A New Pathway for the Preparation of Pyrano[2,3-c]pyrazoles and molecular Docking as Inhibitors of p38 MAP Kinase.

We report a new pathway to synthesize pyrano[2,3-c]pyrazoles and their binding mode to p38 MAP kinase. Pyrano[2,3-c]pyrazole derivatives have been prepared through a four-component reaction of benzyl alcohols, ethyl acetoacetate, phenylhydrazine, and malononitrile in the presence of sulfonated amorphous carbon and eosin Y as catalysts. All products were characterized by melting point, 1H and 13C NMR, and HRMS (ESI). The products were screened in silico for their binding activities to both the ATP-binding pocket and the lipid-binding pocket of p38 MAP kinase, using a structure-based flexible docking provided by the engine ADFR. The results showed that eight synthesized compounds had a higher affinity to the lipid pocket than to the other target site, which implied potential applications as allosteric inhibitors. Finally, the most biologically active compound, 5, had a binding affinity comparable to those of other proven lipid pocket inhibitors, with affinity to the target pocket reaching -10.9932 kcal/mol, and also had the best binding affinity to the ATP-binding pockets in all of our products. Thus, our research provides a novel pathway for synthesizing pyrano[2,3-c]pyrazoles and bioinformatic evidence for their biological capability to block p38 MAP kinase pockets, which could be useful for developing cancer or immune drugs.

Introduction

Pyranopyrazoles are some of the essential compounds in pharmacological chemistry.[1−3] Some of the great biological activities of the pyranopyrazole framework are anti-inflammatory,[4−6] anticancer,[7−9] antimicrobial,[10] antioxidant,[11] anticholinesterase,[12] analgesics,[5] and antimicrobial activity (Scheme ).[13] Generally, pyranopyrazoles have been synthesized from aldehydes, malononitrile, ethyl acetoacetate, and hydrazine derivatives through multicomponent reactions, which was one of the effective methods applied.[14−21] Several protocols have been reported for using various catalysts such as nanoparticles,[22−26] organocatalysts (MDOs),[27] Fe3O4@THAM-SO3H,[22] Fe3O4@SiO2@CPTMO@DEA-SO3H,[23] the water extract of banana peels,[28] chitosan materials,[29] nano-BF3/MNPs,[30] nano-NH4H2PO4/Al2O3,[31] disulfonic acid imidazolium chloroaluminate ([Dsim]AlCl4),[32] and solid catalysts.[33]

Scheme 1

Biologically Active Pyrano[2,3-c]pyrazoles

Sulfonated amorphous carbons have been studied extensively in catalysis, adsorbents, and electrochemistry.[11,34−44] In 2004, Hara and co-workers developed new catalysts, a carbon-based solid acids with sulfonic acid groups (SO3H). The sulfonated amorphous carbons were applied for organic syntheses such as esterification,[45−50] rearrangement,[51,52] C–C coupling reactions,[53] condensations,[54,55] amidations, and amination.[55,56]

Mitogen-activated protein kinase (MAPK) modules belong to a protein kinase family conserved from ancestral unicellular organisms and function as a trigger for phosphorylation switches in substrate proteins related to various roles, including differentiation, proliferation, and cell death decisions.[57] In the multicellular organism group, three well-defined MAPK subfamilies respond to different extracellular signals and could be exploited as drug targets. The extracellular signal regulation using kinases 1 and 2 (ERK1/2) could be activated by mitogens and growth hormones and contribute to cell division, thus targeting chemotherapies. The c-Jun N-terminal kinases (JNKs) oversee apoptosis pathways with regard to environmental stress and growth factors; thus, inhibiting these kinases could be critical in cancer cell treatment. Finally, p38 MAP kinases are activated by bacterial liposaccharides, cytokines, and cellular stresses, including osmotic shock and UV radiation, which could be linked to inflammatory diseases such as asthma and autoimmunity.[57−61]

A typical eukaryote p38 MAPK contains two lobes: a C-terminal lobe functioning as a substrate interaction site (MAPK insert) and an N-terminal lobe consisting of an ATP-binding site and a catalytic site.[62] With the underlying importance of MAPK as a pharmaceutical target, many attempts have been made to achieve optimal p38 MAPK inhibitors by targeting different sites in p38 MAPK. However, most have failed in the clinical phases due to unsatisfying selectivity and tolerance to the human body, as kinases’ structure, especially the ATP-binding site in the N-terminal lobe, is highly conserved among the MAPK family.[61] There have been various patented inhibitor drugs for MAPK targeting different pockets, and they are mainly categorized into six classes from I to VI, with the discovery of the three classes IV–VI having been made possible by new information on the p38 MAPK structure.[59] For the three original inhibitor classes, while the targets are all related to the ATP-binding site, with type I being ATP-competitive inhibitors and type II binding to the inactive form of MAPK, type III inhibitors bind to allosteric sites within the ATP pocket, which would decrease the off-target inhibitions and avoid inhibitor-resistance mutations in the ATP-binding site.[63]

Newly found type IV inhibitors, together with types V and VI, target the allosteric site outside the catalytic domain. With type IV inhibitors, the compound will bind to the allosteric site, thus disrupting the correct conformation for enzymatic activity, such as the D-domain recruitment site (DRS) or the MAPK insert site (hydrophobic pocket). A combination of competitive binding and allosteric interaction is the basis of type V bivalent inhibitors. Finally, type VI inhibitors provide covalent and possibly irreversible effects at both the enzymatic site and the allosteric site.[63] With the advantages of allosteric inhibitors, this study will analyze the affinity of eight potential pyrazoles derivatives to the canonical ATP-binding pocket and a newly found lipid-binding pocket locating near the MAPK insert site, with the anticipation of finding potent type IV–VI inhibitors.[60]

Experimental and Computational Section

Chemicals and Instrumentation

Chemicals and Supplies

Ethyl acetoacetate (grade ReagentPlus, assay 99%), phenylhydrazine (assay 97%), malononitrile (assay, 99%), tert-butyl hydroperoxide (TBHP) (70% in water), choline chloride (98%), 4-methylbenzyl alcohol (assay 98%), 4-methoxybenzyl alcohol (assay 98%), furfuryl alcohol (assay 98%), rhodamine B (≥95%, HPLC), and phenylhydrazine (assay 97%) were obtained from Sigma-Aldrich. Eosin Y (Certistain dry dye used for counterstaining), benzyl alcohol (special grade), urea (GR for analysis ACS, Reag. Ph Eur), hydrazine hydrate (80% solution in water), sulfuric acid 98% (for analysis EMSURE), TLC (silica gel 60 F254), and ethanol (absolute for analysis EMSURE ACS, ISO, Reag. Ph Eur) were obtained from Merck. Ethyl acetate (purity ≥ 99.5%), and n-hexane (purity ≥ 99.5%) were obtained from Xilong Chemical Co., Ltd. (China).

Analytical Techniques

The 1H and 13C NMR spectra were recorded on a Bruker Advance 500 instrument using DMSO-d6 as solvent and the solvent peaks as a reference. HRMS (ESI) data were collected using a Bruker micrOTOF-QII MS instrument at 80 eV. High-performance liquid chromatography was carried out with an Agilent Technologies Infinity instrument with a photodiode array detector (HPLC-PAD).

Preparation of [cholineCl][urea]2

[cholineCl][urea]2 was synthesized according to previous literature.[64−69] A mixture of choline chloride (5 mmol, 0.695 g) and urea (10 mmol, 0.600 g) was heated at 100 °C until a clear homogeneous mixture was obtained (about 30 min). After completion of the reaction, the mixture was cooled to room temperature. Then, the [cholineCl][urea]2 was washed several times with diethyl ether and dried under vacuum before use.

Preparation of Pyrano[2,3-c]pyrazoles

A solution of ethyl acetoacetate (1.0 mmol), phenylhydrazine (1.0 mmol), and [cholineCl][urea]2 (5.0 mmol) was charged in a 25 mL round-bottom flask and reacted at room temperature for 7–10 min to form mixture A. On the other hand, a mixture of benzyl alcohol (1.0 mmol), eosin Y (5 mol %), and TBHP (3 equiv) was irradiated with blue LEDs for 28 h at room temperature. A mixture of aldehyde (1.0 mmol), malononitrile (1.0 mmol), and AC-SO3H catalyst (5 mg) was stirred in another 10 mL round-bottom flask to form mixture B. Blending mixtures A and B was carried out by continuous stirring at room temperature to investigate the reaction time. After the reaction was complete (as checked by TLC), the reaction mixture was washed with water (3 × 5 mL). The products and the catalyst were separated by recrystallization in hot ethanol (10–15 mL). The desired products were obtained pure and characterized by 1H, 13C NMR, HRMS, and melting points.

6-Amino-3-methyl-4-phenyl-2,4-dihydropyrano[2,3-c]pyrazole-5-carbonitrile[22,29] (1)

White solid, mp 258–259 °C (n-hexane/ethyl acetate 9/1, Rf = 0.44).1H NMR (DMSO-d6, 500 MHz): δ 12.06 (s, 1H), 7.30 (t, J = 7.0 Hz, 2H), 7.21 (t, J = 7.0 Hz, 1H), 7.15 (d, J = 7.0 Hz, 2H), 6.82 (s, 2H), 4.57 (s, 1H), 1.77 (s, 3H). 13C NMR (DMSO-d6, 125 MHz): δ 161.4, 155.3, 144.9, 136.0, 128.9, 127.9, 127.19, 121.1, 98.1, 57.8, 36.8, 10.2. HRMS (ESI): m/z [M + H]+ calcd for C14H13N4O+,253.1089; found,253.1081.

6-Amino-3-methyl-4-(p-tolyl)-2,4-dihydropyrano[2,3-c]pyrazole-5-carbonitrile[29,70] (2)

Yellowish solid, mp 206–207 °C (n-hexane/ethyl acetate 8/, Rf = 0.50).1H NMR (DMSO-d6, 500 MHz): δ 12.04 (s, 1H), 7.30 (d, J = 8.0 Hz, 1H), 7.10 (d, J = 8.0 Hz, 1H), 7.03 (d, J = 8.0 Hz, 2H), 6.80 (s, 2H), 4.52 (s, 1H), 2.27 (s, 3H), 1.76 (s, 3H). 13C NMR (DMSO-d6, 125 MHz): δ 161.3, 155.3, 142.0, 136.2, 136.0, 129.5, 127.8, 121.2, 98.2, 58.0, 36.4, 21.1, 10.2. HRMS (ESI): m/z [M + H]+ calcd for C14H15N4O+, 267.1246; found, 267.1239.

6-Amino-4-(4-methoxyphenyl)-3-methyl-2,4-dihydropyrano[2,3-c]pyrazole-5-carbonitrile[29,71] (3)

Yellow solid, mp 209–210 °C (n-hexane/ethyl acetate 8/2, Rf = 0.51). 1H NMR (DMSO-d6, 500 MHz): δ 12.04 (s, 1H), 7.06 (d, J = 8.5 Hz, 2H), 6.85 (d, J = 8.5 Hz, 2H), 6.78 (s, 2H), 4.52 (s, 1H), 3.71 (s, 3H), 1.77 (s, 3H).13C NMR (DMSO-d6, 125 MHz): δ 162.2, 161.2, 160.8, 158.5, 155.3, 137.0, 136.0, 130.4, 128.9, 127.1, 121.2, 114.9, 114.3, 98.4, 58.3, 55.9, 55.5, 36.0, 10.2. HRMS (ESI): m/z [M + H]+ calcd for C14H13N4O+, 283.1195; found, 283.1195.

6-Amino-4-(furan-2-yl)-3-methyl-2,4-dihydropyrano[2,3-c]pyrazole-5-carbonitrile[28,29] (4)

White solid, mp 215–216 °C (n-hexane/ethyl acetate 8/2, Rf = 0.52). 1H NMR (DMSO-d6, 500 MHz): δ 12.11 (s, 1H), 7.51–7.50 (m, 1H), 6.90 (s, 2H), 6.35–6.34 (m, 1H), 6.16 (d, J = 3.0 Hz, 1H), 4.76 (s, 1H), 1.96 (s, 3H). 13C NMR (DMSO-d6, 125 MHz): δ 161.5, 155.7, 154.8, 142.2, 135.81, 120.6, 110.2, 105.6, 95.1, 54.0, 29.8, 9.6. HRMS (ESI): m/z [M + H]+ calcd for C12H11N4O2: 243.0882; found: 243.0875.

6-Amino-3-methyl-1,4-diphenyl-1,4-dihydropyrano[2,3-c]pyrazole-5-carbonitrile[70] (5)

Yellowish solid, mp 170–171 °C (n-hexane/ethyl acetate 8/2, Rf = 0.37). 1H NMR (DMSO-d6, 500 MHz): δ 7.77 (d, J = 8.0 Hz, 2H), 7.48 (t, J = 7.5 Hz, 2H), 7.35–7.29 (m, 3H), 7.26–7.23 (m, 3H), 7.17 (s, 2H), 4.66 (s, 1H), 1.77 (s, 3H).13C NMR (DMSO-d6, 125 MHz): δ 159.9, 145.7, 144.1, 139.6, 139.4, 138.0, 129.8, 129.0, 128.2, 127.5, 126.6, 120.5, 120.4, 99.1, 37.3, 13.0. HRMS (ESI): m/z [M + H]+ calcd for C20H17N4O+, 329.1402; found, 329.1381.

6-Amino-3-methyl-1-phenyl-4-(p-tolyl)-1,4-dihydropyrano[2,3-c]pyrazole-5-carbonitrile[70] (6)

White solid, mp 176–177 °C (n-hexane/ethyl acetate 8/2, Rf = 0.30). 1H NMR (DMSO-d6, 500 MHz): δ 7.77 (dd, J = 1.0 Hz, 8.5 Hz, 2H), 7.47 (t, J = 7.5 Hz, 2H), 7.30 (t, J = 7.5 Hz, 1H), 7.15–7.11 (m, 6H), 4.61 (s, 1H), 2.28 (s, 3H), 1.77 (s, 3H). 13C NMR (DMSO-d6, 125 MHz): δ 168.70, 159.82, 145.77, 144.36, 144.28, 141.12, 138.05, 136.56, 129.79, 129.56, 128.12, 126.61, 120.44, 119.73, 36.89, 21.12, 13.04. HRMS (ESI): m/z [M + H]+ calcd for C21H19N4O+, 343.1559; found, 343.1560.

6-Amino-4-(4-methoxyphenyl)-3-methyl-1-phenyl-1,4-dihydropyrano[2,3-c]pyrazole-5-carbonitrile[70] (7)

White solid, mp 174–175 °C (n-hexane/ethyl acetate 8.2, Rf = 0.32). 1H NMR (DMSO-d6, 500 MHz): δ 7.77 (dd, J = 1.0 Hz, 8.5 Hz, 2H), 7.47 (t, J = 7.5 Hz, 2H), 7.30 (t, J = 7.5 Hz, 1H), 7.15 (d, J = 9.0 Hz, 2H), 7.11 (s, 2H), 6.89 (d, J = 8.5 Hz, 2H), 4.61 (s, 1H), 3.73 (s, 3H), 1.77 (s, 3H). 13C NMR (DMSO-d6, 125 MHz): δ 159.73, 158.71, 145.79, 144.32, 140.23, 138.06, 136.12, 133.97, 129.78, 129.28, 126.59, 120.44, 114.35, 99.34, 55.52, 36.48, 13.04. HRMS (ESI): m/z [M + H]+ calcd for C21H18N4O2+,359.1508; found,359.1511.

6-Amino-4-(furan-2-yl)-3-methyl-1-phenyl-1,4-dihydropyrano[2,3-c]pyrazole-5-carbonitrile (8)

White solid, mp 174–175 °C (n-hexane/ethyl acetate 8/, Rf = 0.32). 1H NMR (DMSO-d6, 500 MHz): δ 8.99 (s, 1H), 7.53 (s, 1H), 7.00 (t, J = 8.5 Hz, 16.0 Hz, 2H), 6.96 (d, J = 8.5 Hz, 2H), 6.56 (t, J = 7.0 Hz, 14.0 Hz, 1H), 6.29 (s, 2H), 4.91 (s, 1H), 1.76 (s, 3H). 13C NMR (DMSO-d6, 125 MHz): δ 169.3, 151.7, 150.1, 148.57, 146.1, 145.0, 144.1, 137.1, 129.3, 119.4, 115.3, 113.2, 112.8, 111.2, 109.7, 61.4, 55.9, 27.2, 14.3.

5-Amino-1-(2,4-dinitrophenyl)-3-phenyl-1H-pyrazole-4-carbonitrile (9)

Orange solid, mp 243–245 °C (n-hexane/ethyl acetate 7/3, Rf = 0.45). 1H NMR (DMSO-d6, 500 MHz): δ 11.64 (s, 1H), 8.85 (d, J = 3.0 Hz, 1H), 8.69 (s, 1H), 8.36 (dd, J = 3.0 Hz, 2.5 Hz, 1H), 8.09 (d, J = 9.5 Hz, 1H), 7.79–7.77 (m, 2H), 7.49–7.46 (m, 3H). 13C NMR (DMSO-d6, 125 MHz): δ 149.9, 145.0, 137.6, 134.3, 131.0, 130.2, 130.04, 129.4, 127.9, 123.4, 117.3. HRMS (ESI): m/z [M + 3H]+ calcd for C16H13N6O4+, 353.0998; found, 353.1038.

5-Amino-1-(2,4-dinitrophenyl)-3-(4-methoxyphenyl)-1H-pyrazole-4-carbonitrile (10)

Red solid, mp 255–259 °C (n-hexane/ethyl acetate 7/3, Rf = 0.55). 1H NMR (DMSO-d6, 500 MHz): δ 11.57 (s, 1H), 8.85 (d, J = 2.5 Hz, 1H), 8.63 (s, 1H), 8.35 (dd, J = 2.5 Hz, 2.5 Hz, 1H), 8.06 (d, J = 10 Hz, 1H), 7.74 (d, J = 9.0 Hz, 2H), 7.04 (d, J = 8.5 Hz, 2H), 3.81 (s, 3H). 13C NMR (DMSO-d6, 125 MHz): δ 161.8, 150.0, 145.0, 137.3, 130.2, 129.6, 126.84, 123.5, 117.2, 115.0, 55.9.

5-Amino-1-(5-bromo-2-hydroxybenzoyl)-3-phenyl-1H-pyrazole-4-carbonitrile (11)

Pinkish solid, mp 253–258 °C (n-hexane/ethyl acetate 7/3, Rf = 0.40). 1H NMR (DMSO-d6, 500 MHz): δ 11.83 (s, 2H), 8.44 (s, 1H), 8.02 (d, J = 2.5 Hz, 1H), 7.75–7.73 (m, 2H), 7.57 (dd, J = 2.5 Hz, 2.5 Hz, 1H), 7.48–7.45 (m, 3H), 6.95 (d, J = 9.0 Hz, 1H). 13C NMR (DMSO-d6, 125 MHz): δ 163.7, 158.4, 149.7, 136.5, 134.5, 131.3, 130.9, 129.3, 127.8, 120.1, 118.9, 110.5. HRMS (ESI): m/z [M – H]− calcd for C17H10BrN4O2–, 379.9914; found, 379.9869.

Receptor File Preparation

Receptor maps were built on the basis of the receptor complex structures of MAP Kinase p38 downloaded from the PDB database with the codes 1A9U (for the ATP binding pocket) and 4DLI (for the lipid-binding pocket) as identifiers. Each of the complex structures had the ligand and the receptor separated by Pymol software, version 2.2. The separated receptor files (both 1A9U and 4DLI receptors) were subsequently removed of water molecules, and hydrogen atoms and Kollman charges were added to the receptors using AutodockTool-1.5.6 and converted to .pdbqt format, before use as a template to build pocket maps with previously separated ligands as reference ligands assisted by the AGFR interface from Mgltools. All possible binding sites adjacent to the canonical binding site between the reference ligands and the ATP or lipid pocket were also included in the docking box, and rigid maps were exported as compressed zip files.

Compound File Preparation

Eight pyrano[2,3-c]pyrazole derivatives were drawn using Chemdraw 8.0 and exported in .mol file format. The structure files were then converted to the .pdb format and then the .pdbqt format using Pymol and the prepare ligand utility from the ADFR software suite, version 1.0.[72]

In order to evaluate the binding strengths of these compounds, structures of 38 ATP-pocket ligands and 26 lipid-pocket ligands in a complex with p38 MAPK were downloaded from the PDB database. The ligands in ATP pockets or lipid pockets were then separated from the complexes using Pymol and Python 3.5.6 and converted to .pdbqt format using the aforementioned method.

Docking Experiment

Docking experiments were performed with the Mgltools2–1.1 package on a 15GB, 4vCPU Instance virtual machine running on the Linux operating system.[72] Each molecule was docked 10 times to retrieve the average binding energy. To perform mass docking for eight molecules to two maps, we used the Python 3.5.6 code to run the adfr.bat file.

Analysis of Docking Results

The docking results were exported in two parts: a log file for each compound detailing the binding affinity in each running epoch, and a .pdbqt file consisting of the coordination of each compound in the best binding conformation with the receptor. The final (best) binding affinity scores in each of 10 iterations were extracted from log files and averaged to compare the scores of each compound.

Possible interactions between the tested compounds and p38 MAPK pockets were analyzed and interpreted using the Ligplot+ program developed by the European Bioinformatics Institute (EMBL-EBI). Inputs were .pdbqt files in the bound state of the compounds and receptors of the two pockets, which were joined together using Pymol software. For each compound, the docking files giving the most negative binding score after 10 iterations were chosen.[73]

Results and Discussion

Synthesis of Pyrano[2,3-c]pyrazole Derivatives

AC-SO3H was synthesized by the carbonization/sulfonation method of rice husks, which was followed a previous literature procedure, and the characterization of the AC-SO3H catalyst was carried out with FTIR, P-XRD, TGA, EDS, and SEM.[68] A [cholineCl][urea]2-deep eutectic solvent was demonstrated by heating a mixture of choline chloride and urea (in the ratio 1:2) according to previous literature.[64−66,74] The purification of [cholineCl][urea]2 was detailed in previous literature.[68]

We investigated the influence of AC-SO3H and [cholineCl][urea]2 in the preparation of pyrano[2,3-c]pyrazole derivatives via the multicomponent reaction of benzyl alcohol, phenylhydrazine, ethyl acetoacetate, and malononitrile. As can be seen in Table S2, the model reaction was tested with benzyl alcohol at room temperature in the presence of eosin Y, rhodamine B, or fluorescein as a catalyst and TBHP or K2S2O8 as an oxidizing agent under visible-light-promoted radicals for 28 h. The influence of solvents was carried out in acetonitrile, dimethylformamide, and dichloromethane and without any solvents. Then, we optimized the reaction conditions by utilizing AC-SO3H and [cholineCl][urea]2 to prepare the desired product. In the presence of [cholineCl][urea]2 as a solvent/catalyst, AC-SO3H used in different catalytic amounts (e.g., 5, 10, and 15 mg); the yields of 6-amino-3-methyl-1,4-diphenyl-1,4-dihydropyrano[2,3-c]pyrazole-5-carbonitrile were determined by an HPLC method. The reaction was carried out in the presence of TsOH as the catalyst (5 mg) (entry 17, Table S2) or H2O2 as the oxidation agent (entry 8, Table S2) under the same reaction conditions. The results show that a reaction does not occur, which may be influenced by the number of −SO3H groups in the catalyst.

The effect of reaction conditions on the formation of the desired product was carefully investigated. On the basis of the survey results, the best yield was observed at room temperature with 5 mg of AC-SO3H for 60 min The reusability of AC-SO3H is detailed in Figure .

Figure 1

(a) Reusability of AC-SO3H. (b) FT-IR spectrum of AC-SO3H and AC-SO3H (after three reuses). (c) SEM image of AC-SO3H. (d) SEM image of AC-SO3H (after three reuses).

After investigating the optimal reaction condition, we focused on investigating the effects of benzyl alcohol derivatives. The study focused on the oxidation of benzyl alcohol to benzaldehyde using a photocatalyst. The model reaction was performed in the presence of eosin Y and TBHP under visible-light-promoted radicals for 28 h. The results are summarized in Table . The reaction was then studied for various benzyl alcohols containing electron-withdrawing or electron-donating functional groups or heterocyclic alcohols with ethyl acetoacetate, hydrazine hydrate, and malononitrile under the standard conditions. Electron-donating groups, including CH3 and OCH3, at the para position of the benzene ring provided yields of the corresponding products in lower than that of benzyl alcohol due to the ability for nucleophilic addition of carbonyl carbon in benzaldehyde, which was performed through the oxidation pathway from the alcohol (compounds 1–3, Table ). Next, furfuryl alcohol was employed for the synthesis of 6-amino-4-(furan-2-yl)-3-methyl-2,4-dihydropyrano[2,3-c]pyrazole-5-carbonitrile; the yield of the product was improved by the resonance of the five-membered ring containing oxygen (compound 4, Table ). Moreover, the condensation of alcohols, malononitrile, and ethyl acetoacetate with phenylhydrazine provided the major products in acceptable yields (compounds 5–8, Table ). As shown, in the presence of AC-SO3H/[cholineCl][urea]2, no condensation reaction occurred between 2,4-dinitrophenylhydrazine or acetohydrazide with benzyl alcohol or furfuryl alcohol, with a prolonging of the reaction time, presumably due to the influence of steric hindrance of a −NO2 group at an ortho position and the acetyl group reducing the activity of the acetohydrazide. The reactions were carried out in the presence of ethyl acetoacetate, malononitrile, benzyl alcohol/2-methoxybenzyl alcohol, and 2,4-dinitrophenylhydrazine/5-bromo-2-hydroxyhydrazide, but the desired products were not observed. As can be seen in Table , the reaction occurs with only three components, including 2,4-dinitrophenylhydrazine/5-bromo-2-hydroxyhydrazide, malononitrile, and benzyl alcohol derivatives. under the same reaction conditions (compound 9–11, Table ).

Table 1

Preparation of Dihydropyrano[2,3-c]pyrazoles by the Current Methoda

Conditions: alcohols (1.0 equiv), ethyl acetoacetate (1.0 equiv), hydrazines (1.0 equiv), and malononitrile (1.0 equiv) with eosin Y (0.05 equiv), TBHP (3 equiv), AC-SO3H (5 mg) and [cholineCl][urea]2 (5 equiv), stirring at room temperature (RT). The product was isolated by recrystallization using hot ethanol (10–15 mL).

The plausible mechanism for the preparation of pyrano[2,3-c]pyrazole through the oxidation of alcohol under visible-light-promoted radicals is presented in Scheme . First, the ethyl acetoacetate was activated with phenylhydrazine in the presence of AC-SO3H to form the intermediate A. Next, the oxidation of benzyl alcohol to form benzaldehyde occurred through the participation of visible-light-induced t-BuO• radicals generated from TBHP. The benzaldehyde was then attacked by malononitrile as a nucleophile to provide the intermediate B. The pyrazolone A reacted with intermediate B to form intermediate C by a Michael addition. The intramolecular cyclization reaction of intermediate C gave the intermediate D. Finally, the major product was produced in a tautomerization reaction.

Scheme 2

Plausible Mechanism for Pyrano[2,3-c]pyrazole Synthesis

Molecular Docking

The resulting binding energies of chemicals were collected, compared, and analyzed using Python 3.5.6 and Microsoft Excel. Scheme A shows the binding affinities to the ATP pocket and the lipid pocket of p38 MAPK of eight molecules derived from pyrazole scaffolds: namely, compounds 1–8. The results reflected that one group had high binding affinity to both the ATP pocket and the lipid pocket, compounds 5–8 (having a phenyl group at the 4-nitrogen atom, group I) and another group with a lower binding affinity, compounds 1–4 (no phenyl group at the 4-nitrogen atom, group II). As a result, compounds with high affinity to the ATP site also have high affinity to the lipid site, with compound 5 having the most potential with binding energies of −9.5215 and −10.9932 kcal/mol for ATP and lipid binding sites, respectively. In contrast, compound 4 is barely bound to either pocket, being the most uonreactive of the weak group with binding affinities of −7.7987 and −8.9249 kcal/mol for ATP and lipid binding sites, respectively. Interestingly, all tested compounds have a higher affinity to the allosteric lipid-binding pocket in comparison to the ATP-binding site, and the degree of difference varies. While group II showed low preference for the lipid pocket, with differences in binding strength to the two pockets being only Δ = 1.5555 kcal/mol at most, group I demonstrated a significant contrast between the sites’ affinity, with the greatest difference coming from compound 2, the binding strength of which to the hydrophobic site is −2.2091 kcal/mol stronger than that of the other site. Therefore, group I provided compounds with high selectivity while group II had an overall higher binding capacity.

Scheme 3

Affinity Plots of Pyrazole Compounds and Reference Ligands of the ATP Pocket and lipid Pocket of MAPK p38 Crystal Structures: (A) Binding Affinity of Pyrano[2,3-c]pyrazole; (B) Binding Affinity of Positive Controls

In order to confirm the effectiveness of these compounds for pharmaceutical applications, 38 known ligands of the ATP binding pocket and 26 known ligands of the lipid-binding pocket were extracted from the complex PDB files, and docking experiments were performed with the two reference receptor structures 1A9U and 4DLI, respectively. From Scheme B, it can be inferred that the ATP-pocket ligand performed better than the lipid counterpart, with binding energies ranging from −8.1206 to −14.6960 kcal/mol, whereas the lipid pocket’s reference ligands had energies ranging from −8.3527 to −11.9219 kcal/mol. Using these values as thresholds, it showed that while all pyrano[2,3-c]pyrazole were within the inhibition range for the lipid pocket, only compounds 5–8 satisfied the affinity required for the ATP pocket. While both groups had affinities within the positive range for the lipid pocket, only compounds 5–8 (the diphenyl group) could theoretically bind to the ATP pocket efficiently (ΔG > −8.12 kcal/mol), with the best ligand candidate being compound 5, having two phenyl groups at positions 1 and 4.

Ligand–Receptor Analysis

With the anticipation of finding a potent ligand for the allosteric site of p38 MAPK, it caught our attention that compound 5 has the highest average binding affinity to both the main pocket and the lipid-binding site, and thus we decided to further analyze this ligand. It was then revealed that both pockets facilitated compound 5 complexes with numerous hydrophobic interactions (Figures , and 3). For the ATP-binding site, pyrano[2,3-c]pyrazole derivatives were well within the phosphate binding region created by the main components of the site, including the “Asp-Phe-Gly” (DFG) motif and the E71 residue (Figure A, the red region surrounding the compound), proving its competitiveness with ATP in the DFG-in state.[62,77] For the DFG motif, the compound interacted with the side chain of Phe169 and Asp168, whereas the main chain of Glu71 might be the main contact with the ligand (Figure A). Apart from that, there were at least 10 other residues within the pocket having hydrophobic interactions with 5, mostly at its two phenyl groups (Figure B).

Figure 2

Interaction between 5 and the ATP binding pocket of the MAPK p38a crystal structure (PDB: 1A9U): (A) compound 5 in a complex with MAPK p38a’s ATP-binding pocket; (B) interaction between compound 5 and amino acids at the ATP binding pocket. The red dotted lines illustrate the hydrophobic bonds.

Figure 3

Interaction between 5 and lipid-binding pocket of MAPK p38a crystal structure (4DLI): (A) compound 5 in a complex with MAPK p38a’s lipid-binding pocket; (B) interaction between compound 5 and the lipid-binding pocket. The red dotted lines illustrate hydrophobic interactions.

On the other hand, compound 5 bound inside the hydrophobic pocket when the receptor was in an inactive, DFG-out state for substrate binding.[60] In particular, compound 5 occupies a microenvironment formed by multiple leucine residues (green) while being stabilized by a π-stacking binding with Trp197 (cyan), which could explain the superior binding affinity in comparison to the ATP-binding site (Figure A). Other than that, the phenyl groups on the ligand still attracted the majority of hydrophobic interactions in the pocket, being in contact with 8 out of 11 possible residues (Figure B). Since the microenvironment inside the pocket is mostly hydrophobic, it was justified that an additional phenyl in the nitrogen atoms in the case of group II would enhance the binding capability, aside from the steric effect that came from such a conformation.

Next, we analyzed another attractive potent allosteric ligand, compound 2, to represent group I pyrano[2,3-c]pyrazole derivatives (Figure ). While the compound was positioned in locations in the receptor similar to those for 5 (data not shown), it is unexpected that compound 2 directly linked to the backbone of Phe169 and Leu171 with hydrogen bonds in the ATP-binding site, despite having lower binding affinity to the canonical pocket in comparison to positive controls. The explanation could be that since hydrophobic interactions are the main driving force of ligand–receptor interactions in the ATP-binding pocket, 2 would have a low binding strength, since it can only interact with nine hydrophobic residues. Meanwhile, 2 was stabilized mainly by hydrophobic interactions in the lipid pockets, thus explaining its high allosteric binding capability. It could be concluded that group I’s inability to bind with the ATP-binding pocket was due to the lack of a truly hydrophobic environment like the lipid-binding counterpart.

Figure 4

Interaction of compound 2 with (A) the ATP-binding pocket and (B) the lipid-binding pocket. The red dotted lines illustrate hydrophobic interactions, while the green dotted lines illustrate hydrogen bonds.

Conclusions

We developed a new approach for the preparation of pyrano[2,3-c]pyrazole through a multicomponent reaction. Eight compounds were successfully synthesized and characterized by NMR and HRMS (ESI). These compounds were tested for their binding affinity to the ATP-binding pocket and the lipid-binding pocket using a molecular docking method. The derivatives could be classified into two main groups, one with 6-amino-1,4-phenyl-3-methyl-2,4-dihydropyrano[2,3-c]pyrazole-5-carbonitrile as the base component (compounds 5–8) and the remaining group without a phenyl group at the 4-nitrogen atom. Both groups were proven to have the tendency to bind to the allosteric lipid-binding site, with the two phenyl groups being been proven to have a greater binding affinity to both the ATP-pocket and the lipid-binding pocket, while the single-phenyl group showed higher selectivity. The results were then compared with positive controls consisting of 38 ATP-pocket ligands and 26 lipid-pocket ligands. While all compounds were comparable with positive lipid binding control, only the two phenyl groups had sufficient binding affinity to the ATP pocket, with the most potent overall inhibitor being compound 5, while compound 2 had the highest affinity to only the hydrophobic pocket. Generally, the eight compounds were shown to interact with p38 MAPK mainly by nonpolar interactions and to tend to interact with the allosteric lipid site rather than the canonical ATP pocket and so could be selective inhibitors and avoid unwanted inhibition of other MAPK. In particular, compound 5 could be a type V inhibitor that attacks the receptor in both active and inactive states, whereas compound 2 provided an excellent type IV activity, only binding the allosteric site. However, these results have not considered receptor–ligand bridging with water molecules, as well as suffering from the limitations of virtual screening, hence requiring experimental confirmations of their inhibiting activity to p38a MAPK. Nevertheless, our result should raise expectations for future applications of pyrano[2,3-c]pyrazole as selective inhibitors for p38a MAPK, while giving clues to the directions for further enhancing these derivatives’ inhibiting strength.