INTRODUCTION

Several reports appeared on the oxidation of propargylic alcohols to produce carbonyl products, as they are of importance as starting materials for a range of organic compounds. For instance, aromatic heterocyclecompounds and other materials [1–8]. Gorden and co-workers have used copper(II) complex of 2-quinoxalinol salen (I) for the selective oxidation of propargylic alcohols to give up to 99% carbonyl compounds for a range of propargylic alcohols in a reasonable time [8, 9]. The same copper(II) complex of 2‑quinoxalinol salen (I) was used in oxidative Mannich reactions for a variety of open-chain and cyclic tertiary amines to yield up to 98% of the products [10] and more important uses [11–16]. A similar Cu(salen) complex (II) was applied in the epoxidation of styrene by tBuOOH [17]. Copper(II) surfactant (III) was of great interest in current rectification [18]. The presence of two NMe2 substituents on the aromatic ring in (IV) facilitates a stable oxidized product formation.

Interestingly, they induce a shift in the oxidation site from the metal to the ligand [19]. This wide range of exciting applications of this class of compounds motivated us to look at the unsymmetrical example, 1c, shown in Fig. 1. This paper reports the synthesis, and characterization via single-crystal X-ray diffraction, DFT studies, and the tetra-coordinate τ4 index of 1c. At the same time, coronavirus was becoming the third most life-threatening disease of animal origin and appeared as one of the severe pandemics and fatal diseases in human history [20–31].

Fig. 1.
figure 1

Tetradentate salen (N2O2 system) copper(II) complexes.

Full size image

Hence, in this study, we aim to contribute in searching for promising drug candidates against COVID-19; thus, we examined our compounds using the free-energy in silico molecular docking method as a preliminary study [32]. Molecular docking is a practical approach for structure-based medicine discovery and a standard procedure for conformational selection, which is developed from Newtonian equations of motion and the concepts of arithmetic thermodynamics [33–37]. The reason for using the in silico method at this stage is that it is a powerful free energy tool that predicts the ligand-receptor complex structure and ranks our compounds with respect to a reportedinhibitor against SARS-CoV-2 Mpro as a greener pasture of biological study.

EXPERIMENTAL

General remarks. Thermogravimetric analysis (TGA) and differential thermal analysis (DTA) were performed on a TA instruments TGA 500, ranging from room temperature to 800°C in nitrogen atmosphere using a ramp rate of 10°C min–1 using aluminum crucibles.

Experimental procedure for the preparation of 1b. A mixture of 6, 6'-(((1E,1'E)-(4-bromo-1,2-phenylene)bis(methaneylylidene))bis(azaneylylidene))bis(2,4-di-tert-butylphenol) 1a (218 mg, 0.352 mmol), Pd(OAc)2 (39.5 mg, 0.176 mmol), PPh3 (74 mg, 0.282 mmol), Na2CO3 (100 mg, 0.723 mmol) and 4‑methoxyphenyl boronic acid (107 mg, 0.704 mmol) were taken in benzene (5 mL) and refluxed at 80°C for 7 h under N2 atmosphere. Upon total consumption of the starting material as evidenced by TLC, the reaction mixture was cooled to r.t. and diluted with brine solution (5 mL) and extracted with ethyl acetate (3 × 10 mL). Combined organic layer was collected and dried over Na2SO4. The organic layer was concentrated, and the residue was passed through a short silica gel column to affordcompound/1b (181 mg, 79%) as red solid. IR: 3225–3510 (O–H), 2951 (CH2), 1601 (C=N), 1415amp; 1515 and 1415 (C=CAr), 1016–1250 (C–O). 1H NMR (400 MHz, CDCl3, δ): 8.61 (s, 1H)(N=CH), 8.55 (s, 1H)(N=CH), 7.98–7.87 (m, 2H)(Ar–H), 7.63–7.48 (m, 4H)(Ar–H), 7.31–7.21 (m, 3H)(Ar–H), 7.07–6.97 (m, 2H)(Ar–H), 3.91(S, 3H)(–OMe), 1.58 (s, 9H)(–tBu), 1.57 (s, 9H)(–tBu), 1.37 (s, 9H) (–tBu), 1.36 (s, 9H)(–tBu). 13C NMR (100 MHz, CDCl3, δ): 166.2, 166.1, 166.0, 159.7, 152.2, 143.7, 142.5, 141.2, 141.2, 141.1, 136.7, 136.6, 136.6, 131.7, 131.6, 131.6, 128.3, 127.7, 127.0, 125.6, 119.8, 115.3, 114.5, 114.2, 55.4, 36.2, 34.0, 31.3, 29.6. HRMS: m/z [M + H]+ calcd. for C43H54N2O3 (%): 647.4213; found (%): 647.4212. Figs. S3, S4, S5: 1H, 13C NMR and MS for 1b.

Experimental procedure for the preparation of copper(II) complex, 1c. Ethanolic solution (10 mL) of 2b (122 mg, 0.189 mmol) was taken into a round bottom flask and Et3N (114 mg, 1.13 mmol) was added dropwise to it through a septa-syringe system. After some time ethanolic solution of Cu(OAc)2 (57 mg, 0.285 mmol) was added and the solution was refluxed at 60°C with a condenser for 1 h under N2 atmosphere. The dark brown precipitated thus obtained was filtered, washed with cold ethanol, and dried in a vacuum over anhydrous CaCl2 to afford copper complex 1c (102 mg, 76%). Recrystallization was done by slow evaporation of ethyl acetate–ethanol (1 : 1) solution of compound 1c. mp: above 250°C. IR: 2952 (CH2), 1571 (C=N), 1471, 1426 (C=CAr), 1006 (C–O), 785 (Cu–N), 696 (Cu–O).

Hirshfeld surface analysis. Two-dimensional (2D) fingerprint plots and Hirshfeld surface analysis (HAS) were calculated using Crystal Explorer [38], which reads a structure input file in CIF format.

SCXRD experimental details. Single crystal X-ray Diffraction data were collected on Bruker APEX II diffractometer using monochromated MoKα radiation (λ = 0.71073 Å) at 100 K using an Oxford cryostream low-temperature device. Unit cell measurements, data integration, scaling and absorption corrections for the crystals were done with Bruker APEXII software [39]. Data reduction was carried out with Bruker SAINT suite [40]. Absorption correction was performed by the numerical method through face-indexing of a single crystal in SADABS [41]. It was solved by direct methods using SIR 2014 [42]. The crystal structure refinements were done in the program package OLEX2 [43]. All non-hydrogen atoms were refined anisotropically by full-matrix least-squares calculations based on F2 with SHELXL-2016 [44], hydrogen atoms were included in calculated positions as riding atoms. The methyl group of the methoxy substituent on one of the two molecules in the asymmetric unit is disordered with occupancy of ~63% for the major part. For stable convergence of the refinement model, the distance and thermal ellipsoid restraints such as SADI, SIMU were employed. The PARST [45] was used for crystal structure analysis and Mercury 4.0 [46] was used for drawing molecular and crystal structures. The crystallographic data were deposited at the Cambridge Crystallographic Data Centre (CCDC) under a CCDC number 2048683. To access this structure please go to (http://www.ccdc.cam.ac.uk/services/structures?Ac-cess=referee&pid=ccdc:2048683&author=hasija) Copies of the data can be obtained via www.ccdc.cam.ac.uk.

Docking in silico studies. The docking calculations of compounds 1b, 1c and N3 using 7BQY (Mpro, PDB, 7BQY, resolution of 1.7 Å) was accomplished using the Autodock Vina wizard in PyRx 0.8. [47]. Settings are made identical for docking in this research study: Grid box center_x = –7.22607850362, center_y = ‒0.699371760206, center_z = 14.5348866365, size_x = 54.8670147657, size_y = 73.1645037698 andsize_z = 64.4162177746. The remaining parameters were used as a default setting in the Autodock Vina-PyRx. All drugs and ligands were converted to SDF file type using Chem. Draw program and were used as input to Autodock vina in PyRx. The proteases (PDB code 7BQY) were saved in PDB format after deleting the water molecules and ligands using Discovery Studio Visualizer v17.2.0.16349. The PyMOL molecular viewer was used to present the output [48]. Schematic diagrams of protein–ligand interactions were generated using the LIGPLOT program [49].

RESULTS AND DISCUSSION

Synthesis

Schiff base compound 1a was synthesized using a previously reported procedure by a one-pot condensation reaction in ethanol in good yields [50, 51]. 4,4'-Methylenedicyclohexanamine was mixed with the aldehyde in a 1 : 2 ratio in ethanol at room temperature. The mixture was then refluxed for two hours. The solvent was reduced to 1/3 and the product was allowed to form at room temperature. The product was recrystallized from hot ethanol at room temperature. Compound 1b was obtained in reasonably high yield using a mixture of 1a, Pd(OAc)2, PPh3, Na2CO3, 4‑methoxyphenylboronic acid in benzene (Scheme 1). The characteristic feature of the IR absorption bands of the free Schiff base, 1b, are: 3225-3510 (O–H), 1601 (C=N) (Fig. S1) and for the 1c are 1571 (C=N), 785 (Cu–N), 696 (Cu–O) cm–1 (Fig. S2) [52, 53]. A broad region (3225–3510 cm–1) comes due to the delocalization in the π-electron pseudo-aromatic six-membered ring forming a H bonding (O─H⋅⋅⋅N) in case of the ligand (1b). Whereas for 1c, Cu(II)L, no such O–H peaks appear due to the absence of any acidic H (Scheme 1, Fig. S2).

Scheme 1 . Synthesis of 1b and the corresponding Cu(II) complex 1c [54].

Thermogravimetric Analyses

Thermogravimetric analysis (TGA) and differential thermal analysis (DTA) of the present Cu-Metal Organic Framework (MOF), 1c, have been performed to examine the thermal behavior and stability of the sample. The results were demonstrated in Fig. 2 as the TGA profile showed four weight-loss stages corresponding to three endothermic peaks in the DTA profile. The TGA analysis revealed an initial weight loss of ca. 0.88% at 110°C, which might be attributed to the loss of adsorbed solvent molecules and free ethanol molecules on the catalyst’s surface [55–58] The second weight loss of around ca. 1.50% was observed on the TGA plot between 110–262°C, which might be attributed to the removal of guest water molecules that coordinate with copper and a tiny exothermic peak at around ca. 260.7°C (DTA curve) [59]. Following, the third weight loss of around ca. 1.98% was observed on the TGA plot between 262–493°C, which is attributed to the complete dehydration of water molecules and dissociation of the N-(4'-methoxy-[1,1'-biphenyl]-3-yl) methanimine group from the imine bond [59]. Additionally, a sharp exothermic peak at 396.8°C is associated with framework decomposition corroborates well with the TGA results. The last significant weight loss of around ca. 4.96% was observed at the temperature range of 493–800°C, which indicates the complete decomposition of framework compound into CO2 and transformation to Cu2O corroborates well with a small exothermic peak 493.6°C from the DTA curve [60, 61]. The thermal stability of the present Cu-MOF at solid-state is 396.8°C, which is higher than the recent literature report [62].

Fig. 2.
figure 2

Thermogravimetric spectra (red line) and differential thermal analysis (blue line) curve of synthesized Cu-MOF 1c.

Full size image

UV-Vis Data of Analysis

Here, the experimental absorption spectra for both 1b and 1c were taken in DCM solvent with total concentration of 20 × 10–5 M. For 1b, two intense bands have appeared at 325.0 nm, and 422.7 nm may be for π–π* and n–π* transitions, respectively. Whereas in the 1c curve, those are observed at 332.1 nm and 459.6 nm, i.e., a slight blue shift occurs. This comes may be due to the very stable complexation via CuII ion. The tetradentate ligand system gets more planarity. Here π–π* is probably happening in the π e– cloud of aromatic rings and the n–π* occurs via the lone pairs of N atom towards the C=N segments for the ligand and CuII complex [63–67] Both transitions show such high wavelengths may be due to extended conjugation through the +R effect of the para methoxy group in 1b and 1c. In the case of 1c, another possibility is that it can be ligand to metal (Cu2+) charge transfer that can cause this type of transition [65, 68–70]. An additional signature hump at 540.3 nm appears in 1c (which is obviously absent in 1b) may be due to d–d transitions [63, 64, 67, 71], resulting Jahn–Teller distortion of d9 Cu(II) ion leading to more stable square planar geometry, which is already proved by theoretical crystal X-ray diffraction data given in this work [63, 64].

Crystal Structure Analysis

The copper complex, 1c crystallized as long, red needles. Despite their shiny appearance, the crystals were weakly diffracting in nature. It crystallized in the centrosymmetric, triclinic space group, P-1, with two molecules of the Cu-complex in the asymmetric unit (depicted in pink (Cu01) and green (Cu02) color in Fig. 4a; Table 1, structure solution and crystal refinement data for 1c). The Cu02 molecule has a disordered methoxy group. As illustrated in Fig. 4b, the asymmetric unit molecules are almost planar and the angle between the molecular planes is ~16°. As depicted in Fig. 4c, on viewing along ac-plane, it is evident that the molecules are molecularly stacked in an …AABBAABB… manner down the a-axis. It is observed that the central core of the complex involving Cu and the rigid part of the ligand is primarily engaged in molecular stacking. In contrast, the peripheral part of the tertiary butyl and phenyl methoxy group is involved in C─H…π interactions and H…H contacts. The C─H…π interactions are formed between C─H of the tertiary butyl group and the rigid-phenyl part of the Cu-complex.

Fig. 3.
figure 3

Absorbance curve of our ligand 1b and Cu(II)-salophen complex 1c.

Full size image
Fig. 4.
figure 4

(a) ORTEP of 1c with 50% thermal ellipsoidal probability showing two molecules in the asymmetric unit (Cu01-pink, Cu02-green); (b) lateral view of molecule showing the deviation in planarity of Cu02 with respect to Cu01; (c) crystalline arrangement of the molecule stabilized via molecular stacking.

Full size image
Table 1.   Structure solution and crystal refinement data of 1c
Full size table

The total angles around the copper center in the 1ccomplex is 360° indicating a square planar geometry, Fig. 5. This is in agreement with the geometry index calculation for the four coordinated copper complexes, τ4 (Eq. (1)) proposed by Houser and co-workers [72]. The tetra-coordinate τ4 index ranges from 1.0 for a perfect tetrahedral geometry to 0.0 for a perfect square planar. The tetra-coordinate index for 1c is τ4 =360 – (176.2 + 176.6)/141 = 7.2/141 = 0.051 using Eq. (1), indication a square planner geometry which is in agreement with the experimental result. α and β are the two largest θ values of bond angles around the central copper metal atom, Table 2.

$${{\tau }_{4}} = \left( {360-\left( {\alpha + \beta } \right)} \right){\text{/}}141.$$
(1)
Fig. 5.
figure 5

Selected bond angles for Cu (1c) complex using Mercury package [46, 78].

Full size image
Table 2.   Selected bond angles for complex 1c
Full size table

Theoretical Study

Gas-phase optimization. With the aid of HSEH1PBE density functional [73–75] using cc-pVDZ basis set [76] for C, H, O and N atoms and LANL2DZ basis set [77, 78] for Cu atom as implemented in Gaussian09 [79], the gas-phase geometry-optimized structures were computed with no imaginary frequency, so the minimum local structures were obtained for 1b and 1c. Although the optimized geometrical parameters of 1c belong to one isolated structure, the obtained theoretical results can compared with those of both molecules in the asymmetric unit. When the experimental and theoretical structures are superimposed, it is seen that there are only minor differences with the r.m.s. the deviation between the conformations being 0.582 Å for 1c, Cu01 molecule and 0.684 Å for 1c, Cu02 molecule (Fig. 6).

Fig. 6.
figure 6

Overlay diagrams for the X-ray of 1c, (black; (a) Cu01 molecule, (b) Cu02 molecule) and calculated (red) structures.

Full size image

Molecular electrostatic potential (MEP). In order to predict and analyze electron-deficient and electron-rich sites for the complex, MEP was calculated by applying the same method and the basis sets as used for geometry optimization. The negative electrostatic potential regions (red, Vs,min) of MEP were related to electrophilic reactivity and the positive electrostatic potential regions (blue, Vs,max) to nucleophilic reactivity are shown in Fig. 7.

Fig. 7.
figure 7

Molecular electrostatic potential map (in a.u.).

Full size image

As shown in Fig. 7, the negative regions are mainly localized on the phenolate rings and the area between them, with an average value of −0.03 a.u. On the other hand, the Vs,min value is slightly lower on methoxyphenyl oxygen atoms with a value of −0.02 a.u. However, the maximum positive regions are associated with the azomethine and methoxyphenyl C─H bonds with an average value of 0.03 a.u.

Hirshfeld surface analysis. Hirshfeld surface analysis (HSA) was used in this study to help understand the supramolecular arrangement of complex 1c, [38]. HSA provides a three-dimensional figure of close contact of the molecule in a crystal. These contacts are summarized in the fingerprint plot (Fig. 6). The distances from the nearest atoms outside the Hirshfeld surface and inside the surface are defined by quantities de and di. Both distances, the di and de surfaces, are displayed in Fig. 2 for clarity. The combination of di and de is presented on the fingerprint plot (Fig. 3), provides more information about all the contacts in the molecule. The present intermolecular interactions (C…C, C…H and H…H) stabilize the molecular packing. These interactions connect 1c in layers as shown in Fig. 4c. The HSA index map aids in analyzing 1c molecular contacts by color code: red indicates high-intensity contacts and blue indicates low-intensity contacts Fig. 4a [25]. The fingerprint plots clearly show that H…H contacts represent a very high percentage of the interactions in 1c layers (71.2%), with total di + de 1.91 Å, as shown in Fig. 6. The H…C interactions show much fewer interactions (16.9%). H…N and H…O contacts show fewer interactions (1.2 and 2.5%, respectively). The spikes due to the C…N and C…O contacts are farther apart on the fingerprint plots, with di + de 3.2 Å and 2.82 Å respectively. Hirshfeld surfaces views mapped over dnorm with enabled surface transparency for both sides of the copper complex of 1c is shown in Figs. 8 and 9.

Fig. 8.
figure 8

Hirshfeld surface fingerprint plots of the nearest internal distance (di) versus the nearest external distance (de) for complex 1c. Intra- and intermolecular contacts are abbreviated di and de, respectively.

Full size image
Fig. 9.
figure 9

Hirshfeld surface dnorm map of thefront and back views of 1c, showing C…C, C…H and H…H interactions detected among molecules as red spots. This is one of two molecules of the Cu-complex in the asymmetric unit. Red indicates high-intensity contacts, whereas blue indicates low-intensity connections.

Full size image

Molecular docking against 7BQY. Due to Covid-19 bad impact on economic and social life, we have been encouraged to study how our ligands/1b and 1c might behave in the main protease’s active site for the virus (MPro, PDB 7BQY) when compared to the well-known N3 inhibitor [80], in an attempt to contribute in solving the outbreak of the pandemic disease. In this study, we used the 7BQY because it has been docked against the well-known inhibitor N3 that appeared recently in Nature journal and has a good resolution value of 1.70 Å [80]. This study was carried out in silico to rank 1b and 1c with respect to N3 inhibitors before allowing any vivo study to save energy [81]. The docking was achieved, against 7BQY, using the same parameters for a fair comparison, Figs. 8 and 9. Worth mentioning here that the ligand was treated as rigid mode during Docking rather than flexible mode [32]. The types of interactions between COVID-19 main protease (Mpro, PDB code, 7BQY) and host (N3, 1b, or 1c, this study) simplify the protease proteolytic activity involved in SARS-CoV-2, which was presented in terms of docking scores [82]. Therefore, molecular docking studies were presented to examine the binding performance of 1b and 1c compared to the well-established inhibitor, N3, against 7BQY, Table 3 and 2 and Figs. 10 and 11. The results show both our compounds are close in affinity scores; however, the total interactions are less when compared to N3, Tables 3 and 4. Compound N3 exhibits four H-bondings, whereas 1b and 1c show 2 and 0, respectively. Compounds 1b and 1c do/not demonstrate further negative (better) values in affinity and display many Van Der Waals interactions, Fig. 10. It is possible to conclude that 1b and 1c merit further studies to investigate the antiviral properties of these drugs against COVID-19. Further studies will enhance explaining this point.

Table 3.   Docking scores on Mpro, kcal/mol
Full size table
Fig. 10.
figure 10

Interaction map of 1b and N3 with the active site residues of Moro. The colored circle (residue) characterizes the type of interaction. Dashed lines also represent the type and strength of interaction presented by Biovia Discovery Studio v.4.5. Color codes are defined, e.g., green lines indicate hydrogen bonds.

Full size image
Fig. 11.
figure 11

A schematic 2D LIGPLOT representation of 1c against 7BQY complex showing the hydrophobic interactions(red semi-circles) [49]. The purple compound is 1c complex.

Full size image
Table 4.   Listing molecular interactions for N3, 1b and 1c
Full size table

CONCLUSIONS

The focus in this work is on the structural and electronic properties of the Cu–Schiff base complex. The tetra-coordinate τ4 index was calculated, indicating almost a perfect square planar in agreement with Xray diffraction results. The molecular electron potential (MEP) map reveals the maximum positive regions in 1c associated with the azomethine and methoxyphenyl C─H bonds with an average value of 0.03 a.u. Hirshfeld Surface Analysis highlighted the significant inter-atomic contacts and their percentage contribution through 2D Fingerprint plot. An interesting comparative molecular docking study of 1b and 1c and the inhibitor N3 against the main protease Mpro, (PDB code 7BQY) using the same parameters and conditions, suggest 1b and 1c worth further study in the context of possible therapeutic agents for COVID-19. In greener pasture gesture free energy approach, in silico molecular docking is used to rank the title compounds with respect to the well-known inhibitor, N3. The binding scores of 1b, 1c, N3 are –7.8, –9.0 and ‒8.4 kcal/mol, respectively.