Introduction

Humanity is now plagued by the coronavirus disease 2019 (Covid-19) pandemic, which is caused by coronavirus 2 that causes severe acute respiratory syndrome (SARS-CoV-2) [1]. Even though COVID-19 is a viral infection, it can also affect a patient's immune system. Therefore, secondary bacterial or fungal infections can take hold [2, 3]. Researchers worldwide have been attempting to investigate the potential treatment and discovery of effective therapeutic drug candidates for the management of COVID-19. Furthermore, some reports proposed potential inhibitory agents against COVID-19 targets. Currently, several molecules are being tested for their efficacy on COVID-19 disease, some of which have reached clinical trials, while others are still in the preclinical phase [1, 4, 5, 6, 7, 8, 9]. Pyrazoline is a five-membered heterocyclic having two adjacent nitrogen atoms within the ring [10, 11]. Nitrogen-containing heterocycles have a wide range of applications in medicine [12, 13, 14]. Therefore, the chemistry of pyrazoline compounds has attracted the attention of researchers due to their important biological activities such as anti-bacterial, anti-inflammatory, anti-cancer, anti-fungal, anti-oxidant, anti-depressant, anti-leishmanial, anti-convulsant, and anti-tumor properties [15, 16, 17, 18, 19, 20, 21, 22, 23], especially as anti-viral agent [24, 25, 26, 27].

Hence, we designed novel pyrazoline derivatives via one-pot three-component condensation reaction, as this became preferred protocol and was used extensively for synthesizing new heterocyclic compounds due to its simple set-up procedure, reduced reaction time, excellent yields and reduced pollutant production. Computationally (in silico) evaluated them as a potential agent against SARS-CoV-2 main protease (Mpro) with the help of molecular docking simulation. Moreover, the current study is going to investigate the different properties including drug-likeness and pharmacokinetics properties, structural characterization, spectroscopic, thermodynamic, and vibrational phenomena with stabilization energy of all the synthesized compounds. Based on the findings of this investigation, researchers will be able to discover an effective drug to treat COVID-19.

Experimental

Preparation of 1-(4-((4-chlorobenzyl)oxy)phenyl)ethan-1-one (3)

In 250-mL round bottom flask, a mixture of 4-hydroxyacetophenone (25 mmol), 4-chlorobenzylchloride (30 mmol) in the presence of catalytic anhydrous potassium carbonate (50 mmol) was refluxed for 6 h with stirring in absolute ethanol (50 mL). The reaction completion was monitored by changing the color, then poured into cold water. The obtained precipitate was filtered, washed with cold ethanol, dried in an oven, and recrystallized from ethanol to give compound 3. White powder; m.p.: 91.3–92.7 °C; yield; 97.0%; FT-IR (KBr) (υmax/cm−1): 3008 (C–H sp2), 2873 (C–H sp3), 1668 (C=O), 1597 (C=C); UV λmax = 336 nm; 1H NMR (400 MHz, CDCl3) (ppm): δ 2.57 (3H, s, CH3), 5.09 (2H, s, OCH2), 6.95 (2H, d, J = 8.4 Hz, Ar–H), 7.35 (4H, m, Ar–H), 7.90 (2H, d, J = 8.4 Hz, Ar–H); 13C NMR (100 MHz, CDCl3) (ppm): δ 196.70 (C=O), 162.30 (C-4), 134.60 (C-6), 134.00 (C-9), 130.70 (C-7 and C-7′), 130.60 (C-2 and C-2′), 128.80 (C-8 and C-8′), 128.7 (C-3), 114.50 (C-3 and C-3′), 69.30 (C-5), 26.30 (CH3).

One pot synthesis of pyrazoline derivatives (A1-A3)

In a round bottom flask, a mixture of compound 3 (7 mmol), substituted benzaldehydes (7 mmol), and thiosemicarbazide (10 mmol) were dissolved in absolute ethanol (30 mL). To this mixture a catalytic amount of sodium hydroxide (12 mmol) was added and refluxed with stirring. The completion of reaction was monitored by thin layer chromatography (TLC) in hexane/ethyl acetate as a solvent system. The solution was cooled to room temperature, filtered the obtained solid, washed with cold ethanol, and dried. The pure target compounds (A1–A3) were obtained by recrystallization from toluene.

5-(3-(Benzyloxy)phenyl)-3-(4-((4-chlorobenzyl)oxy)phenyl)-4,5-dihydro-1H-pyrazole-1-carbothioamide (A1)

Pale yellow; m.p.: 210.3–211.8 °C; yield: 75.5%; Rf = 0.36 in n-Hex:EtOAc (1:1); FT-IR (KBr) (υmax/cm−1): 3487, 3360 (NH2), 1604 (C=C), 1471 (C=N), 1355 (C=S); UV λmax = 360 nm; 1H NMR (400 MHz, CDCl3) (ppm): δ 3.16 (1H, dd, J = 4.0 and 16.0 Hz, Ha-C11), 3.78 (1H, dd, J = 4.0 and 16.0 Hz, Hb-C11), 4.99 (2H, s, OCH2), 5.09 (2H, s, OCH2), 5.99 (1H, dd, J = 4.0 and 12.0 Hz, H-C12), 6.16 (2H, s, NH2), 6.82–7.77 (4H, m, Ar–H), 6.99 (2H, d, J = 8.0, H-7 and H-7′), 7.25–7.38 (16H, m, Ar–H), 7.65 (2H, d, J = 8.0 Hz, H-8 and H-8′); 13C NMR (100 MHz, CDCl3) (ppm): δ 176.37 (C-24), 160.73 (C-6), 158.94 (C-15), 155.68 (C-10), 143.58 (C-13), 135.37 (C-20), 134.77 (C-4), 133.69 (C-23), 130.04 (C-1), 129.02 (C-3 and C-3′), 128.93 (C-21 and C-21′), 128.86 (C-17), 128.74 (C-22 and C-22′), 128.69 (C-2 and C-2′), 128.21 (C-8 and C-8′), 123.55 (C-9), 118.23 (C-18), 115.16 (C-7 and C-7′), 113.60 (C-16), 112.17 (C-14), 69.32 (C-5), 69.20 (C-19), 63.24 (C-12), 43.07 (C-11).

5-(3-(Benzyloxy)phenyl)-3-(4-((4-chlorobenzyl)oxy)phenyl)-4,5-dihydro-1H-pyrazole-1-carbothioamide (A2)

White powder; m.p.: 203.7–205.4 °C; yield; 73.3%; Rf = 0.43 in n-Hex:EtOAc (1:1); FT-IR (KBr) (υmax/cm−1): 3423, 3236 (NH2), 1591 (C=C), 1456 (C=N), 1382 (C=S); UV λmax = 362 nm; 1H NMR (400 MHz, CDCl3) (ppm): δ 3.14 (1H, s, Ha-C11), 3.77 (1H, s, Hb-C11), 5.08 (4H, s, 2× OCH2), 5.99 (1H, s, HX-C12), 6.84–7.65 (16H, m, Ar–H); 13C NMR (100 MHz, CDCl3) (ppm): δ 176.41 (C-24), 160.70 (C-6), 159.28 (C-15), 155.10 (C-10), 137.82 (C-13), 134.72 (C-20), 129.94 (C-4), 128.98 (C-23), 128.84 (C-1), 128.67 (C-3 and C-3′), 128.50 (C-21 and C-21′), 128.17 (C-17), 127.93 (C-22 and C-22′), 127.63 (C-2 and C-2′), 125.25 (C-8 and C-8′), 123.58 (C-9), 117.97 (C-18), 115.11 (C-7 and C-7′), 113.45 (C-16), 112.24 (C-14), 70.10 (C-5), 69.30 (C-19), 63.25 (C-12), 43.06 (C-11).

3-(4-((4-Chlorobenzyl)oxy)phenyl)-5-(3-((3-nitrobenzyl)oxy)phenyl)-4,5-dihydro-1H-pyrazole-1-carbothioamide (A3)

Pale brown; m.p.: 147.4–148.7 °C; yield: 64.6%; Rf = 0.32 in n-Hex:EtOAc (1:1); FT-IR (KBr) (υmax/cm−1): 3471, 3354 (NH2), 1597 (C=C), 1471 (C=N), 1346 (C=S); UV λmax = 294 nm; 1H NMR (400 MHz, CDCl3) (ppm): δ 3.15 (1H, d, J = 16.0 Hz, Ha-C11), 3.72 (1H, dd, J = 16.0 Hz, Hb-C11), 3.75 (1H, s, –CH-HX-C12), 5.07 (2H, s, –OCH2), 5.08 (2H, s, –OCH2), 6.38 (2H, s, NH2), 5.99 (1H, d, J = 12.0 Hz, H-14), 6.85–7.73 (13H, m, Ar–H), 8.99 (1H, d, J = 8.0 Hz, H-23), 8.11 (1H, s, H-21); 13C NMR (100 MHz, CDCl3) (ppm): δ 179.27 (C-26), 160.74 (C-6), 158.56 (C-15), 155.73 (C-10), 148.29 (C-22), 143.83 (C-20), 139.08 (C-13), 134.82 (C-4), 133.95 (C-25), 133.32 (C-1), 130.20 (C-24), 129.54 (C-3 and C-3′), 129.05 (C-17), 128.23 (C-2 and C-2′), 125.31 (C-9), 123.47 (C-8 and C-8′), 122.83 (C-21), 118.53 (C-23), 115.15 (C7 and C-7′), 114.69 (C-18), 113.62 (C-16), 112.28 (C-14), 69.29 (C-5), 68.61 (C-19), 63.21 (C-12), 43.03 (C-11).

Molecular docking study

Ligand preparation

The 3D structures of titled compounds were drawn, and energy of each compound was minimized and saved in the pdb format by using Avogadro [28]. Subsequently, AutoDock Tools 4.2 [29] was used to open the pdb files, and torque adjustment was made by detecting roots and adjusting as desired. Then, the files were saved in the pdbqt format. Meantime, the structure data format (SDF) structure of the control, lopinavir (CID: 92727), was downloaded from the PubChem database (www.pubchem.ncbi.nlm.nih.gov). The structure data file (SDF) format was subjected to the energy minimization process using MMFF94 (Merck Molecular Force Field 94) force field algorithm [30], the minimized structure was converted into PDBQT format using PyRx before performing molecular docking analysis.

Protein preparation

The 3D crystal structure of SARS-CoV-2 (Mpro) was retrieved from the Protein Databank (http://www.rcsb.org) with PDB ID (6LU7). The co-crystallized ligand and water molecules were removed utilizing Discovery Studio Visualizer [31]. Afterward, polar hydrogens and Kollman charges were added. Partial charges of the molecule were calculated using the Gasteiger method. The prepared file was saved as a pdbqt format for molecular docking using AutoDock Tools 4.2 [29].

Molecular docking

Molecular docking of the ligands was investigated by using the AutoDock Vina program [32]. Vina software was run in exhaustiveness = 8. The binding site was selected based on its co-crystallized ligand (N3) with the target protein [33]. Affinity scores (in kcal/mol) given by AutoDock Vina for all the compounds were obtained and ranked based on the free energy binding theory (more negative value means greater binding affinity). The resulting docking of ligand-receptor complexes with higher binding affinity was analyzed and drafted to 2D and 3D figures with the help of the DS Visualizer and UCSF Chimera.

ADMET analysis

According to the reported procedures, as described in the literature [34], prediction of drug-likeness and ADMET (Absorption, Distribution, Metabolism, Excretion, and Toxicity) properties of the synthesized compounds was carried out utilizing online servers such as SwissADME, ProTox-II, and pkCSM.

Density functional theory

In order to understand the spectral assignments, and to study the molecular geometry and electronic transitions of the synthesized compounds (A1, A2, and A3), a computational study was taken into account, to explore the molecular interactions in more detail. The Gaussian 09RevD.01 program [35] was used for theoretical studies. Gradient corrected correlation with Pople's basis set B3LYP/6-311G++ was applied. The gauge independent atomic orbital (GIAO) and Integral Equation Formalism Polarizable Continuum Model (IEFPCM) with the same functional set were used for computing 1H-NMR chemical shifts in the liquid phase [36, 37, 38, 39, 40, 41, 42]. Optimized structure of A1, A2, and A3 was also obtained using the B3LYP/6-311G++ basis set. Highest occupied molecular orbital (HOMO), lowest unoccupied molecular orbital (LUMO), and electrostatic potential map (MEP) were calculated. The calculated IR frequencies through this method are found to be positive, indicating that the optimized structure is at a minimum on the potential energy surface. In the FTIR spectrum, bands that appeared were assigned with full accuracy, using animated modes of vibrations. The frontier molecular orbitals help in determining the chemical stability of the system. Moreover, structure-based molecular properties like atomic charges, total energy, stabilization energy, electronic properties, bond lengths, frontier molecular orbitals, and molecular electrostatic potential were calculated by this theory in the gas phase. For visualization of obtained DFT results, ChemCraft 1.5 software [43] was used [43].

Results and discussion

Chemistry

1-(4-((4-Chlorobenzyl)oxy)phenyl)ethan-1-one (3) was prepared in excellent yield (97%) based on of Williamson synthesis of ether by direct benzylation of para-hydroxy acetophenone with 4-chlorobenzylchloride, in the presence of anhydrous potassium carbonate in ethanol. The synthesis of pyrazoline derivatives (A1–A3) was accomplished via the condensation reaction of compound 3, substituted benzaldehydes with thiosemicarbazide in dilute ethanolic sodium hydroxide solution (Scheme 1). The structures of synthesized pyrazolines (A1–A3) were confirmed by FT-IR, 1H-NMR, 13C-NMR, and UV–Vis spectroscopy. In FT-IR spectra of pyrazolines the disappearance of a band in the range of 1653–1645 cm−1 for carbonyl groups (C=O), were considered as good evidence to produce pyrazoline, all spectra exhibited a characteristic two bands at 3400–3200 cm−1 corresponding to the (NH2) stretching of carbothioamide group attached to pyrazoline ring, also the appearance of the strong band at 1492–1456 cm−1 for C=N stretching vibration, which indicates the presence of pyrazoline ring, two bands at 1608–1597 cm−1 and 1382–1340 cm−1 referring to (C=C) and (C=S) double bond, respectively. The 1H-NMR spectra of pyrazolines display doublet to doublet (dd) signals corresponding to three protons (two germinal and one vicinal) of C11 and C12 of pyrazoline ring; they form a characteristic ABX spin system explaining the inequivalence of protons at C11.

Scheme 1
scheme 1

synthesis of compounds (A1–A3)

Full size image

Molecular docking study

Molecular docking was investigated by AutoDock Vina software due to its excellent speed and providing low root-mean-square deviation (RMSD) [32]. The binding affinity of titled compounds was compared with lopinavir which was used as a positive drug for comparison purposes which was recommended for clinical therapeutic [44]. The binding affinities of the synthesized compounds A1 and A2 (− 8.2and − 8.3 kcal/mol, respectively) were higher than the binding affinity of the positive drug lopinavir (− 7.8 kcal/mol). Thus, these two compounds could be effectively inhibiting the SARS-CoV-2 main protease (Mpro). The compound A3 has a slightly lower binding affinity (− 7.5 kcal/mol) than the lopinavir. The docking results exhibited no significant differences in the binding affinities of compounds A1 (with chloro substituent) and A2 (without substituent), as shown in Table 1.

Table 1 Molecular docking scores and interaction modes between compounds and SARS-CoV-2 (PDB ID: 6LU7)
Full size table

The A2-6LU7 complex established three hydrogen bonds with amino acid residues ASN238, LEU272, and LEU287. An alkyl interaction with LEU272 residue and π-alkyl interactions with TYR237 and LEU286 were also observed. Furthermore, it can form π-sigma interaction with LEU287. In contrast, compound A1 showed additional hydrophobic and electrostatic interactions with ASP197 residue of target receptor via π-sigma and π-anion bonds. Moreover, compound A3 (with nitro substituent) interacted with Mpro, formed five hydrogen bonds at residues THR199, ASN238, LEU271, ALA285, and LEU287. An alkyl and π-alkyl hydrophobic interactions were observed between this compound and the receptor. Additionally, amino acid residues ASP107 and LYS137 are involved in forming π-anion and π-cation electrostatic interactions, respectively. The docking interactions of synthesized compounds with the active site amino acid residues of SARS-CoV-2 (Mpro) are depicted in Fig. 1. The docking scores interaction modes of all docked ligands into the active site and the amino acid residues of Mpro are presented in Table 1.

Fig. 1
figure 1

Three-dimensional and two-dimensional diagrams illustrating the various interaction modes of (A) A1, (B) A2, and (C) A3 within active site of SARS-CoV-2 (6LU7)

Full size image

Drug-likeness and ADMET prediction

The pharmacokinetic properties and toxicity profile of the pyrazolines (A1–A3) were predicted in-silico, as presented in Table 2. The obtained results are in agreement with parameters reported in the literature [45], as well as referring to Lipinski's rule of five (LRO5) [46]. The first parameter observed was molecular weight (Mol. Wt.), where all compounds met the LRO5 criteria, which were ≤ 500 g/mol. However, the Mol. Wt. of all ligands still met the criteria set by Chander et al. [45]. Then, all compounds showed a TPA value of less than 140 A2, thus fulfilling the criteria in LRO5 [46], even though the compound A3 showed a value that was close to the threshold. The number of hydrogen donor and acceptor groups of all compounds meets the LRO5 criteria, with no more than 5 and 10 groups, respectively. All compounds did not meet the criteria for Log P based on LRO5 [46], especially compound A1, whose Log P values ​​were huge (> 7.5). The high Log P value can affect the absorption and distribution of the compound, so it is necessary to consider ways to handle it [47]. Only compound A3 showed optimal Log S values, although all of them had good CaCO2. The compound A3 again is the only one that shows a relatively low Log BB value outside the optimal range, and also shows the highest amount of Rot, even though it is still in the optimal range.

Table 2 ADMET properties of synthesized compounds
Full size table

Finally, all compounds are predicted to have moderate toxicity (harmful if swallowed) with an LD50 of 1000 mg/kg and are in class IV of the globally harmonized system of classification of labeling of chemicals [48]. However, compound A3 was predicted to be a mutagen with a positive Ames test result. Further analysis of the target toxicity of each compound with ProTox-II is presented in Fig. 2. Among synthesized compounds, only A3 showed the probability of toxicity to several targets such as carcinogenicity (carcino) and mutagenicity (mutagen), although with a low probability (< 0.7) [48]. Overall, compound A1 shows absorption and distribution deficiencies, while A3 compounds tend to be more toxic. Thus, compound A2 has the most ideal ADMET properties compared to the two.

Fig. 2
figure 2

Prediction of toxicity parameters with ProTox-II

Full size image

DFT studies

The total SCF energy of the synthesized compounds A1, A2, and A3 after 8, 16, and 40 optimization steps was obtained to be − 2791.37, − 2331.77, and − 2387.12 a.u, respectively. The stabilization energy for A1, A2, and A3 was calculated to be − 288.81 × 103, − 407.88, and − 351.40 × 102 kcal/mol, respectively, which endorsed the stability of the compounds. The optimization of all compounds (A1, A2, and A3) was carried out through the B3LYP/6-311G++ calculations. Figure 3 shows the optimized structures of A1, A2, and A3 having ground state minimum energy, strain-free lattice constants, and atomic coordinates. From the optimized structures, the bond lengths were also obtained, which are represented in Fig. 4 and Table 3. The obtained Mulliken charges of the titled compounds were compared, which showed the shift in Mulliken charges, indicating the compound formation has different structural values. Figure 3 and Table 4 represent the Mulliken atomic numbers scheme and atomic charges, respectively.

Fig. 3
figure 3

Optimized structure of compounds (a) A1, (b) A2, and (c) A3 with Mulliken atom numbering scheme

Full size image
Fig. 4
figure 4

Optimized structure of compounds (a) A1, (b) A2, and (c) A3, showing bond lengths

Full size image
Table 3 The bond lengths of the synthesized compounds obtained through DFT
Full size table
Table 4  Mulliken atomic charges on the atoms of the optimized structure of the titled compounds 
Full size table

The molecular electrostatic potential (MEP) map (Fig. 5) represents the electrostatic potentials strength of reactants and synthesized moieties. The red color represents the electronegative area, and the blue color is for the electropositive region. The interaction between the reactants can be studied by MEP, as the electron-deficient region lies at C and H of reactant and electron-rich at O atom. These results show the preferential binding sites over the molecule for the electrophilic and nucleophilic attacks [49, 50]. MEP map of the designed compounds is plotted from deep red to deep blue color scale as − 5.159 e−2 to + 5.159 e−2, − 4.970 e−2 to + 4.970 e−2, and − 5.148 e−2 to + 5.148 e−2 for A1, A2, and A3, respectively, as shown in Fig. 5. B3LYP/6-311G++ basis set was also used to investigate the IR spectrum theoretically. The comparison of experimental FTIR spectra of the synthesized compounds with the simulated spectra for A1, A2, and A3 scaled at 0.9785, 0.9200, and 0.9190, respectively, are shown in Figs. 6, 7 and 8. The computed IR spectrum in animated modes was found to be compatible with few significant experimental vibrational signals in the FTIR spectra of the compounds [51]. The slight difference between theoretical and experimental frequencies is due to theoretical values being acquired in gas phase and experimental values being obtained in solid phase and also due to neglecting the incompleteness and anharmonicity of the basis set. Therefore, to reduce the differences, the scaling factor is used. The nature of the electronic transitions has been explored in the gas phase through TD-DFT method as shown in Fig. 9. Two absorption bands at 375 and 450 nm can be seen through TD-DFT calculations for A1. The simulated absorption bands at 450 and 375 nm are assigned to HOMO to LUMO and HOMO to LUMO + 1. Similar results were obtained for A2 and A3 from TD-DFT calculations.

Fig. 5
figure 5

MEP surface map of (a), reactant-a (b), reactant-b (c) A1, (d) A2, and (e) A3 with respective color scales

Full size image
Fig. 6
figure 6

FTIR spectra of A1, experimental (green) and computational (black)

Full size image
Fig. 7
figure 7

FTIR spectra of A2, experimental (green) and computational (black)

Full size image
Fig. 8
figure 8

FTIR spectra of A3, experimental (green) and computational (black)

Full size image
Fig. 9
figure 9

UV–visible spectra of (a) A1, (b) A2, and (c) A3 obtained through TD-DFT calculations

Full size image

The 1H-NMR spectra for all the compounds were also obtained through GIAO calculation, as shown in Figs. 10, 11 and 12 [51]. Frontier molecular orbitals (FMOs) are significant factors for explaining chemical behavior; hence, they were also computed theoretically. LUMO are generally electron acceptors, while HOMO are electron donors. The chemical stability of molecules is related to the energy gap between the HOMO and LUMO. According to the literature, compounds with a small energy gap have low kinetic stability, more chemical reactivity, and are soft in nature, whereas compounds with a large energy gap have high kinetic stability, less chemical reactivity, and are hard in nature [52]. Figure 13 represents the spatial arrangements of HOMO and LUMO along with their gap and associated energies. The HOMO–LUMO gap (∆E) for compounds A1, A2, and A3 were calculated to be 3.216, 3.156, and 2.622 eV, respectively [53]. On the basis of ∆E, the chemical reactivity of the compounds is arranged in the following sequence—A1 > A2 > A3. Certain molecular properties in the gas phase related to chemical reactivity for the compounds are shown in Table 5 based on the optimized structure and HOMO–LUMO.

Fig. 10.
figure 10

1H NMR spectrum for A1 calculated through DFT (GIAO) calculations

Full size image
Fig. 11.
figure 11

1H NMR spectrum of A2 calculated through DFT (GIAO) calculations

Full size image
Fig. 12.
figure 12

1H NMR spectrum of A3 calculated through DFT (GIAO) calculations

Full size image
Fig. 13
figure 13

Spatial plot of HOMO and LUMO with their energy gap

Full size image
Table 5 Various other theoretical molecular parameters of the compounds and their constituents
Full size table

Conclusion

We have successfully synthesized three new pyrazoline analogues and evaluated their in silico inhibitory activity against SARS-CoV-2 Mpro. The results exhibited that all of them have potential inhibitory. Compound A2 (− 8.3 kcal/mol) was found to be the most potent SARS-CoV-2 Mpro inhibitor among the docked compounds. Furthermore, compound A2 also showed more ideal ADMET properties compared to the other designed compounds. The optimized structures alone with the molecular properties were obtained for the synthesized compounds through theoretical DFT calculations, which supports the experimental data. Various other theoretical molecular parameters calculated through DFT and in vitro evaluation against SARS-CoV-2 Mpro will help researchers for future investigations.