        aldehyde 1 (or 2) in buffer (15 mM ammonium acetate, pH 7.5) to enable formation of the imine adduct. The modified enzyme was subsequently immobilised on a C18 column, and the imine reduced by eluting (5 mL, 1 mL min -1 ) with NaBH 3 CN (30 mM) in buffer (50 mM sodium phosphate, pH 7.5).    immunoblotting. The inhibitor FG4592 was used as a control.

Analysis of protein activity by peptide turnover assay using the AlphaScreen™ 3
The assay consisted of three parts: pre-incubation with inhibitor, the PHD2 hydroxylation reaction, and incubation with the AlphaScreen™ bead mix. The assay was carried out in 384-well white ProxiPlates μM Fe(II) and 200 μM ascorbate) was incubated with the inhibitor (1 min, 20 min or 1 h) prior to addition to the peptide mix (60 nM biotinylated CODD peptide-HIF-1α 556-574 and 2 μM 2OG). The resulting solution was incubated for 10 min at room temperature before the hydroxylation reaction was quenched by addition of EDTA (5 μL, 30 mM). 5 μL of pre-incubated donor-acceptor bead mix (AlphaScreen streptavidin-conjugated donor and ProteinA-conjugated acceptor beads; PerkinElmer) with HIF-1α hydroxy-Pro546 antibody (Cell Signalling) was added to the reaction mixture followed by incubation for 1 h in the dark at room temperature.
The luminescence signal was acquired with an Envision (Perkin Elmer) plate reader.

Reduction of the imine adducts formed by the aldehyde inhibitors 4
A SepPak cartridge (Waters) was washed with acetonitrile (2 mL) and then conditioned with solvent A (2 mL respectively. The 20 most intense peaks in the full MS scan were selected for fragmentation using either HCD, CID, or using a data-dependent decision tree (DDDT) to select between CID and ETD. In DDDT mode, ETD fragmentation was used for charge states 3, 4 and 5 with m/z less than 750 and for all charge states greater than 5. CID fragmentation was used for all other peptides. Dynamic exclusion was enabled (exclusion list size 75, exclusion duration 5 s).
The raw data files generated were processed using MaxQuant software (Version 1.4.1.2), integrated with the Andromeda search engine as described elsewhere. Data were searched against PHD2 protein sequence, as well as list of common contaminants defined by Andromeda. Protein and PSM false discovery rates (FDR) were set at 0.01. The minimum number of quantified peptides required per protein was set to 2. The maximum number of missed cleavages was set to be 2. Oxidation (M) and acetylation (N-term) were allowed as a variable modifications and carbamidomethylation (C) was used as a fixed modification. The covalent inhibitor modifications were created in Andromeda and allowed to occur on any nucleophilic residue. All assignments of modifications were validated by manual inspection of spectra.

Photoreaction studies
PHD2 (20 µM), Fe(II) (20 µM) and 6 (20 µM) in buffer (15 mM ammonium acetate, pH 7.5) were incubated for 4 h at 37 °C. The sample was cooled to 4 °C before irradiation with UV light (12 min, 310 nm, 4 °C) using a CaproBox TM . Products were analysed by denaturing liquid chromatography-mass spectrometry. Data were processed with MassLynx ™ v4.0 (Waters) and protein peaks were deconvoluted using the Maximum Entropy method (MaxEnt1). Data were processed with MassLynx ™ v4.0 (Waters) and protein peaks were deconvoluted using the Maximum Entropy method (MaxEnt1). The PCR reactions yielded product mixtures containing mutated plasmids and parental (non-mutated) dsDNA templates. It was necessary to remove the non-modified templates before transformation since they both contain the same antibiotic resistance gene. This was achieved by digesting the PCR products with the DpnI restriction enzyme (Stratagene Instruction Manual #200518). Since DpnI works on methylated/hemimethylated DNAs, the plasmid DNA templates were isolated from dam + E. coli strain so that the parent DNA could be digested. DpnI (1 μL) was added to the PCR products, gently mixed and centrifuged for 1 minute before being digested for 3 hours at 37ºC. For tranformation, an aliquot of XL10-Gold competent cells (50 μL) was transferrd to a 1.5 mL sterile Eppendorf vial and the plasmid/PCR product (1 μL) was added with gentle mixing. Cells were treated with β-mercaptoethanol (2 μL) to enhance transformation efficiency.

Preparation of PHD2 and variants
The transformation mixture was chilled on ice for 30 minutes, followed by a heat shock for 30 seconds at 42ºC to induce DNA uptake. The cells were incubated at 37ºC for one hour with 200 μl 2× TY medium and placed on LB-agar plates containing kanamycin (30 μg/mL). Finally, the plates were incubated at 37ºC. A single colony resulting from transformation was inoculated into 10 ml 2× TY medium containing kanamycin.
The starter culture was grown overnight at 37ºC with shaking at 220 rpm. Plasmid DNA was purified from the culture using the Qiagen Miniprep Kit following the manufacturer's protocol. All plasmids were verified by DNA sequencing at the Geneservice Ltd., Department of Biochemistry, University of Oxford. The results were compared to wildtype sequence to ensure the amplified insert or the mutated plasmid sequence was as desired and in the correct reading frame.

Recombinant protein production 6
A single colony was inoculated into 100 mL 2× TY media (overnight, 37ºC, shaken at 200 rpm) containing kanamycin (30 μg/mL). Then 7 ml of overnight starter culture was used to inoculate 2 L flasks containing 600 mL of 2× TY growth media with kanamycin (30 μg/mL). The flasks were incubated at 37ºC until the A 600 reached the range of 0.6-0.8, after which IPTG (600 μL, 0.5 mM) was added and growth was continued for an optimum incubation period (3-4 hours). Cells were harvested by centrifugation (9,000 rpm, 4ºC, 15 minutes) and the wet pellets were stored at -80ºC for protein purification. The frozen cell pellets were resuspended in 5 mL lysis buffer (see below) per gram of wet cell pellet. The cell suspension was then lysed by sonication on ice (4 × 45 sec sonication, 45 sec rest in between), using a SONICS VibraCell sonicator 150 sonicator, with a microtip at 60% amplitude. DNAse I (0.02 mg/mL) and MgCl 2 (5 mM) were added to the lysate to digest chromosomal DNA. The solution was left on ice for 20 minutes and then centrifuged (9,000 rpm, 20 minutes, The proteins were concentrated to around 2 mL using an Amicon ultra-15, Millipore centrifugal filter membrane, and loaded onto the column using a 2mL/ 10mL loop at 0.5 mL/min. After loading of proteins, the flow rate was increased to 2 mL/min with the UV baseline corrected to zero. The protein fractions were collected by an automatic fraction collector when the UV peak level rose above 25 mAU, in typical fractions of 5 mL. The eluted fractions were analysed for purity by SDS-PAGE. Those fractions containg pure protein were pooled and buffer exchanged into 50 mM Tris•HCl pH 7.5 using a PD-10 desalting column (Amersham Biosciences) and stored at -80ºC.

Synthesis
All chemicals, reagents, and solvents were from Sigma-Aldrich (Dorset, UK) and used without further

Methyl 6-(formyl)pyridine-2-carboxylate (S4z)
To a solution of compound S4y (589 mg, 3.52 mmol) in CHCl 3 (25 mL) was added manganese dioxide (6.00 g, 69 mmol) and the mixture was heated to reflux for 5 hrs. The reaction mixture was then cooled down and filtered through a pad of silica gel, washed with EtOAc and evaporated affording S4z as a white solid (212 mg, 36%).
Then it was cooled to room temperature and the solid residue was filtered. The filtrate was evaporated and resulting residue purified by column chromatography (PE to PE: EA 9:1) to afford 5.21 g (54%) of 4x as a colourless oil.

5-Bromo-2-hydroxy-3-nitrobenzoic acid (7x)
To a suspension of 5-bromosalicylic acid (23 g, 105 mmol) in concentrated sulfuric acid (70 ml) at 0°C, a mixture of nitric acid (6.6 ml) and conc. sulfuric acid (15.4 ml) was added dropwise whilst maintaining a reaction temperature of < 10°C. The reaction was stirred on ice for 1 hour before the ice bath was removed and the reaction stirred at room temperature for 3 hours at which point the reaction was observed to be complete by LCMS. The resulting mixture was decanted into a conical flask containing crushed ice, which gave a bright yellow precipitate which was collected by büchner filtration, washed with cold water and dried to leave a yellow solid. After recrystallised from EtOAc, a crystalline bright yellow solid  The reaction was cooled to room temperature before being quenched with 1.0M HCl (aq) (~ 10 ml). The resulting suspension was passed through a short pad of silica gel and the filtrate was extracted with EtOAc (3