Discovery of (5-Phenylfuran-2-yl)methanamine Derivatives as New Human Sirtuin 2 Inhibitors

Human sirtuin 2 (SIRT2), a member of the sirtuin family, has been considered as a promising drug target in cancer, neurodegenerative diseases, type II diabetes, and bacterial infections. Thus, SIRT2 inhibitors have been involved in effective treatment strategies for related diseases. Using previously established fluorescence-based assays for SIRT2 activity tests, the authors screened their in-house database and identified a compound, 4-(5-((3-(quinolin-5-yl)ureido)methyl)furan-2-yl)benzoic acid (20), which displayed 63 ± 5% and 35 ± 3% inhibition against SIRT2 at 100 μM and 10 μM, respectively. The structure-activity relationship (SAR) analyses of a series of synthesized (5-phenylfuran-2-yl)methanamine derivatives led to the identification of a potent compound 25 with an IC50 value of 2.47 μM, which is more potent than AGK2 (IC50 = 17.75 μM). Meanwhile, 25 likely possesses better water solubility (cLogP = 1.63 and cLogS = −3.63). Finally, the molecular docking analyses indicated that 25 fitted well with the induced hydrophobic pocket of SIRT2.

Next, the desired target compound 30, a hydroxamic acid derivative, was prepared by a three-step sequence starting from the synthesized intermediate 4a (Scheme 2). Sodium cyanoborohydride (NaBH 3 CN)-mediated reduction reaction was firstly performed to reduce the aldoxime group of intermediate 4a to the hydroxylamine of intermediate 27 (54% yield), followed by condensation with 2-phenylacetyl chloride in the presence of NaHCO 3 to give the compound 29. Further, hydrolysis of compound 29 using 3.0 equiv NaOH led to the white solid target compound 30. The synthesis of target compounds 32-37 are also depicted in Scheme 2. The reactions of commercially available amines (aniline, phenylmethanamine, and pyridin-3-ylmethanamine) or hydrazide (nicotinohydrazide) with intermediates 3a or 3i in the presence of hantzschester (1.2 equiv), catalytic amount of molecular sieve and trifluoroacetic acid, resulted in the reductive amination products 31-34. The resulting compounds 31-33 were subsequently hydrolyzed to give the desired compounds 35-37 in high yields. Finally, Scheme 3 presents the synthetic routes for compounds 39 and 43-52, which contain a sulfonamide or amide linker. For sulfonamide linker compound 39, intermediate 5a was used to react with benzenesulfonyl chloride in the presence of Et3N at room temperature, and the resulting compound 38 underwent a hydrolysis reaction to give the desired target compound 39, in 80% yield for two steps. The synthetic access to structurally diverse amide linker compounds 41-48 was achieved using a condensation reaction of carboxylic acid (40) with amine (5a, 5c-5f) in the presence of 1-hydroxybenzotriazole (HOBT), 1-(3-(dimethylamino)propyl)-3-ethylcarbodiimide hydrochloride (EDCI), and N,N-diisopropylethylamine (DIPEA). The resulting ester-contained compounds 41, 42, 46 and 47 were subjected to hydrolyzation to afford the target compounds 49-52 in good yields. Finally, Scheme 3 presents the synthetic routes for compounds 39 and 43-52, which contain a sulfonamide or amide linker. For sulfonamide linker compound 39, intermediate 5a was used to react with benzenesulfonyl chloride in the presence of Et 3 N at room temperature, and the resulting compound 38 underwent a hydrolysis reaction to give the desired target compound 39, in 80% yield for two steps. The synthetic access to structurally diverse amide linker compounds 41-48 was achieved using a condensation reaction of carboxylic acid (40) with amine (5a, 5c-5f) in the presence of 1-hydroxybenzotriazole (HOBT), 1-(3-(dimethylamino)propyl)-3-ethylcarbodiimide hydrochloride (EDCI), and N,N-diisopropylethylamine (DIPEA). The resulting ester-contained compounds 41, 42, 46 and 47 were subjected to hydrolyzation to afford the target compounds 49-52 in good yields.
Collectively, the structural optimization and SAR studies led to the discovery of compound 25, which exhibited high potency against SIRT2, better than the hit compound 20 and positive control AGK2. Subsequently, the IC 50 value of 25 was then measured against SIRT2, and the IC 50 curve has been presented in Figure 3. The study observed that compound 25 inhibited SIRT2 via a dose dependent manner with an IC 50 value of 2.47 µM, which is more potent than AGK2 (with an IC 50 value of 17.75 µM). Molecular docking was then used to investigate the possible binding mode of 25 with SIRT2. The results indicated that 25 appeared to fit well with the induced hydrophobic pocket (Figure 4) [44,45]. The carboxyl acid group of 25 is likely positioned to make hydrogen-bonding interactions with the main chain of Asp170 and the side chain of Thr171 and Tyr139. The furan and pyridine moiety likely have hydrophobic contacts with hydrophobic residues Phe119, Phe234, Phe131, Leu138, and Ile169 ( Figure 4). Notably, the pyridine appears to form edge-to-face aromatic interactions with Phe119, and fits well with the pocket around Phe119, Phe131, and Phe234, suggesting that introducing substituents on pyridine may result in a clash with these three residues. Together, these docking results may explain why the replacement of the carboxyl acid group or the introduction of substituents on pyridine leads to a decrease in SIRT2 inhibition, and indicates the possible inhibition mode for this series of compounds.  1 Each compound was tested in triplicate; the data are presented as the mean ± SD (n = 2).
Collectively, the structural optimization and SAR studies led to the discovery of compound 25, which exhibited high potency against SIRT2, better than the hit compound 20 and positive control AGK2. Subsequently, the IC50 value of 25 was then measured against SIRT2, and the IC50 curve has been presented in Figure 3. The study observed that compound 25 inhibited SIRT2 via a dose dependent manner with an IC50 value of 2.47 μM, which is more potent than AGK2 (with an IC50 value of 17.75 μM). Molecular docking was then used to investigate the possible binding mode of 25 with SIRT2. The results indicated that 25 appeared to fit well with the induced hydrophobic pocket   1 Each compound was tested in triplicate; the data are presented as the mean ± SD (n = 2).
Collectively, the structural optimization and SAR studies led to the discovery of compound 25, which exhibited high potency against SIRT2, better than the hit compound 20 and positive control AGK2. Subsequently, the IC50 value of 25 was then measured against SIRT2, and the IC50 curve has been presented in Figure 3. The study observed that compound 25 inhibited SIRT2 via a dose dependent manner with an IC50 value of 2.47 μM, which is more potent than AGK2 (with an IC50 value of 17.75 μM). Molecular docking was then used to investigate the possible binding mode of 25 with SIRT2. The results indicated that 25 appeared to fit well with the induced hydrophobic pocket 40  Collectively, the structural optimization and SAR studies led to the discovery of compound 25, which exhibited high potency against SIRT2, better than the hit compound 20 and positive control AGK2. Subsequently, the IC50 value of 25 was then measured against SIRT2, and the IC50 curve has been presented in Figure 3. The study observed that compound 25 inhibited SIRT2 via a dose dependent manner with an IC50 value of 2.47 μM, which is more potent than AGK2 (with an IC50 value of 17.75 μM). Molecular docking was then used to investigate the possible binding mode of 25 with SIRT2. The results indicated that 25 appeared to fit well with the induced hydrophobic pocket 20  Collectively, the structural optimization and SAR studies led to the discovery of compound 25, which exhibited high potency against SIRT2, better than the hit compound 20 and positive control AGK2. Subsequently, the IC50 value of 25 was then measured against SIRT2, and the IC50 curve has been presented in Figure 3. The study observed that compound 25 inhibited SIRT2 via a dose dependent manner with an IC50 value of 2.47 μM, which is more potent than AGK2 (with an IC50 value of 17.75 μM). Molecular docking was then used to investigate the possible binding mode of 25 with SIRT2. The results indicated that 25 appeared to fit well with the induced hydrophobic pocket  Collectively, the structural optimization and SAR studies led to the discovery of compound 25, which exhibited high potency against SIRT2, better than the hit compound 20 and positive control AGK2. Subsequently, the IC50 value of 25 was then measured against SIRT2, and the IC50 curve has been presented in Figure 3. The study observed that compound 25 inhibited SIRT2 via a dose dependent manner with an IC50 value of 2.47 μM, which is more potent than AGK2 (with an IC50 value of 17.75 μM). Molecular docking was then used to investigate the possible binding mode of 25 18  Collectively, the structural optimization and SAR studies led to the discovery of compound 25, which exhibited high potency against SIRT2, better than the hit compound 20 and positive control AGK2. Subsequently, the IC50 value of 25 was then measured against SIRT2, and the IC50 curve has been presented in Figure 3. The study observed that compound 25 inhibited SIRT2 via a dose dependent manner with an IC50 value of 2.47 μM, which is more potent than AGK2 (with an IC50 value of 17.75 μM). Molecular docking was then used to investigate the possible binding mode of 25 3  Phe131, Leu138, and Ile169 ( Figure 4). Notably, the pyridine appears to form edge-to-face aromatic interactions with Phe119, and fits well with the pocket around Phe119, Phe131, and Phe234, suggesting that introducing substituents on pyridine may result in a clash with these three residues. Together, these docking results may explain why the replacement of the carboxyl acid group or the introduction of substituents on pyridine leads to a decrease in SIRT2 inhibition, and indicates the possible inhibition mode for this series of compounds.

Synthesis
As previously reported, proton ( 1 H) and carbon ( 13 C) NMR spectra were recorded on a Bruker AV-400 (Bruker Company, Billerica, Germany) instrument and are reported in ppm relative to tetramethylsilane (TMS) and referenced to the solvent in which the spectra were collected. Unless otherwise noted, all of the commercially available starting materials, reagents, and solvents and reagents were used without further purification. The analytical thin-layer chromatography (TLC) was run on Merck silica gel 60 F-254 (Qingdao Haiyang, Qingdao, China). The spots on the plates were visualized under UV light (λ = 254 nm). Purification was performed on silica gel chromatography with EtOAc-petroleum ether or CH2Cl2-MeOH solvent systems. The melting points were measured on an electrothermal melting point apparatus without correction (JIAHANG, Shanghai, China. ESI-MS was obtained on a Shimadzu-2010EV series liquid chromatograph mass spectrometer (Shimadzu, Tokyo, Japan). High-resolution mass spectra (HRMS) were determined using a SCIEX X500 QTOF mass spectrometer (Shanghai Sciex Analytical Instrument Trading Co., Shanghai, China). All target compounds were purified to >95% purity, as determined by the high-performance liquid chromatography (HPLC). The HPLC analysis was performed on a Waters 2695 HPLC system equipped with a Kromasil C18 column (4.6 mm × 250 mm, 5 μm, Waters, Milford, MA, USA).

Synthesis
As previously reported, proton ( 1 H) and carbon ( 13 C) NMR spectra were recorded on a Bruker AV-400 (Bruker Company, Billerica, Germany) instrument and are reported in ppm relative to tetramethylsilane (TMS) and referenced to the solvent in which the spectra were collected. Unless otherwise noted, all of the commercially available starting materials, reagents, and solvents and reagents were used without further purification. The analytical thin-layer chromatography (TLC) was run on Merck silica gel 60 F-254 (Qingdao Haiyang, Qingdao, China). The spots on the plates were visualized under UV light (λ = 254 nm). Purification was performed on silica gel chromatography with EtOAc-petroleum ether or CH 2 Cl 2 -MeOH solvent systems. The melting points were measured on an electrothermal melting point apparatus without correction (JIAHANG, Shanghai, China. ESI-MS was obtained on a Shimadzu-2010EV series liquid chromatograph mass spectrometer (Shimadzu, Tokyo, Japan). High-resolution mass spectra (HRMS) were determined using a SCIEX X500 QTOF mass spectrometer (Shanghai Sciex Analytical Instrument Trading Co., Shanghai, China). All target compounds were purified to >95% purity, as determined by the high-performance liquid chromatography (HPLC). The HPLC analysis was performed on a Waters 2695 HPLC system equipped with a Kromasil C18 column (4.6 mm × 250 mm, 5 µm, Waters, Milford, MA, USA).
To a solution of the coupling products 3a-3i (12 mmol) in EtOH (25 mL), hydroxylamine hydrochloride (NH 2 OH.HCl, 14.4 mmol) and sodium acetate (NaOAc, 14.4 mmol) were added and the mixture was stirred at reflux for 0.5 h. When TLC indicated that the reaction was finished, the reaction solution was concentrated and the residue was partitioned between water (50) and ethyl acetate (3 × 50 mL). The combined organic layer was dried over MgSO 4 , filtered and concentrated in vacuo to give the crude products 4a-4i, which were used without further purification. Subsequently, to a stirring solution of condensation products, 4a-4i (12 mmol) in EtOH (25 mL) was added to zinc powder (Zn, 12 mmol) and 3 M hydrochloric acid (HCl, 8.0 mL) at ambient temperatures. The reaction mixture was heated to 80 • C for further 2 h. After completion (monitored by TLC), the solvent was removed in vacuo, the crude residue was treated with 100 mL of ice water, and the pH was adjusted to 7-8 with saturated NaHCO 3 . Then, the mixture was filtered by diatomite and extracted with ethyl acetate (3 × 80 mL). The combined extracts were dried, concentrated and purified by column chromatography with appropriate eluents with three ethylamine (Et 3 N, TEA) to afford the desired intermediates 5a-5i in high yields.

Hantzsch-Involved Reductive Amination Used for Compounds 31-34
To a solution of substituted 5-phenylfuran-2-carbaldehydes (3a and 3i, 1.5 mmol), different amines (1.8 mmol) and diethyl 2,6-dimethyl-1,4-dihydropyridine-3,5-dicarboxylate (hantzschester, 1.8 mmol) in DCM (25 mL), catalytic amount of molecular sieve and trifluoroacetic acid were added at room temperature, and the reaction was warmed to 45 • C and reacted for 6-12 h. After completion (monitored by TLC), the reaction was filtered, and the crude residue was obtained by concentrating the filtrate in vacuo. Finally, the crude residue was purified by column chromatography to give the desired compounds 31-34 in high yields.

Inhibition Assays
This study tested the inhibitory activities of the synthesized compounds against recombinant human SIRT2 proteins using a fluorogenic substrate p2270(Ac-Glu-Thr-Asp-Lys(Dec)-AMC)-coupled trypsin assay. The assay buffer is 25 mM Tris-HCl pH 8.0, 150 mM NaCl, and 10% glycerol. The test compounds were added to 60 µL of reaction mixture containing SIRT2 enzymes (0.2 µM), and each compound was prepared in a 3-fold dilution series (300 µM-15 nM) with the final DMSO concentration < 1%. After incubation at 25 • C for 30 min, the reaction started by the addition of the substrate p2270 (10 mM) and NAD + (400 mM) at 25 • C. After 2 h, 50 µL 3~4 U/µL trypsin and 4 mM nicotinamide were added to terminate the reaction, followed by further incubation for 30 min at 25 • C. The fluorescence intensity was measured using a microplate reader (λ ex = 380 nm, λ em = 460 nm). All determinations were performed in triplicate. The IC 50 values were obtained using GraphPad Prism software as described previously.

Molecular Docking Assays
All the docking simulations were performed using AutoDock Vina. The crystal structure of SIRT2 complexed with an N-(3-(phenoxymethyl)phenyl)acetamide derivative (24a) (PDB ID: 5YQO) and was used as the docking template. All the water and solvant molecules, as well as 24a were removed, and clean protein structure coordinates were obtained. AutoDockTools was used to assign Gasteiger-Marsili charges to the protein structure model, and merge non-polar hydrogens onto their respective heavy atoms of the protein structure (saved as pdbqt format). The 3D coordinates of the compound structures were prepared using the Discovery Studio viewer, followed by assigning atom types and partial charges using AutoDockTools (saved as pdbqt format). The binding site was defined as a rectangular grid, with the grid center coordinates of [x, y, z = −13.5, −10.1, −18.4] and the grid size of [25,25,25], to encompass the entire binding site. The number of possible docking poses were set as 10, and the other docking parameters were set as default. The docking results were inspected using PyMOL.

Conclusions
In this study, a series of (5-phenylfuran-2-yl)methanamine derivatives were synthesized. The SAR analyses of these compounds with SIRT2 led to the identification of compound 25 with 99 ± 2% @ 100 µM and 90 ± 3 % @ 10 µM inhibition against SIRT2. Meanwhile, 25 likely possesses better water solubility (cLogP = 1.63 and cLogS = −3.63). The IC 50 measurements revealed that 25 had considerable potency against SIRT2 with an IC 50 value of 2.47 µM, which is more potent than AGK2. The molecular docking analyses indicated that 25 fits well with the induced hydrophobic pocket of SIRT2. This study will aid future investigations to discover new potent and selective SIRT2 inhibitors to provide potential treatments for relevant diseases.