A Simple and Efficient Approach to the Synthesis of 4-Aryl-2-dialkylphosphonomethyl-4-oxobutanenitrile

In this work, we describe a simple and easy synthetic approach to variously 4-aryl-2alkylphosphonomethyl-4-oxobutanenitrile based on the reaction of aromatic aldehydes with phosphorylated Michael’s acceptors in good yields. A general mechanism for the reactions was also proposed. Characterization of the products was carried out by several spectroscopic tools, including Infrared and Nuclear Magnetic Resonance Spectroscopies (1H, 13C, and 31P-NMR). Molecular docking studies were conducted on the synthesized materials against (1UK4) the crystal structure of the SARS Coronavirus Main Proteinase (3CLpro) to study the antiviral activity of these compounds and against (1E3K) the Human Progesterone Receptor to study the anticancer activity of these compounds. We found that compound (5i) was the best one in both antiviral and anticancer activity (according to the binding energy values).


Introduction
For a long time, compounds with both a carbonyl and a nitrile function piqued the interest of several researchers due to their potential utility as reagents in organic synthesis [1]. Many of these compounds have several biological activities [2], and numerous methods of synthesis are described in the literature [3,4]. The starting substrate used is most often acrylonitrile.
To our knowledge, 4-aryl-2-alkylphosphonomethyl-4-oxobutanenitrile is not described in the literature. We were inspired by these works, particularly those of Stetter and others [5,6]. At the start of this year, the world was shocked when severe acute respiratory syndrome (SARS) was discovered in the Chinese province of Hubei and spread rapidly. There are more than 200 districts in China nowadays. Around the world, there are a variety of countries and territories [7]. The genome shares a lot of similarities with the SARS-Cove genome. The outbreak began in early 2003 and lasted into the summer of that year [8]. The majority of the Coronaviridae genome encodes two polypeptides [9] pp1a and pp1ab, which are translated via ribosomal frame shifting. The two proteases, 3CLpro (3C-like protease) and 3Cpro (3C protease), break these polypeptides and turn them into mature nonstructural proteins (NSPs). The open reading frame 1 encodes PLpro (Papain Like Protease) [10]. The structure of 3CLpro from SARS-CoV-2 (PDB code 6LU7) differs from The condensation of one equivalent of the aromatic aldehyde and one equivalent of 2-dialkylaminomethyl-propenenitrile 3 in DMF in the presence of a catalytic amount of 3ethyl-5-(2-hydroxyethyl)-4-methylthiazolium bromide and triethylamine leads to product 5 in a good yield, and the product obtained is 4-aryl-2-dialkylaminomethyl-4-oxobutanenitrile (Scheme 2). The structure of isolable products was characterized by spectral (IR, 1 H -NMR, MS) and elemental analysis data (see experimental). (Supplementary file S9-S26).
The IR spectra produced on some 4-aryl-2-methylamino-4-oxobutanenitrile 5 have a nitrile band at around 2247 cm −1 , characteristic of the CN group, and a band around 1685 cm −1 , characteristic of the carbonyl group. The absence of the absorption band of the nitrile bond on the IR spectra allowed us to follow the evolution of the formation reaction of the product 5. The analysis of proton and 13 C-NMR spectra confirms the proposed structures. The addition of aldehyde to the double bond of alkylphosphonopropenenitriles 3 results in 1 H-NMR by the disappearance of the ethylene protons to 5.5-6.0 ppm and the appearance of a signal at 3.2 ppm attributable to the protons of the CH2-CO motif. Examination of the 13 C-NMR spectra of 4-aryl-2-methylamino-4-oxobutanenitrile 5 shows the disappearance of signals of the two ethylenic carbons of 2-phosphonomethylpropenenitrile, which resonate at 118 and 121 ppm, and the appearance of a signal at about 198 ppm relative to the carbon of the C = O unit. The 1 H, 13 C, and 31 P-NMR attributions of the 4aryl-2-methylamino-4-oxobutanenitrile compounds were carried out by referring to the Scheme 1. Synthesis of 2-dialkylphosphonomethylpropenenitrile 3.
The condensation of one equivalent of the aromatic aldehyde and one equivalent of 2-dialkylaminomethyl-propenenitrile 3 in DMF in the presence of a catalytic amount of 3-ethyl-5-(2-hydroxyethyl)-4-methylthiazolium bromide and triethylamine leads to product 5 in a good yield, and the product obtained is 4-aryl-2-dialkylaminomethyl-4oxobutanenitrile (Scheme 2). The structure of isolable products was characterized by spectral (IR, 1 H-NMR, MS) and elemental analysis data (see experimental). (Figures S9-S26). The IR spectra produced on some 4-aryl-2-methylamino-4-oxobutanenitrile 5 have a nitrile band at around 2247 cm −1 , characteristic of the CN group, and a band around 1685 cm −1 , characteristic of the carbonyl group. The absence of the absorption band of the nitrile bond on the IR spectra allowed us to follow the evolution of the formation reaction of the product 5. The analysis of proton and 13 C-NMR spectra confirms the proposed structures. The addition of aldehyde to the double bond of alkylphosphonopropenenitriles 3 results in 1 H-NMR by the disappearance of the ethylene protons to 5.5-6.0 ppm and the appearance of a signal at 3.2 ppm attributable to the protons of the CH 2 -CO motif. Examination of the 13 C-NMR spectra of 4-aryl-2-methylamino-4-oxobutanenitrile 5 shows the disappearance of signals of the two ethylenic carbons of 2-phosphonomethylpropenenitrile, which resonate at 118 and 121 ppm, and the appearance of a signal at about 198 ppm relative to the carbon of the C=O unit. The 1 H, 13 C, and 31 P-NMR attributions of the 4-aryl-2-methylamino-4oxobutanenitrile compounds were carried out by referring to the bibliographic data. The proton, 31 P, and 13 C-NMR data of the compounds 5 are recorded in the experimental part. bibliographic data. The proton, 31 P, and 13 C-NMR data of the compounds 5 are recorded in the experimental part. Scheme 2. Synthesis of 4-aryl-2-dialkylphosphonomethyl-4-oxobutanenitrile 5.
The first step of this reaction mechanism (Scheme 3) goes through the formation of catalyst II by deprotonation of the precatalyst, the thiazolium salt I. This mechanism is identical to that which was proposed by Breslow for the benzoin reaction [26]. Once the catalyst is formed, it can be added to the aldehyde to form the intermediary III, which, by proton exchange, will generate the compound IV, known as the Breslow intermediate. This is the last one who will be able to add 1.4 to Michael's acceptor to form the new intermediate V, which, after proton exchange, forms compound VI. The removal of catalyst II releases the product of the Stetter reaction 5, and a new catalytic cycle begins. There may be some competition between the additions of 1, 2 and 1, 4 of the Breslow Scheme 2. Synthesis of 4-aryl-2-dialkylphosphonomethyl-4-oxobutanenitrile 5.
The first step of this reaction mechanism (Scheme 3) goes through the formation of catalyst II by deprotonation of the precatalyst, the thiazolium salt I. This mechanism is identical to that which was proposed by Breslow for the benzoin reaction [26].
Molbank 2022, 2022, x FOR PEER REVIEW 3 of 12 bibliographic data. The proton, 31 P, and 13 C-NMR data of the compounds 5 are recorded in the experimental part.
The first step of this reaction mechanism (Scheme 3) goes through the formation of catalyst II by deprotonation of the precatalyst, the thiazolium salt I. This mechanism is identical to that which was proposed by Breslow for the benzoin reaction [26]. Once the catalyst is formed, it can be added to the aldehyde to form the intermediary III, which, by proton exchange, will generate the compound IV, known as the Breslow intermediate. This is the last one who will be able to add 1.4 to Michael's acceptor to form the new intermediate V, which, after proton exchange, forms compound VI. The removal of catalyst II releases the product of the Stetter reaction 5, and a new catalytic cycle begins. There may be some competition between the additions of 1, 2 and 1, 4 of the Breslow Scheme 3. A general mechanism for the reactions.
Once the catalyst is formed, it can be added to the aldehyde to form the intermediary III, which, by proton exchange, will generate the compound IV, known as the Breslow intermediate. This is the last one who will be able to add 1.4 to Michael's acceptor to form the new intermediate V, which, after proton exchange, forms compound VI. The removal of catalyst II releases the product of the Stetter reaction 5, and a new catalytic cycle begins. There may be some competition between the additions of 1, 2 and 1, 4 of the Breslow intermediate IV. Indeed, it can be added to Michael's acceptor to generate product V, or react with a second equivalent of aldehyde to give the product benzoin. However, the formation of the product benzoin is reversible, which ultimately favors the formation of product V.

Molecular Docking Study
For each synthesized compound, the docking simulation process was completed, and the best conformation was chosen as the compound with the highest negative binding energy value. Figures S1-S4 illustrate the 2D and 3D structures of the ligand-receptor interactions of the synthesized compounds with (1UK4). Figures S5-S8 illustrate the 2D and 3D structures of the ligand-receptor interactions of the synthesized compounds with (1E3K). Table 1 displays the estimated binding energies produced by docking the synthesized materials with 1UK4 and 1E3K. All the compounds studied formed stable complexes with receptors that had a high binding energy. According to our findings, compound 5i had the best docking energy (highest binding energy), with a binding affinity of −8.65 kcal/mol with 1UK4 and −10.41 kcal/mol with 1E3K). As a result, the compounds studied, particularly compound 5i, have the potential to be used in antiviral and anticancer applications. According to molecular docking, the most interacting residues of 1UK4 in the 5i compound active site were HIS 161, CYS 143, SER 142, and MET 47. But in 1E3K, it interacted with CYS 209, TYR 208, PHE 96, MET 77, GLN 43, LEU 39, and LEU 38.

Chemistry
Solvents and reagents were obtained from commercial sources and were dried and purified when necessary using standard techniques. IR spectra of the compound 5 derivatives were made in chloroform on a Perkin-Elmer Paragon 1000 PC spectrometer. The wave numbers were expressed in cm −1 . The 1 H, 31 P, and 13 C-NMR spectra were recorded in solution in CDCl 3 on a Brucker AC 300 using TMS and H 3 PO 4 as an internal reference. The chemical shifts were expressed in ppm. The multiplicity of signals was indicated by the following abbreviations: s: singlet, d: doublet, t: triplet, m: multiplet, md: doubled multiplet. Coupling constants were reported in Hertz (Hz). Purification of the products was carried out by chromatography on silica gel using a mixture of ether and petroleum ether as eluent, the proportion of which was 20/80. Chromatography on a thin layer of silica gel 0.2 mm thick with a fluorescent indicator at 254 nm using ether and petroleum ether in a 20/80 ratio as eluent was used to monitor the progress of the reaction. To the quaternary ammonium obtained previously, 20 mmol of alkoxyphosphite and 20 mL of anhydrous benzene were added. The mixture was heated under reflux for two hours. After cooling, the residual ammonium salt was filtered off, the benzene was evaporated off, and the product obtained was either distilled under reduced pressure or purified by column chromatography using ethyl acetate as an eluent.
Dimethyl We describe, for example, the preparation of compound 5a. In a 500 mL three-necked flask equipped with a refrigerant fitted with a drying tube and a dropping funnel mounted with a tube of nitrogen, a solution of 3-ethyl-5-(2-hydroxyethyl)-4-methylthiazolium bromide 2.52 g (0.01 mol) in 150 mL of dry N,N-dimethylformamide was introduced after thorough purging with dry nitrogen. The flask was immersed in a water bath maintained at 60 • C and 6.06 g of triethylamine (0.06mol) was added rapidly from the dropping funnel. After 30 min of agitation, 10.6 g of benzaldehyde (0.1mol) was added dropwise to this solution over a period of 30 min. After stirring for one hour, 2-dialkylaminopropenenitrile (3a), freshly distilled at 17.5 g (0.1 mol), was added over a period of one hour. The solution became more and more viscous. After 12 h of stirring, 30 mL of acetic acid (1 M) was added and stirring continued for another 5 min. The solvent was removed with a rotary evaporator, and the residue was dissolved in 100 mL of water. The solution was extracted several times with chloroform (4 × 100 mL). After drying with anhydrous magnesium sulfate, the solvent was evaporated under reduced pressure and the residual liquid was purified on a column of silica gel (ether, petroleum ether 20/80).

Protein and Ligand Preparation
The progesterone receptor protein structure was received in PDB format from RCSB PDB (ID: 1E3K) and the structure of 3CLpro from SARS-CoV19 (ID: 1UK4), and they were prepared using Discovery Studio 2019 to remove ligands and water molecules [27]. ChemDraw3D Ultra provided the SDF format for the ligand structures.

Molecular Docking
Polar hydrogens and Kollman charges were applied to the protein, and a pdbqt format file was generated using the AutoDockTools 1.5.6 program. The protein was created using the protein preparation wizard in AutoDockTools 1.5.6. Polar hydrogens and Kollman charges were applied to the protein, and a pdbqt format file was generated using AutoDockTools. Both 1UK4 and 1E3K were completely devoid of water molecules. The ligand torsions were calculated by first detecting the roots in AutoDockTools 1.5.6, and then setting the aromaticity parameters to 7.5. The receptor was given a grid size of 60 Å × 60 Å × 60 Å, and the molecular docking operation was assigned to the Lamarckian genetic algorithm (LGA). After docking, the best pose was chosen based on binding energy, ligand-receptor interactions, and active site residues. The docked posture was simply compared to the cocrystallized structure, and the root mean square deviation (RMSD) was less than 1.0 Å. All torsions were allowed to rotate during docking. The traditional docking procedure for rigid and fluid ligand docking included 10 separate runs per ligand, 2.5 × 10 6 energy measurements, a total of 27,000 iterations, a mutation rate of 0.02, a crossover rate of 0.80, and an elitism value of 1. The likelihood of conducting a local search on a person in the population was 0.06 using a limit of 300 iterations per local search. Following docking, the 10 solutions were classified as having RMS differences of less than 1.0. The clusters were sorted based on the cluster's lowest energy representation. The effects of the docking process were visualized using the BIOVIA Discovery Studio program.

Conclusions
A molecular docking study was used to determine the binding energy for nonbonding interactions between the ligand (synthesized compounds) and receptors (1UK4 and 1E3K). All the compounds studied formed stable complexes with receptors that had a high binding energy. Compound 5i had the best docking energy (highest binding energy) according to our findings, with a binding affinity of (−8.65 kcal/mol with 1UK4, and −10.41 kcal/mol with 1E3K), as in Table 1.
Supplementary Materials: The following supporting information can be downloaded, Mol file of 4-Aryl-2-dialkylphosphonomethyl-4-oxobutanenitrile; Figure S1: molecular docked model of compounds 3a-5a with (1uk4) : (the target is presented as thin sticks; the ligands are drawn as ball-and-

Protein and Ligand Preparation
The progesterone receptor protein structure was received in PDB format from RCSB PDB (ID: 1E3K) and the structure of 3CLpro from SARS-CoV19 (ID: 1UK4), and they were prepared using Discovery Studio 2019 to remove ligands and water molecules [27]. ChemDraw3D Ultra provided the SDF format for the ligand structures.

Molecular Docking
Polar hydrogens and Kollman charges were applied to the protein, and a pdbqt format file was generated using the AutoDockTools 1.5.6 program. The protein was created using the protein preparation wizard in AutoDockTools 1.5.6. Polar hydrogens and Kollman charges were applied to the protein, and a pdbqt format file was generated using AutoDockTools. Both 1UK4 and 1E3K were completely devoid of water molecules. The ligand torsions were calculated by first detecting the roots in AutoDockTools 1.5.6, and then setting the aromaticity parameters to 7.5. The receptor was given a grid size of 60 Å × 60 Å × 60 Å, and the molecular docking operation was assigned to the Lamarckian genetic algorithm (LGA). After docking, the best pose was chosen based on binding energy, ligand-receptor interactions, and active site residues. The docked posture was simply compared to the cocrystallized structure, and the root mean square deviation (RMSD) was less than 1.0 Å. All torsions were allowed to rotate during docking. The traditional docking procedure for rigid and fluid ligand docking included 10 separate runs per ligand, 2.5 × 10 6 energy measurements, a total of 27,000 iterations, a mutation rate of 0.02, a crossover rate of 0.80, and an elitism value of 1. The likelihood of conducting a local search on a person in the population was 0.06 using a limit of 300 iterations per local search. Following docking, the 10 solutions were classified as having RMS differences of less than 1.0. The clusters were sorted based on the cluster's lowest energy representation. The effects of the docking process were visualized using the BIOVIA Discovery Studio program.

Conclusions
A molecular docking study was used to determine the binding energy for nonbonding interactions between the ligand (synthesized compounds) and receptors (1UK4 and 1E3K). All the compounds studied formed stable complexes with receptors that had a high binding energy. Compound 5i had the best docking energy (highest binding energy) according to our findings, with a binding affinity of (−8.65 kcal/mol with 1UK4, and −10.41 kcal/mol with 1E3K), as in Table 1.
Supplementary Materials: The following supporting information can be downloaded, Mol file of 4-Aryl-2-dialkylphosphonomethyl-4-oxobutanenitrile; Figure S1: molecular docked model of compounds 3a-5a with (1uk4): (the target is presented as thin sticks; the ligands are drawn as ball-and-stick) where (a-d) represent the 3D docking styles for 1uk4 with compounds 3a, 3b, 3e and 5a, respectively, and (e-h) represent the 2D docking styles for 1uk4 with compounds 3a, 3b, 3e and 5a, respectively; Figure S2: molecular docked model of compounds 5b-5e with (1uk4): (the target is presented as thin sticks; the ligands are drawn as ball-and-stick) where (i-l) represent the 3D docking styles for 1uk4 with compounds 5b, 5c, 5d and 5e, respectively, and (m-p) represent the 2D docking styles for 1uk4 with compounds 5b, 5c, 5d and 5e, respectively; Figure S3: molecular docked model of compounds 5f-5i with (1uk4): (the target is presented as thin sticks; the ligands are

Data Availability Statement:
The data presented in this study are available on request from the corresponding author.