- Email: [email protected]
Identification of a pyrimidinetrione derivative as the potent DprE1 inhibitor by structure-based virtual ligand screening
Bioorganic Chemistry 85 (2019) 168–178
Contents lists available at ScienceDirect
Bioorganic Chemistry journal homepage: www.elsevier.com/locate/bioorg
Identification of a pyrimidinetrione derivative as the potent DprE1 inhibitor by structure-based virtual ligand screening Ya Gaoa, Jinshan Xiea, Ruotian Tanga, Kaiyin Yanga, Yahan Zhanga, Lixia Chenb, , Hua Lia,b, ⁎
T
⁎
a
Hubei Key Laboratory of Natural Medicinal Chemistry and Resource Evaluation, School of Pharmacy, Tongji Medical College, Huazhong University of Science and Technology, Wuhan 430030, China b Wuya College of Innovation, School of Traditional Chinese Materia Medica, Key Laboratory of Structure-Based Drug Design & Discovery, Ministry of Education, Shenyang Pharmaceutical University, Shenyang 110016, China
ARTICLE INFO
ABSTRACT
Keywords: Pyrimidinetrione DprE1 Zinc database Virtual ligand screening Tuberculosis
Despite the increasing need of new antituberculosis drugs, the number of agents approved for the market has fallen to an all-time low. In response to the emerging drug resistance followed, structurally unique chemical entities will be highlighted. decaprenylphosphoryl-β-D-ribose oxidase (DprE1) participating in the biosynthesis of mycobacterium cell wall is a highly vulnerable and validated antituberculosis target. On the basis of it, a systematic strategy was applied to identify a high-quality lead compound (compound 50) that inhibits the essential enzyme DprE1, thus blocking the synthesis of the mycobacterial cell wall to kill M. tuberculosis in vitro and in vivo. Correspondingly, the rational design and synthetic strategy for compound 50 was reported. Notably, the compound 50 has been confirmed to be no toxicity. Altogether, our data suggest the compound 50 targeting DprE1 is a promising candidate for the tuberculosis (TB) therapy.
1. Introduction It was claimed that M. tuberculosis, an intracellular pathogenic bacteria, emerged since the times known to us [1]. The terrible thing is that the transmission, evolution, and existence of the bacteria till now have shown enormously success, which plagues humanity all the time [1,2]. According to the World Health Organization (WHO) reports [3], tuberculosis (TB) is one of the leading cause of death in infectious disease in the world. Treatment of TB has been always depending on the drugs used for a long time, such as isoniazid, rifampicin, ethambutol, pyrazinamide, and so on. While, the multidrug resistant (MDR-TB) and extensively drug resistant (XDR-TB) render these drugs less effective. Unfortunately, no new and effective anti-TB drugs have been developed in the past 40 years. The research on drugs against drug-resistant strains of M. tuberculosis with special structures and patterns of action remains a significant challenge [4]. DprE1 is reported to be essential for survival of M. tuberculosis including multi-drug resistant strains [5]. The enzyme is crucial and conservative for mycobacterial cell wall biosynthesis. DprE1 catalyzes the FAD-dependent oxidation of decaprenylphosphoryl-β-D-
ribose (DPR) to decaprenyl-phosphoryl-2′-keto-D-erythro- pentofuranose (DPX). Then DPX is then reduced by decaprenyl-phospho-2′-ketoD- arabinose reductase (DprE2) to generate DPA, which is the indispensable component of mycobacterium cell wall [5–7]. Since there is no alternative way to synthesis DPX, DprE1 becomes a druggable target. Correspondingly, DprE1 inhibitors will be low toxic to human cell because there are no cell walls for human and thus has no similar metabolic pathway. Targeting DprE1 with small-molecule inhibitors has emerged as a promising strategy for anti-TB therapy. For the past few years, DprE1 inhibitors, like BTZs and TCA1, were found to cause mycobacteria death by covalently or non-covalently binding to the enzyme DprE1. These compounds have showed promising anti-tubercular efficacy, and provided powerful evidence for designing DprE1 inhibitors and treating TB [8,9]. In this study, 63 compounds were cherry-picked from more than 6.2 million small molecules through virtual ligand screening. Compound 50 was further confirmed to be a significantly effective bactericidal agent compared with isoniazid (INH) through inhibition test using M. smegmatis and M. tuberculosis in vitro. From these combined screening efforts, we identified a novel inhibitor that targets DprE1. Furthermore, in the
Abbreviations: DprE1, decaprenylphosphoryl-β-D-ribose oxidase; MDR, multidrug resistant; XDR, extensively drug resistant; MST, microscale thermophoresis; TB, tuberculosis; DPR, decaprenylphosphoryl-β-D-ribose; DPX, decaprenyl-phosphoryl-2′-keto-D-erythro- pentofuranose; INH, isoniazid; DprE2, Decaprenyl-phospho-2′keto-D-arabinose reductase; mp, melting point; MST, microscale thermophoresis ⁎ Corresponding authors. E-mail addresses: [email protected] (L. Chen), [email protected] (H. Li). https://doi.org/10.1016/j.bioorg.2018.12.018 Received 19 October 2018; Received in revised form 30 November 2018; Accepted 12 December 2018 Available online 13 December 2018 0045-2068/ © 2018 Elsevier Inc. All rights reserved.
Bioorganic Chemistry 85 (2019) 168–178
Y. Gao et al.
MST assay, compound 50 was verified to specifically bind to DprE1 with high affinity. It is gratifying that compound 50 showed comparable therapeutic effect to INH for the treatment of acute infections in mice. It was noteworthy that this compound showed no any obvious toxicity in cytotoxicity testing and acute toxicity assessment. Our study yielded new antituberculosis lead compounds, and the structure-based drug discovery strategy will also provide an effective method for other major diseases.
Table 1 Binding affinity and inhibitory activities of compounds in vitro.
2. Materials and methods 2.1. Compounds database a. Subset of ZINC database, ZINC lead like compound library containing 6,053,287 molecules [8]. b. ZINC natural compounds library, a total of approximately 150,000 compounds, including IBS screen, Princeton NP, Taiwan and AnalytiCon Discovery NP database, is the relatively large database that has undergone ZINC screening of drugs and remove duplicate or similar structure [10].
No.a
Scoreb (Kal·cal−1)
Kdc (nM)
Inhibition zoned (mm)
1 10 16 19 28 33 34 38 49 50 54 62 INH
−44.93 −43.87 −38.21 −38.61 −37.7 −37.19 −37.86 −36.9 −37.13 −37.8 −40.58 −37.62
> 500000 > 500000 65600 ± 6140 891 ± 109 > 500000 > 500000 154000 ± 13500 > 500000 14500 ± 1020 25 ± 2 > 500000 11400 ± 663 > 500000
20 16 14 24 20 16 14 14 14 20 14 16 20
a Among the 63 compounds from virtual screening, 12 of them displayed notable activity against M. smegmatis, with inhibition zone greater than 14 mm. b Docking score/interaction potential of compounds with DprE1 (kcal/mol). c The Kd value was automatic calculated by the curve fitting. d Inhibition zone was conducted through the Kirby–Bauer disc agar diffusion assay.
2.2. Structure-based virtual ligand screening
DMSO to 25 mM. INH was also diluted in DMSO to 500 mg/L as positive control. And the solvent DMSO was used as negative control. Sterile discs which were impregnated with 10 μL of DMSO or compounds were put onto 7H10 (BD) plate. The plates were incubated for 30 h at 37 °C. Zones of inhibitor were measured to estimate the antibacterial effect [15,16].
ICM 3.8.2 modeling software on an Intel i7 4960 processor (MolSoft LLC, San Diego, CA) was used to identify possible drug candidates for DprE1 enzyme. For the structure-based virtual screening, ligand was continuously resiliently made to dock with DprE1 (PDB ID: 4FDO) [7] that was represented in potential energy maps and obtained online (http://www.rcsb.org/pdb). 3D compounds obtained from ZINC were scored according to the internal coordinate mechanics (Internal Coordinate Mechanics, ICM) [11–13]. Based on Monte Carlo simulations, stochastic global optimization procedure and pseudo-Brownian positional/torsional steps, the position of intrinsic molecular was optimized [14]. By visually inspecting, compounds outside the active site, as well as those weakly fitting to the active site were eliminated. According with Lipinski (Rule of 5) and scoring more than −32 (generally represents strong interactions) was priority to be selected.
2.5. In vitro binding affinity Recombinant DprE1 was labeled with the Monolith NT™ Protein Labeling Kit RED (Cat#L001) according to the supplied protocol. Labeled DprE1 was kept constant at 100 nM. All samples tested were diluted in a 20 mM HEPES (pH 7.5) and 0.05 (v/v) % Tween-20. Compounds were diluted covering the range from 500 μM to 2 nM. After 10 min incubation at room temperature, samples were loaded into Monolith™ standard-treated capillaries and the thermophoresis was measured at 25 °C after 30 min incubation on a Monolith NT.115 instrument (NanoTemper Technologies, München, Germany) [17]. The LED power was set to 100%. The dissociation constant Kd values were fitted by using the NT Analysis software (NanoTemper Technologies, München, Germany).
2.3. Protein expression and purification The gene of full-length decaprenylphosphoryl-β-D-ribose oxidase was cloned from M. smegmatis strain ATCC 607 (China General Microbiological Culture Collection Center). The gene was ligated into PET28a vector (Novagen). After the recombinant plasmid was verified by sequencing, it was transformed into E. coli strain BL21 (Invitrogen) at 293 K which was grown in LB medium at 37 °C to an OD600 (0.8–1.0) and induced by 0.4 mM isopropyl-D-thiogalactopyranoside (IPTG) at 18 °C for 16 h. Bacterial cells were lysed on ice in buffer containing 100 mM Tris-HCl pH 8.8, 200 mM NaCl, 10% glycerol, 1% TritonX-100, 5 mM β-mercaptoethanol. Soluble N-terminally hexa-histidine tagged DprE1 was bound to Ni-agrose affinity resin (Qiagen), washed with a buffer containing 20 mM Tris-HCl pH 8.8, 200 mM NaCl and 10 mM imidazole and eluted with a buffer containing 20 mM Tris-HCl pH 8.8, 250 mM NaCl, and 150 mM imidazole. The eluted protein was concentrated and diluted with a buffer containing 20 mM Tris-HCl pH 8.8, 250 mM NaCl. The protein was further purified with size exclusion chromatography (GE Health) at 20 mM Tris-HCl pH 8.8 and 200 mM NaCl [5].
2.6. In vitro antibacterial activity test using M. tuberculosis In vitro antibacterial activity test using M. tuberculosis was carried in accordance with MABA. 100 µL sterile normal saline was added to outer-perimeter wells of clean 96-well plates (Falcon; BD), to avoid the error by evaporation of the liquid during the test. Every three horizontal tested one compound which was diluted by 7H9 broth in 9 dilution steps. All compounds were done in identical serial 1:1 dilutions. And the wells in column 11 were served as drug-free controls. After adding 100 μL drug liquid, 100 μL 7H9 broth seeded M. tuberculosis was put into every well tested. The microplates were resealed with Parafilm and were incubated for an additional 24 h at 37 °C. Fifty microliters of a freshly prepared 1:1 mixture of Alamar Blue (Accumed International, Westlake, Ohio) reagent and 10% Tween 80 were added to drug-free wells. If drug-free wells turned pink, the reagent mixture was added to all wells in the microplates. The microplates were resealed with Parafilm and were incubated for an additional 24 h at 37 °C, and the colors of all wells were recorded. A blue color in the well was interpreted as no growth, and a pink color was scored as growth. A few wells appeared violet after 24 h of incubation, but they invariably changed into pink after another day of incubation and thus were scored as growth (while the adjacent blue wells remained blue) [18]. The MIC
2.4. In vitro antibacterial activity test using M. smegmatis Because of the highly infectious of M. tuberculosis, the homology and safety M. smegmatis ATCC 607 (China General Microbiological Culture Collection Center) was identified as a good model to screen M. tuberculosis inhibitors. It was selected to test the effects of all compounds using a Kirby-Bauer disc diffusion assay. All compounds were diluted in 169
Bioorganic Chemistry 85 (2019) 168–178
Y. Gao et al.
Fig. 1. MST assay to measure the binding between compounds and DprE1. Compound 50 displayed the equilibrium dissociation constant (Kd) of 25 ± 2 nM. Kd values were automatic calculated by the curve fitting.
was defined as the lowest drug concentration which prevented a color change from blue to pink. And the values will be recorded on Microplate Reader at 520/590 nm.
2.7.2. Synthetic procedure for allylurea 2 A 100 mL three-necked bottle was charged with allylamine hydrochloride (0.1 mol), KOCN (0.11 mol, 1.1 equiv), H2O (50 mL), and stirred at 55 °C for 3 h, the progress of the reaction was monitored by TLC. Upon completion of the reaction, the solvent was evaporated under vacuum. The residue was dispersed with 100 mL ethanol then filtered, washed with ethanol. The filtrate was evaporated under vacuum to afford the target allylurea 2 as white solid. Yield 76.8%. 1H NMR (400 MHz, DMSO‑d6) δ 6.05–5.97 (m, 1H), 5.81–5.72 (m, 1H), 5.44 (s, 2H), 5.10–5.04 (m, 1H), 5.00–4.96 (m, 1H), 3.57–3.55 (m, 2H).
2.7. Synthesis of compound 50 2.7.1. General experimental methods All reactions were carried out in flame-dried glassware under an atmosphere of argon, all reagents were procured from commercial sources (purity > 99%) and used without further purification. Silica gel GF254 and silica gel (200–300 mesh) were respectively used for thin-layer chromatography and column chromatography. NMR spectra were recorded on a Bruker AM-400 spectrometer, with 1 H and 13C NMR chemical shifts referenced to the solvent or solvent impurity peaks for DMSO‑d6 (δH 2.50 and δC 39.52). HRESIMS data were acquired using a Thermo Fisher LC-LTQ-Orbitrap XL spectrometer and an electrospray ionization and a hybrid quadrupole time-of-flight (q-TOF) mass spectrometer (model 6540, Agilent).
2.7.3. Synthetic procedure for pyrimidinetrione 3 To a 100 mL three-necked bottle was added allylurea 2 (0.1 mol), dimethyl malonate (0.1 mol, 1.0 equiv), NaOMe (0.12 mol, 1.2 equiv) and MeOH (100 mL), the mixture was refluxed for 6 h until the completion of the reaction monitored by TLC. The solvent was evaporated under vacuum and H2O (50 mL) was added, the mixture was cooled in ice water and the solid filtered, washed with water to give pure 170
Bioorganic Chemistry 85 (2019) 168–178
Y. Gao et al.
Fig. 2. Chemical structures of the selected compounds.
pyrimidinetrione 3 as white solid. Yield 71.5%. 1H NMR (400 MHz, DMSO‑d6) δ 11.34 (s, 1H), 5.79–5.69 (m, 1H), 5.16–5.12 (m, 1H), 5.07–5.04 (m, 1H), 4.25–4.23 (m, 2H).
2.8. Cytotoxicity testing Vero Cells and L929 were grown in Roswell Park Memorial Institute medium (RPMI), supplemented with penicillin G (100 UI/mL), streptomycin (100 mg/mL) and 10% fetal calf serum at 37 °C in 5% CO2 atmosphere. Lo2 were grown in Dulbecco’s modified eagle medium (DMEM), supplemented with penicillin G (100 UI/mL), streptomycin (100 mg/mL) and 10% fetal calf serum at 37 °C in 5% CO2 atmosphere. For cytotoxicity evaluation of compounds, the Vero, L929 and Lo2 cells were plated at a density of 1 × 104 cells/well in a 96 well plate at 37 °C in 5% CO2 atmosphere. Cells were then cultivated for 24 h, medium in the wells was replaced with fresh medium containing compound 50 which was diluted in 8 dilution steps covering the range from 500 μM to 3.9 μM. And DMSO was used as negative control. After 48 h incubation, 10 μL of CCK8 solution was added to each well. After 30 min, the absorbance of each well was measured at wavelength of 450 nm on spectrophotometer. Data was corrected for background (nocell control) and expressed as a percentage of the value for untreated cells. IC50 was calculated by the test and cytotoxicity of the compound to normal cells can be evaluated.
2.7.4. Synthetic procedure for formylpyrimidinetrione 4 To a 100 mL three-necked bottle was added pyrimidinetrione 3 (10 mmol), trimethylorthoformate (0.1 mol, 10 equiv), PTSA·H2O (1.0 mmol, 0.1 equiv), the mixture was stirred for 6 h at 80 °C until the completion of the reaction monitored by TLC. The solvent was evaporated under vacuum and the residue was recrystallized from ethanol to give formylpyrimidinetrione 4 as yellow solid. Yield 50.1%. 1H NMR (400 MHz, DMSO‑d6) δ 10.47–10.46 (m, 2H), 5.82–5.73 (m, 1H), 5.04–4.99 (m, 2H), 4.51 (s, 1H), 4.28–4.27 (d, J = 4.0 Hz, 2H). 2.7.5. Synthetic procedure for compound 50 To a solution of formylpyrimidinetrione 4 (10 mmol) in 100 mL dry THF was added isoniazid (10 mmol, 1.0 equiv), the mixture was stirred for 4 h under room temperature. The solvent was evaporated under vacuum and the residue was washed with ethanol to afford the target compound 50 as yellow solid. Yield 88.4%. mp. 150–152 °C. 1H NMR (400 MHz, DMSO‑d6) δ 11.14–11.02 (m, 1H), 8.78–8.77 (m, 2H), 8.21–8.20 (m, 1H), 7.77–7.76 (m, 2H), 5.85–5.75 (m, 1H), 5.09–5.01 (m, 2H), 4.32 (t, J = 4.0 Hz, 2H), 3.43–3.38 (m, 1H). 13C NMR (100 MHz, DMSO‑d6) δ 164.52, 164.28, 163.69, 163.50, 162.73, 162.55, 156.34, 156.30, 151.01, 150.95, 150.72, 139.46, 139.42, 133.61, 133.54, 121.83, 116.50, 116.22. HRMS (ESI) m/z 316.1035 [M +H]+, m/z 338.0840 [M+Na]+ (calcd for C14H13N5O4, 315.0968).
2.9. In vivo efficacy of compound 50 in mouse acute infection model Forty specific pathogen-free, female C57BL/6 mice aged 6 weeks were obtained from Beijing HFK Bio-Technology. All experiments were done by Biosafety of ABSL3 Lab of Wuhan University, State Key Laboratory of Virology. All mice were randomly divided into ten cages, feeding one week, to adapt to the environment. Mice were infected by nasal drops with one dose of 1 × 106 CFU of M. tuberculosis H37Rv (provided by State Key Laboratory of Virology). Infection dose was 171
Bioorganic Chemistry 85 (2019) 168–178
Y. Gao et al.
Fig. 3. Low-energy binding conformations of compound 50 bound to the active site of DprE1 generated by virtual ligand docking. A. Compound 50 depicted as the ball-and-stick model showing carbon (yellow), hydrogen (grey), oxygen (red) and nitrogen (blue) atoms. Compound 50 formed obvious hydrophobic interactions with DprE1 and FAD. B. Hydrogen bonds were predicted between compound 50 with DprE1. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
verified by incubating the inoculation fluid, whole-lung and wholespleen homogenate on 7H10 plates for 48 h postinfection. Mice treated with PBS were negative controls. After two days from infected, the lung and spleen bacterial load were assessed to make sure the real infected dose was 100 CFU per mouse. After infected, mice were divided randomly into five groups, nine of one group. After making sure the dose of the infection, C57BL/6 mice were administered by oral gavage with PBS, INH (10 mg/kg) in two groups. The other three groups, C57BL/6
mice were administered by oral gavage with 10 mg/kg, 30 mg/kg of compound 50. After treatment of ten days, the drug therapy efficiency was ascertained by evaluation of the bacterial load in lung, spleen and their histopathology compared with negative group and organs before treated. Organs were homogenized in PBS and various dilutions were placed on 7H11 (bought from BD) plates containing ampicillin (100 mg/L), carbenicillin (50 mg/L), amphotericin B (2.5 mg/L), polymyxin B (25 mg/L), TCH (2 mg/L). Plates were incubated at 37 °C for 172
Bioorganic Chemistry 85 (2019) 168–178
Y. Gao et al.
Fig. 4. The anti-TB activity of selected compounds in vitro. The values were recorded by Microplate Reader at 520/590 nm (n = 3).
3 week, CFU were recorded of all groups [19–21]. And histochemical staining and acid-fast staining were performed of the lung tissue following formaldehyde fixation. All animals were maintained in accordance with the Regulations of Experimental Animal Administration issued by the State Committee of Science and Technology of People’s Republic of China. All animal experimental protocols were approved by the Laboratory Animal Center of Huazhong University of Science and Technology and the Ethics Committee.
2.11. Statistical analysis All data are reported as means ± SD. Differences were analyzed by Student’s t-test (with 95% confidence interval). p values < 0.05 were considered as statistically significant. 3. Results 3.1. Structure-based virtual ligand screening identified a small molecule set binding to DprE1
2.10. Acute toxicity
The virtual screening had become an efficient, specifically targeted technology compared to traditional high-throughput screening (HST) [11–13]. DprE1 (PDB ID: 4FDO) was virtually screened against ∼6,200,000 small molecules from ZINC lead like database and natural products database [10]. According to the occupancy of the active binding pocket, many scaffolds were further excluded. Lipinski Rule (Rule of 5) is a rule of thumb to evaluate druglikeness. Scoring more negative than −32 generally represents strong interaction between target and ligand in ICM-Pro. According to Rule of 5 and scoring more negative than −32, the final set was reduced to a total of 63 compounds from 6.2 million (Table 1, Table S1). All of these 63 compounds have not been used in clinic and were purchased from Molport in our study. Moreover, none of them was reported as DprE1 inhibitors, or
KM mice of both sex (10 females and 10 males) were divided into control and experimental groups, which were obtained from Laboratory animal center Tongji Medical College, Huazhong University of Science and Technology. The first group served as a normal control treated with distilled water (10 mL/kg). And compound groups were orally administrated compound 50 on 2000 mg/kg a single time among 24 h. Along the 14 consecutive days, the signs of toxicity were observed every day. The changes were also noted on body weight, mortality, behavior of the animal. Acute toxicity caused a set of experimental animals with 50% subject compounds dose for quantitative evaluation of death (LD50). Formaldehyde fixation and histochemical staining of the lung, heart, liver, spleen and kidney tissue were assessed at the end [22]. 173
Bioorganic Chemistry 85 (2019) 168–178
Y. Gao et al.
with antibacterial and anti-tubercular activities. 3.2. A phenotypic screen verified M. smegmatis growth inhibition of compound 50 Then, the in vitro antibacterial activities of compounds 1–63 were conducted through the Kirby–Bauer disc agar diffusion assay. INH was picked as a reference drug. Since M. tuberculosis is infectious, the amenable substitute M. smegmatis was used [15,16]. Among the 63 compounds, 12 of them displayed notable activity, with inhibition zone greater than 14 mm (Tables 1, S1 and Fig. 2). 3.3. Microscale thermophoresis (MST) confirmed binding affinity of compounds to DprE1 To investigate the results of molecular docking, microscale thermophoresis (MST) assay was subsequently employed to validate the binding interactions between the 12 effective compounds with DprE1. MST, a precious method, can capture the fluorescence changes of proteins during thermal immortalization, so as to realize the study of protein-protein or protein-small molecule interaction [17]. The dissociation constant (Kd) measured by MST reflects the affinity between protein and ligand. Among all of the selected candidates, compound 50 demonstrated the highest affinity to DprE1 with Kd of 25 ± 2 nM (Fig. 1). 3.4. Molecular docking elucidated the binding mode of compound 50 with DprE1 In order to further elucidate the binding mode of compound 50 with DprE1, molecular docking was carried out. The lowest-energy binding conformation of compound 50 was shown as Fig. 3. Docking results demonstrated that compounds bind to the active site of the enzyme. The ligand binding pocket of DprE1 was zig-zag shaped, where many hydrophobic amino acid, including His132, Ile113, Tyr314, Phe320, TRP323, Val365, and Phe369, form a relatively hydrophobic envelop (Fig. 3A). Compound 50 formed obvious hydrophobic interactions with many of these amino acids. The pyrimidinetrione ring of compound 50 also formed strong hydrophobic interaction with flavine adenine dinucleotide (FAD) (Fig. 3A). Hydrogen bonds were predicted between two carbonyl oxygens of compound 50 with Lys418, and pyridine nitrogen with Asn324 (Fig. 3B).
Fig. 5. Anti-TB efficacy in agents-treated C57BL/6 acute infection model. Mice (n = 10) were i.v. challenged with 1.2 × 106 CFU of virulent M. tuberculosis H37Rv. After treatment, weight was recorded (A); lung and spleen tissue sections were fixed and embedded for HE staining and acid-fast (B). Arrows indicate AF staining positive bacteria. Pathological changes were scored.
3.5. A phenotypic screen verified M. tuberculosis growth inhibition of compound 50 For the compounds that have both activity against M. smegmatis and high affinity towards DprE1, a test to determine minimal inhibitory
Scheme 1. The synthesis of compound 50. A. Rational design and synthetic strategy for the DprE1 inhibitor. Reagents and conditions: (i) KOCN, in H2O, 55 °C for 4 h. (ii) Dimethyl malonate, NaOMe in MeOH, reflux for 6 h. (iii) Trimethyl orthoformate, TsOH. (iv) Isonicotinic acid hydrazide, in MeOH, rt for 6 h. B. The three tautomers of enamine 5 (compound 50) in solution. 174
Bioorganic Chemistry 85 (2019) 168–178
Y. Gao et al.
Fig. 6. The bacterial load of organs was enumerated in agents-treated C57BL/6 acute infection model. After treatment, lungs (A) and spleen (B) were harvested and the bacterial load was enumerated. Results are shown as mean ( ± SEM) log10 CFU/lung, *p < 0.01, **0.001 < p < 0.01, ***p < 0.001, PBS vs. other groups.
Fig. 7. Cytotoxicity testing of compound 50. The cytotoxicity of compound 50 was measured on Vero, L929 and Lo2 (n = 3).
concentration against M. tuberculosis was performed. Compound 50 presented a comparable activity to that of INH (1 μg/mL, 7.29 μM), with an MIC of 3.1 μg/mL (9.75 μM) (Fig. 4).
were treated with compound 50 (10 mg/kg, 30 mg/kg), or INH (10 mg/ kg). After treatment, bacterial load of the lung and spleen along with those histopathological changes were calculated (Fig. 6). Compared with PBS control group, the CFU dropped 1.0 log in lung and 0.6 logs in spleen for the group of compound 50 (10 mg/kg) similar to the potency of INH group (Fig. 6). And the CFU dropped 2.1 and 1.3 logs in lung and spleen, respectively, for the group of the compound 50 (30 mg/kg), better than the INH group (10 mg/kg) (Fig. 6). As the figure showed, the most severe granuloma-like histopathological changes were clearly visible improvement in lungs, especially for the group of compound 50 (30 mg/kg), little M. tuberculosis bacterium could be found in lungs (Fig. 5). Thus, compound 50 indicated a significant efficacy in vivo on reducing the bacterial burden in lungs in a dose-dependent tendency compared to PBS controls.
3.6. The synthesis of compound 50 In order to carry out subsequent studies, we synthesized compound 50. As shown in Scheme 1, allylurea (2) was prepared from potassium cyanate and allyl amine hydrochloride according to the literature procedure [23–25], then was cyclized with dimethyl malonate in the presence of sodium methoxide. Refluxing the obtained pyrimidinetrione (3) in trimethyl orthoformate with a catalytic amount of PTSA for several hours afforded the formylpyrimidinetrione (4) in good yield [26]. The condensation between 4 and isonicotinyl hydrazide followed by a 1,3-proton transfer process in one-pot under room temperature gave the target compound 50. Similar to tetramic acid derivatives [27–29], the resulting enamine 5 (compound 50) can exist as endo and exo-enol and keto tautomers in solution, which was a mixture of the E- and Z-isomers [30–32]. Since the two configurations (E/Z) of this enamine-typed compound could interconvert into each other rapidly in polar solvent through an imine intermediate, we speculate that the geometry of the double bound would not affect its interaction with DprE1, or the different interactions of two isomers with the enzyme could not be measured.
3.8. Compound 50 showed no obvious toxicity in vitro and in vivo To check out the safety of compound 50, we test its cytotoxicity on Vero, Lo2 and L929 cells in vitro. As shown in Fig. 7, IC50 values for all cell lines were greater than 500 μM, which were more than 100 times higher compared with the effective concentration. These results demonstrated no obvious cytotoxicity of compound 50, suggesting that it would have a wide therapeutic window in vivo. In order to verify the conjecture above, an acute toxicity assay in mice was conducted [22]. During the assay, no decline on the body weight was observed (Fig. 8A). The weight of mice and the formaldehyde fixation of lung, spleen, heart, liver, kidney tissues were no significantly different compared to control groups, indicating that no any toxicity of compound 50 orally gavage in a dose of 2000 mg/kg (Fig. 8A and B). During 14 days of observation, no mouse died. The LD50 of compound 50 was higher than 2000 mg/kg. These results demonstrated compound 50 would be safe if it is used for treatment.
3.7. Compound 50 is efficacious in acute M. tuberculosis infection mouse model To identify the efficacy of compound 50, we carried out an in vivo experiment in an acute infection model. Specific pathogen-free, female C57BL/6 mice were infected by nasal drops with the dose of 1 × 106 CFU of M. tuberculosis H37Rv. After infection for 14 days, mice 175
Bioorganic Chemistry 85 (2019) 168–178
Y. Gao et al.
Fig. 8. Acute toxicity in vivo. A. body weight was recorded; B. Spleen, lung, heart, liver and kidney tissue sections were fixed and embedded for HE staining. Groups a, b represent male gavage with saline and compound 50, respectively; groups c, d represent female gavage with saline and compound 50, respectively.
4. Discussion and conclusions
Comprising a similar scaffold, compound 49 and 50 showed different anti-TB effect. Indeed, it may clarify the mode of compound 50. As the result of molecular docking, Fig. 3 showed that compound 50 deeply inserted into the active site of DprE1, forming several obvious hydrogen bonding interactions. Among them, the hydrogen bond between pyridine nitrogen with Asn324 is worthy of attention, which may cause compound 50 to have better anti-TB effects than compound 49. MST confirmed better binding potency of compound 50. Following that, inhibition against M. smegmatis and M. tuberculosis verified compound 49 was really weaker than compound 50. Therefore, compound 50 was preferentially picked up, rather than compound 49. Compound 50 with the isoniazid fragment is a potent DprE1 inhibitor, but INH is not. So compound 50 and INH must kill M. tuberculosis in different mode. Both of them achieved similar antimicrobial effects on M. smegmatis and M. tuberculosis in vitro and in vivo. Since compound 50 is stable in vitro, it was confirmed that compound 50 could behave an
The economic and social damage caused by M. tuberculosis and the toxicity of anti-tuberculosis drugs in clinic are driving the emergence of new drugs with the anti-tuberculosis effects and the safe treatment profile. Here we have applied a structure-based virtual ligand screening approach to explore target-based anti-tuberculosis agents among 6.2 million compounds. This method is proved to be quick and efficient to search for lead-like compounds. And the pyrimidinetrione, compound 50 was identified as a novel and potent DprE1 inhibitor with high affinity at nanomolar scale through MST assay. Moreover, compound 50 showed commendable anti-tuberculosis efficiency in vitro and in vivo, which was comparable to the first-line anti-tuberculosis drugs. What need to be emphasized is that compound 50 did not show any obvious toxicity in cytotoxicity testing and acute toxicity assessment. It exactly predicted a favorable treatment window of compound 50. 176
Bioorganic Chemistry 85 (2019) 168–178
Y. Gao et al.
inhibitory effect against M. tuberculosis in vitro. This exclude the possibility that compound 50 has to metabolize to INH to take effects. We concluded that compound 50 come into effects in a different way from INH. Taken together, compound 50 could be a novel type of anti-tuberculosis agent or lead compound via inhibiting DprE1. Of course, it is worthy further to investigate whether compound 50 can be converted to INH to generate additional anti-tubercular effects in vivo. In addition, our research process of structure-based drug discovery will also provide a vital method for the drug discovery in major diseases. Furthermore, other important species known for human pathogenesis, such as M. leprae, the causative agent of leprosy persisting in developing countries, and M. avium intracellular complex (MAC), an important group including nontuberculous mycobacteria, are responsible for opportunistic infections in immunocompromised individuals [33,34]. Since the two species both depend on DprE1 to synthesize their cell walls, these diseases could be also solved by compound 50, which is worthy to be further validated. Additionally, these studies emphasize the great potential of inhibiting critical mycobacteria enzymes and reveal a potential lead compound worthy further investigation.
11354–11359, https://doi.org/10.1073/pnas.1205735109. [8] R.V. Chikhale, M.A. Barmade, P.R. Murumkar, M.R. Yadav, Overview of the development of DprE1 inhibitors for combating the menace of tuberculosis, J. Med. Chem. 61 (2018) 8563–8593, https://doi.org/10.1021/acs.jmedchem.8b00281. [9] R. Sommer, J. Neres, J. Piton, N. Dhar, A. van der Sar, R. Mukherjee, T. Laroche, P.J. Dyson, J.D. McKinney, W. Bitter, V. Makarov, S.T. Cole, Fluorescent benzothiazinone analogues efficiently and selectively label dpre1 in mycobacteria and actinobacteria, ACS Chem. Biol. 13 (2018) 3184–3192, https://doi.org/10.1021/ acschembio.8b00790. [10] J.J. Irwin, T. Sterling, M.M. Mysinger, E.S. Bolstad, R.G. Coleman, ZINC: a free tool to discover chemistry for biology, J. Chem. Inf. Model. 52 (2012) 1757–1768, https://doi.org/10.1021/ci3001277. [11] R. Abagyan, M. Totrov, D. Kuznetsov, ICM? A new method for protein modeling and design: applications to docking and structure prediction from the distorted native conformation, J. Comput. Chem. 15 (1994) 488–506, https://doi.org/10.1002/jcc. 540150503. [12] E.-L. Lyle, G.L. Hayward, M. Thompson, Acoustic coupling of transverse waves as a mechanism for the label-free detection of protein–small molecule interactions, The Analyst 127 (2002) 1596–1600, https://doi.org/10.1039/B209051C. [13] D.-L. Ma, D.S.-H. Chan, C.-H. Leung, Molecular docking for virtual screening of natural product databases, Chem. Sci. 2 (2011) 1656–1665, https://doi.org/10. 1039/C1SC00152C. [14] L. Borriello, M. Montès, Y. Lepelletier, B. Leforban, W.-Q. Liu, L. Demange, B. Delhomme, S. Pavoni, R. Jarray, J.L. Boucher, S. Dufour, O. Hermine, C. Garbay, R. Hadj-Slimane, F. Raynaud, Structure-based discovery of a small non-peptidic Neuropilins antagonist exerting in vitro and in vivo anti-tumor activity on breast cancer model, Cancer Lett. 349 (2014) 120–127, https://doi.org/10.1016/j.canlet. 2014.04. 004. [15] V. Chaturvedi, N. Dwivedi, R.P. Tripathi, S. Sinha, Evaluation of Mycobacterium smegmatis as a possible surrogate screen for selecting molecules active against multi-drug resistant Mycobacterium tuberculosis, J. Gen. Appl. Microbiol. 53 (2007) 333–337, https://doi.org/10.2323/jgam.53.333. [16] G. Etienne, The cell envelope structure and properties of Mycobacterium smegmatis mc2155: is there a clue for the unique transformability of the strain? Microbiology 151 (2005) 2075–2086, https://doi.org/10.1099/mic.0.27869-0. [17] Y. Mao, L. Yu, R. Yang, L. Qu, P.deB. Harrington, A novel method for the study of molecular interaction by using microscale thermophoresis, Talanta 132 (2015) 894–901, https://doi.org/10.1016/j.talanta.2014.09.038. [18] P. Escalante, P. Escalante, S. Ramaswamy, H. Sanabria, H. Soini, X. Pan, O. Valiente-Castillo, J.M. Musser, Genotypic characterization of drug-resistant Mycobacterium tuberculosis isolates from Peru, Tuber. Lung Dis. 79 (1998) 111–118, https://doi.org/10.1054/tuld.1998.0013. [19] F. Wang, D. Sambandan, R. Halder, J. Wang, S.M. Batt, B. Weinrick, I. Ahmad, P. Yang, Y. Zhang, J. Kim, M. Hassani, S. Huszar, C. Trefzer, Z. Ma, T. Kaneko, K.E. Mdluli, S. Franzblau, A.K. Chatterjee, K. Johnsson, K. Mikusova, G.S. Besra, K. Futterer, S.H. Robbins, S.W. Barnes, J.R. Walker, W.R. Jacobs, P.G. Schultz, Identification of a small molecule with activity against drug-resistant and persistent tuberculosis, Proc. Natl. Acad. Sci. 110 (2013) E2510–E2517, https://doi.org/10. 1073/pnas.1309171110. [20] J. Liang, X. Teng, X. Yuan, Y. Zhang, C. Shi, T. Yue, L. Zhou, J. Li, X. Fan, Enhanced and durable protective immune responses induced by a cocktail of recombinant BCG strains expressing antigens of multistage of Mycobacterium tuberculosis, Mol. Immunol. 66 (2015) 392–401, https://doi.org/10.1016/j.molimm.2015.04.017. [21] P.S. Shirude, R. Shandil, C. Sadler, M. Naik, V. Hosagrahara, S. Hameed, V. Shinde, C. Bathula, V. Humnabadkar, N. Kumar, J. Reddy, V. Panduga, S. Sharma, A. Ambady, N. Hegde, J. Whiteaker, R.E. McLaughlin, H. Gardner, P. Madhavapeddi, V. Ramachandran, P. Kaur, A. Narayan, S. Guptha, D. Awasthy, C. Narayan, J. Mahadevaswamy, K. Vishwas, V. Ahuja, A. Srivastava, K. Prabhakar, S. Bharath, R. Kale, M. Ramaiah, N.R. Choudhury, V.K. Sambandamurthy, S. Solapure, P.S. Iyer, S. Narayanan, M. Chatterji, Azaindoles: noncovalent DprE1 inhibitors from scaffold morphing efforts, kill Mycobacterium tuberculosis and are efficacious in vivo, J. Med. Chem. 56 (2013) 9701–9708, https://doi.org/10.1021/ jm401382v. [22] W. Li, M. Zheng, S. Wu, S. Gao, M. Yang, Z. Li, Q. Min, W. Sun, L. Chen, G. Xiang, H. Li, Benserazide, a dopadecarboxylase inhibitor, suppresses tumor growth by targeting hexokinase 2, J. Exp. Clin. Cancer Res. 36 (2017) 58, https://doi.org/10. 1186/s13046-017-0530-4. [23] R.C. Moschel, W.R. Hudgins, A. Dipple, Reactivity effects on site selectivity in nucleoside aralkylation: a model for the factors influencing the sites of carcinogennucleic acid interactions, J. Org. Chem. 51 (1986) 4180–4185, https://doi.org/10. 1021/jo00372a015. [24] S. Fouladdel, A. Khalaj, N. Adibpour, E. Azizi, Synthesis and cytotoxicity of some biurets against human breast cancer T47D cell line, Bioorg. Med. Chem. Lett. 20 (2010) 5772–5775, https://doi.org/10.1016/j.bmcl.2010.07.137. [25] X. Min, J. Liu, Y. Dong, M. Hussain, One-pot assembly towards ω-substituted arylbiurets from aromatic amines, potassium cyanate, and glacial acetic acid, Synthesis 50 (2018) 341–348, https://doi.org/10.1055/s-0036-1590934. [26] D.M. Neumann, A. Cammarata, G. Backes, G.E. Palmer, B.S. Jursic, Synthesis and antifungal activity of substituted 2,4,6-pyrimidinetrione carbaldehyde hydrazones, Bioorg. Med. Chem. 22 (2014) 813–826, https://doi.org/10.1016/j.bmc.2013.12. 010. [27] Y.-C. Jeong, M.G. Moloney, Synthesis of and tautomerism in 3-acyltetramic acids, J. Org. Chem. 76 (2011) 1342–1354, https://doi.org/10.1021/jo102304y. [28] Y.-C. Jeong, Z. Bikadi, E. Hazai, M.G. Moloney, A detailed study of antibacterial 3acyltetramic acids and 3-acylpiperidine-2,4-diones, ChemMedChem (2014), https://doi.org/10.1002/cmdc.201402093 n/a-n/a.
Funding This work was supported by the National Natural Science Foundation of China (NSFC) [grant number 81473254, 81773637, U1703111, and 81773594] and National Major Scientific and Technological Special Project for “Significant New Drugs Development” during the 13th Five-Year Plan Period [grant number 2019ZX09721001-004-007]. Acknowledgement We acknowledge support from Schrödinger Release 2018-4: Maestro, Schrödinger, LLC, New York, NY, 2018. Competing interests The authors have declared that no conflict of interest exists. Appendix A. Supplementary material Supplementary data to this article can be found online at https:// doi.org/10.1016/j.bioorg.2018.12.018. References [1] M. Chatterji, R. Shandil, M.R. Manjunatha, S. Solapure, V. Ramachandran, N. Kumar, R. Saralaya, V. Panduga, J. Reddy, P. Kr, S. Sharma, C. Sadler, C.B. Cooper, K. Mdluli, P.S. Iyer, S. Narayanan, P.S. Shirude, 1,4-Azaindole, a potential drug candidate for treatment of tuberculosis, Antimicrob. Agents Chemother. 58 (2014) 5325–5331, https://doi.org/10.1128/AAC.03233-14. [2] A. Nader, T.A. deRoon-Cassini, T.B. Meier, J.B. Hoelzle, L.D. Nelson, Relationship between somatic symptoms, peripheral inflammation, and clinical outcomes after traumatic injury, Brain Behav. Immun. 66 (2017) e26, , https://doi.org/10.1016/j. bbi.2017.07.099. [3] Organisation mondiale de la santé, Global Tuberculose Report: 2015, Word Health Organization, Genève (Suisse), 2015. [4] S. Crunkhorn, Antibacterial agents: new routes to tuberculosis treatment, Nat. Rev. Drug Discov. 16 (2017) 600–601, https://doi.org/10.1038/nrd.2017.156. [5] H. Li, G. Jogl, Crystal structure of decaprenylphosphoryl-β-D-ribose 2′-epimerase from Mycobacterium smegmatis, Prot.: Struct. Funct. Bioinform. 81 (2013) 538–543, https://doi.org/10.1002/prot.24220. [6] J. Neres, F. Pojer, E. Molteni, L.R. Chiarelli, N. Dhar, S. Boy-Rottger, S. Buroni, E. Fullam, G. Degiacomi, A.P. Lucarelli, R.J. Read, G. Zanoni, D.E. Edmondson, E. De Rossi, M.R. Pasca, J.D. McKinney, P.J. Dyson, G. Riccardi, A. Mattevi, S.T. Cole, C. Binda, Structural basis for benzothiazinone-mediated killing of Mycobacterium tuberculosis, Sci. Transl. Med. 4 (2012), https://doi.org/10.1126/ scitranslmed.3004395 150ra121-150ra121. [7] S.M. Batt, T. Jabeen, V. Bhowruth, L. Quill, P.A. Lund, L. Eggeling, L.J. Alderwick, K. Futterer, G.S. Besra, Structural basis of inhibition of Mycobacterium tuberculosis DprE1 by benzothiazinone inhibitors, Proc. Natl. Acad. Sci. 109 (2012)
177
Bioorganic Chemistry 85 (2019) 168–178
Y. Gao et al. [29] Y.-C. Jeong, M.G. Moloney, Synthesis and antibacterial activity of monocyclic 3carboxamide tetramic acids, Beilstein J. Org. Chem. 9 (2013) 1899–1906, https:// doi.org/10.3762/bjoc.9.224. [30] Y.-C. Jeong, M. Moloney, Antibacterial barbituric acid analogues inspired from natural 3-acyltetramic acids; Synthesis, tautomerism and structure and physicochemical property-antibacterial activity relationships, Molecules 20 (2015) 3582–3627, https://doi.org/10.3390/molecules20033582. [31] I.V. Ukrainets, L. Yangyang, N.L. Bereznyakova, A.V. Turov, 4-hydroxy-2-quinolones. 169*. synthesis and bromination of 1-allyl-3-(arylamino-methylene)quinoline-2,4-(1h,3h)-diones, Chem. Heterocycl. Compd. 45 (2009) 1235–1240, https:// doi.org/10.1007/s10593-010-0412-6.
[32] S. Han, F.-F. Zhang, H.-Y. Qian, L.-L. Chen, J.-B. Pu, X. Xie, J.-Z. Chen, Development of quinoline-2,4(1 H,3 H)-diones as potent and selective ligands of the cannabinoid type 2 receptor, J. Med. Chem. 58 (2015) 5751–5769, https://doi.org/10.1021/acs. jmedchem.5b00227. [33] J. Gawad, C. Bonde, Current affairs, future perspectives of tuberculosis and antitubercular agents, Indian J. Tuberc. 65 (2018) 15–22, https://doi.org/10.1016/j. ijtb.2017.08.011. [34] P.K. Crellin, R. Brammananth, R.L. Coppel, Decaprenylphosphoryl-β-D-Ribose 2′Epimerase, the target of benzothiazinones and dinitrobenzamides, is an essential enzyme in Mycobacterium smegmatis, PLoS One 6 (2011) e16869, , https://doi.org/ 10.1371/journal.pone.0016869.
178
Contents lists available at ScienceDirect
Bioorganic Chemistry journal homepage: www.elsevier.com/locate/bioorg
Identification of a pyrimidinetrione derivative as the potent DprE1 inhibitor by structure-based virtual ligand screening Ya Gaoa, Jinshan Xiea, Ruotian Tanga, Kaiyin Yanga, Yahan Zhanga, Lixia Chenb, , Hua Lia,b, ⁎
T
⁎
a
Hubei Key Laboratory of Natural Medicinal Chemistry and Resource Evaluation, School of Pharmacy, Tongji Medical College, Huazhong University of Science and Technology, Wuhan 430030, China b Wuya College of Innovation, School of Traditional Chinese Materia Medica, Key Laboratory of Structure-Based Drug Design & Discovery, Ministry of Education, Shenyang Pharmaceutical University, Shenyang 110016, China
ARTICLE INFO
ABSTRACT
Keywords: Pyrimidinetrione DprE1 Zinc database Virtual ligand screening Tuberculosis
Despite the increasing need of new antituberculosis drugs, the number of agents approved for the market has fallen to an all-time low. In response to the emerging drug resistance followed, structurally unique chemical entities will be highlighted. decaprenylphosphoryl-β-D-ribose oxidase (DprE1) participating in the biosynthesis of mycobacterium cell wall is a highly vulnerable and validated antituberculosis target. On the basis of it, a systematic strategy was applied to identify a high-quality lead compound (compound 50) that inhibits the essential enzyme DprE1, thus blocking the synthesis of the mycobacterial cell wall to kill M. tuberculosis in vitro and in vivo. Correspondingly, the rational design and synthetic strategy for compound 50 was reported. Notably, the compound 50 has been confirmed to be no toxicity. Altogether, our data suggest the compound 50 targeting DprE1 is a promising candidate for the tuberculosis (TB) therapy.
1. Introduction It was claimed that M. tuberculosis, an intracellular pathogenic bacteria, emerged since the times known to us [1]. The terrible thing is that the transmission, evolution, and existence of the bacteria till now have shown enormously success, which plagues humanity all the time [1,2]. According to the World Health Organization (WHO) reports [3], tuberculosis (TB) is one of the leading cause of death in infectious disease in the world. Treatment of TB has been always depending on the drugs used for a long time, such as isoniazid, rifampicin, ethambutol, pyrazinamide, and so on. While, the multidrug resistant (MDR-TB) and extensively drug resistant (XDR-TB) render these drugs less effective. Unfortunately, no new and effective anti-TB drugs have been developed in the past 40 years. The research on drugs against drug-resistant strains of M. tuberculosis with special structures and patterns of action remains a significant challenge [4]. DprE1 is reported to be essential for survival of M. tuberculosis including multi-drug resistant strains [5]. The enzyme is crucial and conservative for mycobacterial cell wall biosynthesis. DprE1 catalyzes the FAD-dependent oxidation of decaprenylphosphoryl-β-D-
ribose (DPR) to decaprenyl-phosphoryl-2′-keto-D-erythro- pentofuranose (DPX). Then DPX is then reduced by decaprenyl-phospho-2′-ketoD- arabinose reductase (DprE2) to generate DPA, which is the indispensable component of mycobacterium cell wall [5–7]. Since there is no alternative way to synthesis DPX, DprE1 becomes a druggable target. Correspondingly, DprE1 inhibitors will be low toxic to human cell because there are no cell walls for human and thus has no similar metabolic pathway. Targeting DprE1 with small-molecule inhibitors has emerged as a promising strategy for anti-TB therapy. For the past few years, DprE1 inhibitors, like BTZs and TCA1, were found to cause mycobacteria death by covalently or non-covalently binding to the enzyme DprE1. These compounds have showed promising anti-tubercular efficacy, and provided powerful evidence for designing DprE1 inhibitors and treating TB [8,9]. In this study, 63 compounds were cherry-picked from more than 6.2 million small molecules through virtual ligand screening. Compound 50 was further confirmed to be a significantly effective bactericidal agent compared with isoniazid (INH) through inhibition test using M. smegmatis and M. tuberculosis in vitro. From these combined screening efforts, we identified a novel inhibitor that targets DprE1. Furthermore, in the
Abbreviations: DprE1, decaprenylphosphoryl-β-D-ribose oxidase; MDR, multidrug resistant; XDR, extensively drug resistant; MST, microscale thermophoresis; TB, tuberculosis; DPR, decaprenylphosphoryl-β-D-ribose; DPX, decaprenyl-phosphoryl-2′-keto-D-erythro- pentofuranose; INH, isoniazid; DprE2, Decaprenyl-phospho-2′keto-D-arabinose reductase; mp, melting point; MST, microscale thermophoresis ⁎ Corresponding authors. E-mail addresses: [email protected] (L. Chen), [email protected] (H. Li). https://doi.org/10.1016/j.bioorg.2018.12.018 Received 19 October 2018; Received in revised form 30 November 2018; Accepted 12 December 2018 Available online 13 December 2018 0045-2068/ © 2018 Elsevier Inc. All rights reserved.
Bioorganic Chemistry 85 (2019) 168–178
Y. Gao et al.
MST assay, compound 50 was verified to specifically bind to DprE1 with high affinity. It is gratifying that compound 50 showed comparable therapeutic effect to INH for the treatment of acute infections in mice. It was noteworthy that this compound showed no any obvious toxicity in cytotoxicity testing and acute toxicity assessment. Our study yielded new antituberculosis lead compounds, and the structure-based drug discovery strategy will also provide an effective method for other major diseases.
Table 1 Binding affinity and inhibitory activities of compounds in vitro.
2. Materials and methods 2.1. Compounds database a. Subset of ZINC database, ZINC lead like compound library containing 6,053,287 molecules [8]. b. ZINC natural compounds library, a total of approximately 150,000 compounds, including IBS screen, Princeton NP, Taiwan and AnalytiCon Discovery NP database, is the relatively large database that has undergone ZINC screening of drugs and remove duplicate or similar structure [10].
No.a
Scoreb (Kal·cal−1)
Kdc (nM)
Inhibition zoned (mm)
1 10 16 19 28 33 34 38 49 50 54 62 INH
−44.93 −43.87 −38.21 −38.61 −37.7 −37.19 −37.86 −36.9 −37.13 −37.8 −40.58 −37.62
> 500000 > 500000 65600 ± 6140 891 ± 109 > 500000 > 500000 154000 ± 13500 > 500000 14500 ± 1020 25 ± 2 > 500000 11400 ± 663 > 500000
20 16 14 24 20 16 14 14 14 20 14 16 20
a Among the 63 compounds from virtual screening, 12 of them displayed notable activity against M. smegmatis, with inhibition zone greater than 14 mm. b Docking score/interaction potential of compounds with DprE1 (kcal/mol). c The Kd value was automatic calculated by the curve fitting. d Inhibition zone was conducted through the Kirby–Bauer disc agar diffusion assay.
2.2. Structure-based virtual ligand screening
DMSO to 25 mM. INH was also diluted in DMSO to 500 mg/L as positive control. And the solvent DMSO was used as negative control. Sterile discs which were impregnated with 10 μL of DMSO or compounds were put onto 7H10 (BD) plate. The plates were incubated for 30 h at 37 °C. Zones of inhibitor were measured to estimate the antibacterial effect [15,16].
ICM 3.8.2 modeling software on an Intel i7 4960 processor (MolSoft LLC, San Diego, CA) was used to identify possible drug candidates for DprE1 enzyme. For the structure-based virtual screening, ligand was continuously resiliently made to dock with DprE1 (PDB ID: 4FDO) [7] that was represented in potential energy maps and obtained online (http://www.rcsb.org/pdb). 3D compounds obtained from ZINC were scored according to the internal coordinate mechanics (Internal Coordinate Mechanics, ICM) [11–13]. Based on Monte Carlo simulations, stochastic global optimization procedure and pseudo-Brownian positional/torsional steps, the position of intrinsic molecular was optimized [14]. By visually inspecting, compounds outside the active site, as well as those weakly fitting to the active site were eliminated. According with Lipinski (Rule of 5) and scoring more than −32 (generally represents strong interactions) was priority to be selected.
2.5. In vitro binding affinity Recombinant DprE1 was labeled with the Monolith NT™ Protein Labeling Kit RED (Cat#L001) according to the supplied protocol. Labeled DprE1 was kept constant at 100 nM. All samples tested were diluted in a 20 mM HEPES (pH 7.5) and 0.05 (v/v) % Tween-20. Compounds were diluted covering the range from 500 μM to 2 nM. After 10 min incubation at room temperature, samples were loaded into Monolith™ standard-treated capillaries and the thermophoresis was measured at 25 °C after 30 min incubation on a Monolith NT.115 instrument (NanoTemper Technologies, München, Germany) [17]. The LED power was set to 100%. The dissociation constant Kd values were fitted by using the NT Analysis software (NanoTemper Technologies, München, Germany).
2.3. Protein expression and purification The gene of full-length decaprenylphosphoryl-β-D-ribose oxidase was cloned from M. smegmatis strain ATCC 607 (China General Microbiological Culture Collection Center). The gene was ligated into PET28a vector (Novagen). After the recombinant plasmid was verified by sequencing, it was transformed into E. coli strain BL21 (Invitrogen) at 293 K which was grown in LB medium at 37 °C to an OD600 (0.8–1.0) and induced by 0.4 mM isopropyl-D-thiogalactopyranoside (IPTG) at 18 °C for 16 h. Bacterial cells were lysed on ice in buffer containing 100 mM Tris-HCl pH 8.8, 200 mM NaCl, 10% glycerol, 1% TritonX-100, 5 mM β-mercaptoethanol. Soluble N-terminally hexa-histidine tagged DprE1 was bound to Ni-agrose affinity resin (Qiagen), washed with a buffer containing 20 mM Tris-HCl pH 8.8, 200 mM NaCl and 10 mM imidazole and eluted with a buffer containing 20 mM Tris-HCl pH 8.8, 250 mM NaCl, and 150 mM imidazole. The eluted protein was concentrated and diluted with a buffer containing 20 mM Tris-HCl pH 8.8, 250 mM NaCl. The protein was further purified with size exclusion chromatography (GE Health) at 20 mM Tris-HCl pH 8.8 and 200 mM NaCl [5].
2.6. In vitro antibacterial activity test using M. tuberculosis In vitro antibacterial activity test using M. tuberculosis was carried in accordance with MABA. 100 µL sterile normal saline was added to outer-perimeter wells of clean 96-well plates (Falcon; BD), to avoid the error by evaporation of the liquid during the test. Every three horizontal tested one compound which was diluted by 7H9 broth in 9 dilution steps. All compounds were done in identical serial 1:1 dilutions. And the wells in column 11 were served as drug-free controls. After adding 100 μL drug liquid, 100 μL 7H9 broth seeded M. tuberculosis was put into every well tested. The microplates were resealed with Parafilm and were incubated for an additional 24 h at 37 °C. Fifty microliters of a freshly prepared 1:1 mixture of Alamar Blue (Accumed International, Westlake, Ohio) reagent and 10% Tween 80 were added to drug-free wells. If drug-free wells turned pink, the reagent mixture was added to all wells in the microplates. The microplates were resealed with Parafilm and were incubated for an additional 24 h at 37 °C, and the colors of all wells were recorded. A blue color in the well was interpreted as no growth, and a pink color was scored as growth. A few wells appeared violet after 24 h of incubation, but they invariably changed into pink after another day of incubation and thus were scored as growth (while the adjacent blue wells remained blue) [18]. The MIC
2.4. In vitro antibacterial activity test using M. smegmatis Because of the highly infectious of M. tuberculosis, the homology and safety M. smegmatis ATCC 607 (China General Microbiological Culture Collection Center) was identified as a good model to screen M. tuberculosis inhibitors. It was selected to test the effects of all compounds using a Kirby-Bauer disc diffusion assay. All compounds were diluted in 169
Bioorganic Chemistry 85 (2019) 168–178
Y. Gao et al.
Fig. 1. MST assay to measure the binding between compounds and DprE1. Compound 50 displayed the equilibrium dissociation constant (Kd) of 25 ± 2 nM. Kd values were automatic calculated by the curve fitting.
was defined as the lowest drug concentration which prevented a color change from blue to pink. And the values will be recorded on Microplate Reader at 520/590 nm.
2.7.2. Synthetic procedure for allylurea 2 A 100 mL three-necked bottle was charged with allylamine hydrochloride (0.1 mol), KOCN (0.11 mol, 1.1 equiv), H2O (50 mL), and stirred at 55 °C for 3 h, the progress of the reaction was monitored by TLC. Upon completion of the reaction, the solvent was evaporated under vacuum. The residue was dispersed with 100 mL ethanol then filtered, washed with ethanol. The filtrate was evaporated under vacuum to afford the target allylurea 2 as white solid. Yield 76.8%. 1H NMR (400 MHz, DMSO‑d6) δ 6.05–5.97 (m, 1H), 5.81–5.72 (m, 1H), 5.44 (s, 2H), 5.10–5.04 (m, 1H), 5.00–4.96 (m, 1H), 3.57–3.55 (m, 2H).
2.7. Synthesis of compound 50 2.7.1. General experimental methods All reactions were carried out in flame-dried glassware under an atmosphere of argon, all reagents were procured from commercial sources (purity > 99%) and used without further purification. Silica gel GF254 and silica gel (200–300 mesh) were respectively used for thin-layer chromatography and column chromatography. NMR spectra were recorded on a Bruker AM-400 spectrometer, with 1 H and 13C NMR chemical shifts referenced to the solvent or solvent impurity peaks for DMSO‑d6 (δH 2.50 and δC 39.52). HRESIMS data were acquired using a Thermo Fisher LC-LTQ-Orbitrap XL spectrometer and an electrospray ionization and a hybrid quadrupole time-of-flight (q-TOF) mass spectrometer (model 6540, Agilent).
2.7.3. Synthetic procedure for pyrimidinetrione 3 To a 100 mL three-necked bottle was added allylurea 2 (0.1 mol), dimethyl malonate (0.1 mol, 1.0 equiv), NaOMe (0.12 mol, 1.2 equiv) and MeOH (100 mL), the mixture was refluxed for 6 h until the completion of the reaction monitored by TLC. The solvent was evaporated under vacuum and H2O (50 mL) was added, the mixture was cooled in ice water and the solid filtered, washed with water to give pure 170
Bioorganic Chemistry 85 (2019) 168–178
Y. Gao et al.
Fig. 2. Chemical structures of the selected compounds.
pyrimidinetrione 3 as white solid. Yield 71.5%. 1H NMR (400 MHz, DMSO‑d6) δ 11.34 (s, 1H), 5.79–5.69 (m, 1H), 5.16–5.12 (m, 1H), 5.07–5.04 (m, 1H), 4.25–4.23 (m, 2H).
2.8. Cytotoxicity testing Vero Cells and L929 were grown in Roswell Park Memorial Institute medium (RPMI), supplemented with penicillin G (100 UI/mL), streptomycin (100 mg/mL) and 10% fetal calf serum at 37 °C in 5% CO2 atmosphere. Lo2 were grown in Dulbecco’s modified eagle medium (DMEM), supplemented with penicillin G (100 UI/mL), streptomycin (100 mg/mL) and 10% fetal calf serum at 37 °C in 5% CO2 atmosphere. For cytotoxicity evaluation of compounds, the Vero, L929 and Lo2 cells were plated at a density of 1 × 104 cells/well in a 96 well plate at 37 °C in 5% CO2 atmosphere. Cells were then cultivated for 24 h, medium in the wells was replaced with fresh medium containing compound 50 which was diluted in 8 dilution steps covering the range from 500 μM to 3.9 μM. And DMSO was used as negative control. After 48 h incubation, 10 μL of CCK8 solution was added to each well. After 30 min, the absorbance of each well was measured at wavelength of 450 nm on spectrophotometer. Data was corrected for background (nocell control) and expressed as a percentage of the value for untreated cells. IC50 was calculated by the test and cytotoxicity of the compound to normal cells can be evaluated.
2.7.4. Synthetic procedure for formylpyrimidinetrione 4 To a 100 mL three-necked bottle was added pyrimidinetrione 3 (10 mmol), trimethylorthoformate (0.1 mol, 10 equiv), PTSA·H2O (1.0 mmol, 0.1 equiv), the mixture was stirred for 6 h at 80 °C until the completion of the reaction monitored by TLC. The solvent was evaporated under vacuum and the residue was recrystallized from ethanol to give formylpyrimidinetrione 4 as yellow solid. Yield 50.1%. 1H NMR (400 MHz, DMSO‑d6) δ 10.47–10.46 (m, 2H), 5.82–5.73 (m, 1H), 5.04–4.99 (m, 2H), 4.51 (s, 1H), 4.28–4.27 (d, J = 4.0 Hz, 2H). 2.7.5. Synthetic procedure for compound 50 To a solution of formylpyrimidinetrione 4 (10 mmol) in 100 mL dry THF was added isoniazid (10 mmol, 1.0 equiv), the mixture was stirred for 4 h under room temperature. The solvent was evaporated under vacuum and the residue was washed with ethanol to afford the target compound 50 as yellow solid. Yield 88.4%. mp. 150–152 °C. 1H NMR (400 MHz, DMSO‑d6) δ 11.14–11.02 (m, 1H), 8.78–8.77 (m, 2H), 8.21–8.20 (m, 1H), 7.77–7.76 (m, 2H), 5.85–5.75 (m, 1H), 5.09–5.01 (m, 2H), 4.32 (t, J = 4.0 Hz, 2H), 3.43–3.38 (m, 1H). 13C NMR (100 MHz, DMSO‑d6) δ 164.52, 164.28, 163.69, 163.50, 162.73, 162.55, 156.34, 156.30, 151.01, 150.95, 150.72, 139.46, 139.42, 133.61, 133.54, 121.83, 116.50, 116.22. HRMS (ESI) m/z 316.1035 [M +H]+, m/z 338.0840 [M+Na]+ (calcd for C14H13N5O4, 315.0968).
2.9. In vivo efficacy of compound 50 in mouse acute infection model Forty specific pathogen-free, female C57BL/6 mice aged 6 weeks were obtained from Beijing HFK Bio-Technology. All experiments were done by Biosafety of ABSL3 Lab of Wuhan University, State Key Laboratory of Virology. All mice were randomly divided into ten cages, feeding one week, to adapt to the environment. Mice were infected by nasal drops with one dose of 1 × 106 CFU of M. tuberculosis H37Rv (provided by State Key Laboratory of Virology). Infection dose was 171
Bioorganic Chemistry 85 (2019) 168–178
Y. Gao et al.
Fig. 3. Low-energy binding conformations of compound 50 bound to the active site of DprE1 generated by virtual ligand docking. A. Compound 50 depicted as the ball-and-stick model showing carbon (yellow), hydrogen (grey), oxygen (red) and nitrogen (blue) atoms. Compound 50 formed obvious hydrophobic interactions with DprE1 and FAD. B. Hydrogen bonds were predicted between compound 50 with DprE1. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
verified by incubating the inoculation fluid, whole-lung and wholespleen homogenate on 7H10 plates for 48 h postinfection. Mice treated with PBS were negative controls. After two days from infected, the lung and spleen bacterial load were assessed to make sure the real infected dose was 100 CFU per mouse. After infected, mice were divided randomly into five groups, nine of one group. After making sure the dose of the infection, C57BL/6 mice were administered by oral gavage with PBS, INH (10 mg/kg) in two groups. The other three groups, C57BL/6
mice were administered by oral gavage with 10 mg/kg, 30 mg/kg of compound 50. After treatment of ten days, the drug therapy efficiency was ascertained by evaluation of the bacterial load in lung, spleen and their histopathology compared with negative group and organs before treated. Organs were homogenized in PBS and various dilutions were placed on 7H11 (bought from BD) plates containing ampicillin (100 mg/L), carbenicillin (50 mg/L), amphotericin B (2.5 mg/L), polymyxin B (25 mg/L), TCH (2 mg/L). Plates were incubated at 37 °C for 172
Bioorganic Chemistry 85 (2019) 168–178
Y. Gao et al.
Fig. 4. The anti-TB activity of selected compounds in vitro. The values were recorded by Microplate Reader at 520/590 nm (n = 3).
3 week, CFU were recorded of all groups [19–21]. And histochemical staining and acid-fast staining were performed of the lung tissue following formaldehyde fixation. All animals were maintained in accordance with the Regulations of Experimental Animal Administration issued by the State Committee of Science and Technology of People’s Republic of China. All animal experimental protocols were approved by the Laboratory Animal Center of Huazhong University of Science and Technology and the Ethics Committee.
2.11. Statistical analysis All data are reported as means ± SD. Differences were analyzed by Student’s t-test (with 95% confidence interval). p values < 0.05 were considered as statistically significant. 3. Results 3.1. Structure-based virtual ligand screening identified a small molecule set binding to DprE1
2.10. Acute toxicity
The virtual screening had become an efficient, specifically targeted technology compared to traditional high-throughput screening (HST) [11–13]. DprE1 (PDB ID: 4FDO) was virtually screened against ∼6,200,000 small molecules from ZINC lead like database and natural products database [10]. According to the occupancy of the active binding pocket, many scaffolds were further excluded. Lipinski Rule (Rule of 5) is a rule of thumb to evaluate druglikeness. Scoring more negative than −32 generally represents strong interaction between target and ligand in ICM-Pro. According to Rule of 5 and scoring more negative than −32, the final set was reduced to a total of 63 compounds from 6.2 million (Table 1, Table S1). All of these 63 compounds have not been used in clinic and were purchased from Molport in our study. Moreover, none of them was reported as DprE1 inhibitors, or
KM mice of both sex (10 females and 10 males) were divided into control and experimental groups, which were obtained from Laboratory animal center Tongji Medical College, Huazhong University of Science and Technology. The first group served as a normal control treated with distilled water (10 mL/kg). And compound groups were orally administrated compound 50 on 2000 mg/kg a single time among 24 h. Along the 14 consecutive days, the signs of toxicity were observed every day. The changes were also noted on body weight, mortality, behavior of the animal. Acute toxicity caused a set of experimental animals with 50% subject compounds dose for quantitative evaluation of death (LD50). Formaldehyde fixation and histochemical staining of the lung, heart, liver, spleen and kidney tissue were assessed at the end [22]. 173
Bioorganic Chemistry 85 (2019) 168–178
Y. Gao et al.
with antibacterial and anti-tubercular activities. 3.2. A phenotypic screen verified M. smegmatis growth inhibition of compound 50 Then, the in vitro antibacterial activities of compounds 1–63 were conducted through the Kirby–Bauer disc agar diffusion assay. INH was picked as a reference drug. Since M. tuberculosis is infectious, the amenable substitute M. smegmatis was used [15,16]. Among the 63 compounds, 12 of them displayed notable activity, with inhibition zone greater than 14 mm (Tables 1, S1 and Fig. 2). 3.3. Microscale thermophoresis (MST) confirmed binding affinity of compounds to DprE1 To investigate the results of molecular docking, microscale thermophoresis (MST) assay was subsequently employed to validate the binding interactions between the 12 effective compounds with DprE1. MST, a precious method, can capture the fluorescence changes of proteins during thermal immortalization, so as to realize the study of protein-protein or protein-small molecule interaction [17]. The dissociation constant (Kd) measured by MST reflects the affinity between protein and ligand. Among all of the selected candidates, compound 50 demonstrated the highest affinity to DprE1 with Kd of 25 ± 2 nM (Fig. 1). 3.4. Molecular docking elucidated the binding mode of compound 50 with DprE1 In order to further elucidate the binding mode of compound 50 with DprE1, molecular docking was carried out. The lowest-energy binding conformation of compound 50 was shown as Fig. 3. Docking results demonstrated that compounds bind to the active site of the enzyme. The ligand binding pocket of DprE1 was zig-zag shaped, where many hydrophobic amino acid, including His132, Ile113, Tyr314, Phe320, TRP323, Val365, and Phe369, form a relatively hydrophobic envelop (Fig. 3A). Compound 50 formed obvious hydrophobic interactions with many of these amino acids. The pyrimidinetrione ring of compound 50 also formed strong hydrophobic interaction with flavine adenine dinucleotide (FAD) (Fig. 3A). Hydrogen bonds were predicted between two carbonyl oxygens of compound 50 with Lys418, and pyridine nitrogen with Asn324 (Fig. 3B).
Fig. 5. Anti-TB efficacy in agents-treated C57BL/6 acute infection model. Mice (n = 10) were i.v. challenged with 1.2 × 106 CFU of virulent M. tuberculosis H37Rv. After treatment, weight was recorded (A); lung and spleen tissue sections were fixed and embedded for HE staining and acid-fast (B). Arrows indicate AF staining positive bacteria. Pathological changes were scored.
3.5. A phenotypic screen verified M. tuberculosis growth inhibition of compound 50 For the compounds that have both activity against M. smegmatis and high affinity towards DprE1, a test to determine minimal inhibitory
Scheme 1. The synthesis of compound 50. A. Rational design and synthetic strategy for the DprE1 inhibitor. Reagents and conditions: (i) KOCN, in H2O, 55 °C for 4 h. (ii) Dimethyl malonate, NaOMe in MeOH, reflux for 6 h. (iii) Trimethyl orthoformate, TsOH. (iv) Isonicotinic acid hydrazide, in MeOH, rt for 6 h. B. The three tautomers of enamine 5 (compound 50) in solution. 174
Bioorganic Chemistry 85 (2019) 168–178
Y. Gao et al.
Fig. 6. The bacterial load of organs was enumerated in agents-treated C57BL/6 acute infection model. After treatment, lungs (A) and spleen (B) were harvested and the bacterial load was enumerated. Results are shown as mean ( ± SEM) log10 CFU/lung, *p < 0.01, **0.001 < p < 0.01, ***p < 0.001, PBS vs. other groups.
Fig. 7. Cytotoxicity testing of compound 50. The cytotoxicity of compound 50 was measured on Vero, L929 and Lo2 (n = 3).
concentration against M. tuberculosis was performed. Compound 50 presented a comparable activity to that of INH (1 μg/mL, 7.29 μM), with an MIC of 3.1 μg/mL (9.75 μM) (Fig. 4).
were treated with compound 50 (10 mg/kg, 30 mg/kg), or INH (10 mg/ kg). After treatment, bacterial load of the lung and spleen along with those histopathological changes were calculated (Fig. 6). Compared with PBS control group, the CFU dropped 1.0 log in lung and 0.6 logs in spleen for the group of compound 50 (10 mg/kg) similar to the potency of INH group (Fig. 6). And the CFU dropped 2.1 and 1.3 logs in lung and spleen, respectively, for the group of the compound 50 (30 mg/kg), better than the INH group (10 mg/kg) (Fig. 6). As the figure showed, the most severe granuloma-like histopathological changes were clearly visible improvement in lungs, especially for the group of compound 50 (30 mg/kg), little M. tuberculosis bacterium could be found in lungs (Fig. 5). Thus, compound 50 indicated a significant efficacy in vivo on reducing the bacterial burden in lungs in a dose-dependent tendency compared to PBS controls.
3.6. The synthesis of compound 50 In order to carry out subsequent studies, we synthesized compound 50. As shown in Scheme 1, allylurea (2) was prepared from potassium cyanate and allyl amine hydrochloride according to the literature procedure [23–25], then was cyclized with dimethyl malonate in the presence of sodium methoxide. Refluxing the obtained pyrimidinetrione (3) in trimethyl orthoformate with a catalytic amount of PTSA for several hours afforded the formylpyrimidinetrione (4) in good yield [26]. The condensation between 4 and isonicotinyl hydrazide followed by a 1,3-proton transfer process in one-pot under room temperature gave the target compound 50. Similar to tetramic acid derivatives [27–29], the resulting enamine 5 (compound 50) can exist as endo and exo-enol and keto tautomers in solution, which was a mixture of the E- and Z-isomers [30–32]. Since the two configurations (E/Z) of this enamine-typed compound could interconvert into each other rapidly in polar solvent through an imine intermediate, we speculate that the geometry of the double bound would not affect its interaction with DprE1, or the different interactions of two isomers with the enzyme could not be measured.
3.8. Compound 50 showed no obvious toxicity in vitro and in vivo To check out the safety of compound 50, we test its cytotoxicity on Vero, Lo2 and L929 cells in vitro. As shown in Fig. 7, IC50 values for all cell lines were greater than 500 μM, which were more than 100 times higher compared with the effective concentration. These results demonstrated no obvious cytotoxicity of compound 50, suggesting that it would have a wide therapeutic window in vivo. In order to verify the conjecture above, an acute toxicity assay in mice was conducted [22]. During the assay, no decline on the body weight was observed (Fig. 8A). The weight of mice and the formaldehyde fixation of lung, spleen, heart, liver, kidney tissues were no significantly different compared to control groups, indicating that no any toxicity of compound 50 orally gavage in a dose of 2000 mg/kg (Fig. 8A and B). During 14 days of observation, no mouse died. The LD50 of compound 50 was higher than 2000 mg/kg. These results demonstrated compound 50 would be safe if it is used for treatment.
3.7. Compound 50 is efficacious in acute M. tuberculosis infection mouse model To identify the efficacy of compound 50, we carried out an in vivo experiment in an acute infection model. Specific pathogen-free, female C57BL/6 mice were infected by nasal drops with the dose of 1 × 106 CFU of M. tuberculosis H37Rv. After infection for 14 days, mice 175
Bioorganic Chemistry 85 (2019) 168–178
Y. Gao et al.
Fig. 8. Acute toxicity in vivo. A. body weight was recorded; B. Spleen, lung, heart, liver and kidney tissue sections were fixed and embedded for HE staining. Groups a, b represent male gavage with saline and compound 50, respectively; groups c, d represent female gavage with saline and compound 50, respectively.
4. Discussion and conclusions
Comprising a similar scaffold, compound 49 and 50 showed different anti-TB effect. Indeed, it may clarify the mode of compound 50. As the result of molecular docking, Fig. 3 showed that compound 50 deeply inserted into the active site of DprE1, forming several obvious hydrogen bonding interactions. Among them, the hydrogen bond between pyridine nitrogen with Asn324 is worthy of attention, which may cause compound 50 to have better anti-TB effects than compound 49. MST confirmed better binding potency of compound 50. Following that, inhibition against M. smegmatis and M. tuberculosis verified compound 49 was really weaker than compound 50. Therefore, compound 50 was preferentially picked up, rather than compound 49. Compound 50 with the isoniazid fragment is a potent DprE1 inhibitor, but INH is not. So compound 50 and INH must kill M. tuberculosis in different mode. Both of them achieved similar antimicrobial effects on M. smegmatis and M. tuberculosis in vitro and in vivo. Since compound 50 is stable in vitro, it was confirmed that compound 50 could behave an
The economic and social damage caused by M. tuberculosis and the toxicity of anti-tuberculosis drugs in clinic are driving the emergence of new drugs with the anti-tuberculosis effects and the safe treatment profile. Here we have applied a structure-based virtual ligand screening approach to explore target-based anti-tuberculosis agents among 6.2 million compounds. This method is proved to be quick and efficient to search for lead-like compounds. And the pyrimidinetrione, compound 50 was identified as a novel and potent DprE1 inhibitor with high affinity at nanomolar scale through MST assay. Moreover, compound 50 showed commendable anti-tuberculosis efficiency in vitro and in vivo, which was comparable to the first-line anti-tuberculosis drugs. What need to be emphasized is that compound 50 did not show any obvious toxicity in cytotoxicity testing and acute toxicity assessment. It exactly predicted a favorable treatment window of compound 50. 176
Bioorganic Chemistry 85 (2019) 168–178
Y. Gao et al.
inhibitory effect against M. tuberculosis in vitro. This exclude the possibility that compound 50 has to metabolize to INH to take effects. We concluded that compound 50 come into effects in a different way from INH. Taken together, compound 50 could be a novel type of anti-tuberculosis agent or lead compound via inhibiting DprE1. Of course, it is worthy further to investigate whether compound 50 can be converted to INH to generate additional anti-tubercular effects in vivo. In addition, our research process of structure-based drug discovery will also provide a vital method for the drug discovery in major diseases. Furthermore, other important species known for human pathogenesis, such as M. leprae, the causative agent of leprosy persisting in developing countries, and M. avium intracellular complex (MAC), an important group including nontuberculous mycobacteria, are responsible for opportunistic infections in immunocompromised individuals [33,34]. Since the two species both depend on DprE1 to synthesize their cell walls, these diseases could be also solved by compound 50, which is worthy to be further validated. Additionally, these studies emphasize the great potential of inhibiting critical mycobacteria enzymes and reveal a potential lead compound worthy further investigation.
11354–11359, https://doi.org/10.1073/pnas.1205735109. [8] R.V. Chikhale, M.A. Barmade, P.R. Murumkar, M.R. Yadav, Overview of the development of DprE1 inhibitors for combating the menace of tuberculosis, J. Med. Chem. 61 (2018) 8563–8593, https://doi.org/10.1021/acs.jmedchem.8b00281. [9] R. Sommer, J. Neres, J. Piton, N. Dhar, A. van der Sar, R. Mukherjee, T. Laroche, P.J. Dyson, J.D. McKinney, W. Bitter, V. Makarov, S.T. Cole, Fluorescent benzothiazinone analogues efficiently and selectively label dpre1 in mycobacteria and actinobacteria, ACS Chem. Biol. 13 (2018) 3184–3192, https://doi.org/10.1021/ acschembio.8b00790. [10] J.J. Irwin, T. Sterling, M.M. Mysinger, E.S. Bolstad, R.G. Coleman, ZINC: a free tool to discover chemistry for biology, J. Chem. Inf. Model. 52 (2012) 1757–1768, https://doi.org/10.1021/ci3001277. [11] R. Abagyan, M. Totrov, D. Kuznetsov, ICM? A new method for protein modeling and design: applications to docking and structure prediction from the distorted native conformation, J. Comput. Chem. 15 (1994) 488–506, https://doi.org/10.1002/jcc. 540150503. [12] E.-L. Lyle, G.L. Hayward, M. Thompson, Acoustic coupling of transverse waves as a mechanism for the label-free detection of protein–small molecule interactions, The Analyst 127 (2002) 1596–1600, https://doi.org/10.1039/B209051C. [13] D.-L. Ma, D.S.-H. Chan, C.-H. Leung, Molecular docking for virtual screening of natural product databases, Chem. Sci. 2 (2011) 1656–1665, https://doi.org/10. 1039/C1SC00152C. [14] L. Borriello, M. Montès, Y. Lepelletier, B. Leforban, W.-Q. Liu, L. Demange, B. Delhomme, S. Pavoni, R. Jarray, J.L. Boucher, S. Dufour, O. Hermine, C. Garbay, R. Hadj-Slimane, F. Raynaud, Structure-based discovery of a small non-peptidic Neuropilins antagonist exerting in vitro and in vivo anti-tumor activity on breast cancer model, Cancer Lett. 349 (2014) 120–127, https://doi.org/10.1016/j.canlet. 2014.04. 004. [15] V. Chaturvedi, N. Dwivedi, R.P. Tripathi, S. Sinha, Evaluation of Mycobacterium smegmatis as a possible surrogate screen for selecting molecules active against multi-drug resistant Mycobacterium tuberculosis, J. Gen. Appl. Microbiol. 53 (2007) 333–337, https://doi.org/10.2323/jgam.53.333. [16] G. Etienne, The cell envelope structure and properties of Mycobacterium smegmatis mc2155: is there a clue for the unique transformability of the strain? Microbiology 151 (2005) 2075–2086, https://doi.org/10.1099/mic.0.27869-0. [17] Y. Mao, L. Yu, R. Yang, L. Qu, P.deB. Harrington, A novel method for the study of molecular interaction by using microscale thermophoresis, Talanta 132 (2015) 894–901, https://doi.org/10.1016/j.talanta.2014.09.038. [18] P. Escalante, P. Escalante, S. Ramaswamy, H. Sanabria, H. Soini, X. Pan, O. Valiente-Castillo, J.M. Musser, Genotypic characterization of drug-resistant Mycobacterium tuberculosis isolates from Peru, Tuber. Lung Dis. 79 (1998) 111–118, https://doi.org/10.1054/tuld.1998.0013. [19] F. Wang, D. Sambandan, R. Halder, J. Wang, S.M. Batt, B. Weinrick, I. Ahmad, P. Yang, Y. Zhang, J. Kim, M. Hassani, S. Huszar, C. Trefzer, Z. Ma, T. Kaneko, K.E. Mdluli, S. Franzblau, A.K. Chatterjee, K. Johnsson, K. Mikusova, G.S. Besra, K. Futterer, S.H. Robbins, S.W. Barnes, J.R. Walker, W.R. Jacobs, P.G. Schultz, Identification of a small molecule with activity against drug-resistant and persistent tuberculosis, Proc. Natl. Acad. Sci. 110 (2013) E2510–E2517, https://doi.org/10. 1073/pnas.1309171110. [20] J. Liang, X. Teng, X. Yuan, Y. Zhang, C. Shi, T. Yue, L. Zhou, J. Li, X. Fan, Enhanced and durable protective immune responses induced by a cocktail of recombinant BCG strains expressing antigens of multistage of Mycobacterium tuberculosis, Mol. Immunol. 66 (2015) 392–401, https://doi.org/10.1016/j.molimm.2015.04.017. [21] P.S. Shirude, R. Shandil, C. Sadler, M. Naik, V. Hosagrahara, S. Hameed, V. Shinde, C. Bathula, V. Humnabadkar, N. Kumar, J. Reddy, V. Panduga, S. Sharma, A. Ambady, N. Hegde, J. Whiteaker, R.E. McLaughlin, H. Gardner, P. Madhavapeddi, V. Ramachandran, P. Kaur, A. Narayan, S. Guptha, D. Awasthy, C. Narayan, J. Mahadevaswamy, K. Vishwas, V. Ahuja, A. Srivastava, K. Prabhakar, S. Bharath, R. Kale, M. Ramaiah, N.R. Choudhury, V.K. Sambandamurthy, S. Solapure, P.S. Iyer, S. Narayanan, M. Chatterji, Azaindoles: noncovalent DprE1 inhibitors from scaffold morphing efforts, kill Mycobacterium tuberculosis and are efficacious in vivo, J. Med. Chem. 56 (2013) 9701–9708, https://doi.org/10.1021/ jm401382v. [22] W. Li, M. Zheng, S. Wu, S. Gao, M. Yang, Z. Li, Q. Min, W. Sun, L. Chen, G. Xiang, H. Li, Benserazide, a dopadecarboxylase inhibitor, suppresses tumor growth by targeting hexokinase 2, J. Exp. Clin. Cancer Res. 36 (2017) 58, https://doi.org/10. 1186/s13046-017-0530-4. [23] R.C. Moschel, W.R. Hudgins, A. Dipple, Reactivity effects on site selectivity in nucleoside aralkylation: a model for the factors influencing the sites of carcinogennucleic acid interactions, J. Org. Chem. 51 (1986) 4180–4185, https://doi.org/10. 1021/jo00372a015. [24] S. Fouladdel, A. Khalaj, N. Adibpour, E. Azizi, Synthesis and cytotoxicity of some biurets against human breast cancer T47D cell line, Bioorg. Med. Chem. Lett. 20 (2010) 5772–5775, https://doi.org/10.1016/j.bmcl.2010.07.137. [25] X. Min, J. Liu, Y. Dong, M. Hussain, One-pot assembly towards ω-substituted arylbiurets from aromatic amines, potassium cyanate, and glacial acetic acid, Synthesis 50 (2018) 341–348, https://doi.org/10.1055/s-0036-1590934. [26] D.M. Neumann, A. Cammarata, G. Backes, G.E. Palmer, B.S. Jursic, Synthesis and antifungal activity of substituted 2,4,6-pyrimidinetrione carbaldehyde hydrazones, Bioorg. Med. Chem. 22 (2014) 813–826, https://doi.org/10.1016/j.bmc.2013.12. 010. [27] Y.-C. Jeong, M.G. Moloney, Synthesis of and tautomerism in 3-acyltetramic acids, J. Org. Chem. 76 (2011) 1342–1354, https://doi.org/10.1021/jo102304y. [28] Y.-C. Jeong, Z. Bikadi, E. Hazai, M.G. Moloney, A detailed study of antibacterial 3acyltetramic acids and 3-acylpiperidine-2,4-diones, ChemMedChem (2014), https://doi.org/10.1002/cmdc.201402093 n/a-n/a.
Funding This work was supported by the National Natural Science Foundation of China (NSFC) [grant number 81473254, 81773637, U1703111, and 81773594] and National Major Scientific and Technological Special Project for “Significant New Drugs Development” during the 13th Five-Year Plan Period [grant number 2019ZX09721001-004-007]. Acknowledgement We acknowledge support from Schrödinger Release 2018-4: Maestro, Schrödinger, LLC, New York, NY, 2018. Competing interests The authors have declared that no conflict of interest exists. Appendix A. Supplementary material Supplementary data to this article can be found online at https:// doi.org/10.1016/j.bioorg.2018.12.018. References [1] M. Chatterji, R. Shandil, M.R. Manjunatha, S. Solapure, V. Ramachandran, N. Kumar, R. Saralaya, V. Panduga, J. Reddy, P. Kr, S. Sharma, C. Sadler, C.B. Cooper, K. Mdluli, P.S. Iyer, S. Narayanan, P.S. Shirude, 1,4-Azaindole, a potential drug candidate for treatment of tuberculosis, Antimicrob. Agents Chemother. 58 (2014) 5325–5331, https://doi.org/10.1128/AAC.03233-14. [2] A. Nader, T.A. deRoon-Cassini, T.B. Meier, J.B. Hoelzle, L.D. Nelson, Relationship between somatic symptoms, peripheral inflammation, and clinical outcomes after traumatic injury, Brain Behav. Immun. 66 (2017) e26, , https://doi.org/10.1016/j. bbi.2017.07.099. [3] Organisation mondiale de la santé, Global Tuberculose Report: 2015, Word Health Organization, Genève (Suisse), 2015. [4] S. Crunkhorn, Antibacterial agents: new routes to tuberculosis treatment, Nat. Rev. Drug Discov. 16 (2017) 600–601, https://doi.org/10.1038/nrd.2017.156. [5] H. Li, G. Jogl, Crystal structure of decaprenylphosphoryl-β-D-ribose 2′-epimerase from Mycobacterium smegmatis, Prot.: Struct. Funct. Bioinform. 81 (2013) 538–543, https://doi.org/10.1002/prot.24220. [6] J. Neres, F. Pojer, E. Molteni, L.R. Chiarelli, N. Dhar, S. Boy-Rottger, S. Buroni, E. Fullam, G. Degiacomi, A.P. Lucarelli, R.J. Read, G. Zanoni, D.E. Edmondson, E. De Rossi, M.R. Pasca, J.D. McKinney, P.J. Dyson, G. Riccardi, A. Mattevi, S.T. Cole, C. Binda, Structural basis for benzothiazinone-mediated killing of Mycobacterium tuberculosis, Sci. Transl. Med. 4 (2012), https://doi.org/10.1126/ scitranslmed.3004395 150ra121-150ra121. [7] S.M. Batt, T. Jabeen, V. Bhowruth, L. Quill, P.A. Lund, L. Eggeling, L.J. Alderwick, K. Futterer, G.S. Besra, Structural basis of inhibition of Mycobacterium tuberculosis DprE1 by benzothiazinone inhibitors, Proc. Natl. Acad. Sci. 109 (2012)
177
Bioorganic Chemistry 85 (2019) 168–178
Y. Gao et al. [29] Y.-C. Jeong, M.G. Moloney, Synthesis and antibacterial activity of monocyclic 3carboxamide tetramic acids, Beilstein J. Org. Chem. 9 (2013) 1899–1906, https:// doi.org/10.3762/bjoc.9.224. [30] Y.-C. Jeong, M. Moloney, Antibacterial barbituric acid analogues inspired from natural 3-acyltetramic acids; Synthesis, tautomerism and structure and physicochemical property-antibacterial activity relationships, Molecules 20 (2015) 3582–3627, https://doi.org/10.3390/molecules20033582. [31] I.V. Ukrainets, L. Yangyang, N.L. Bereznyakova, A.V. Turov, 4-hydroxy-2-quinolones. 169*. synthesis and bromination of 1-allyl-3-(arylamino-methylene)quinoline-2,4-(1h,3h)-diones, Chem. Heterocycl. Compd. 45 (2009) 1235–1240, https:// doi.org/10.1007/s10593-010-0412-6.
[32] S. Han, F.-F. Zhang, H.-Y. Qian, L.-L. Chen, J.-B. Pu, X. Xie, J.-Z. Chen, Development of quinoline-2,4(1 H,3 H)-diones as potent and selective ligands of the cannabinoid type 2 receptor, J. Med. Chem. 58 (2015) 5751–5769, https://doi.org/10.1021/acs. jmedchem.5b00227. [33] J. Gawad, C. Bonde, Current affairs, future perspectives of tuberculosis and antitubercular agents, Indian J. Tuberc. 65 (2018) 15–22, https://doi.org/10.1016/j. ijtb.2017.08.011. [34] P.K. Crellin, R. Brammananth, R.L. Coppel, Decaprenylphosphoryl-β-D-Ribose 2′Epimerase, the target of benzothiazinones and dinitrobenzamides, is an essential enzyme in Mycobacterium smegmatis, PLoS One 6 (2011) e16869, , https://doi.org/ 10.1371/journal.pone.0016869.
178