New N-Alkylated Heterocyclic Compounds as Prospective NDM1 Inhibitors: Investigation of In Vitro and In Silico Properties

A new family of pyrazole-based compounds (1–15) was synthesized and characterized using different physicochemical analyses, such as FTIR, UV-Visible, 1H, 13C NMR, and ESI/LC-MS. The compounds were evaluated for their in vitro antifungal and antibacterial activities against several fungal and bacterial strains. The results indicate that some compounds showed excellent antibacterial activity against E. coli, S. aureus, C. freundii, and L. monocytogenes strains. In contrast, none of the compounds had antifungal activity. Molecular electrostatic potential (MEP) map analyses and inductive and mesomeric effect studies were performed to study the relationship between the chemical structure of our compounds and the biological activity. In addition, molecular docking and virtual screening studies were carried out to rationalize the antibacterial findings to characterize the modes of binding of the most active compounds to the active pockets of NDM1 proteins.


Introduction
Resistance to antibiotics pushes researchers to discover new antibacterial candidates as prospective treatments for different infectious diseases with another class of antibiotics with specific mechanisms of action.

Antibacterial and Antifungal Activities
The antibacterial potential of the compounds against two Gram-negative bacterial strains (Escherichia coli and Citrobacter freundii) and two Gram-positive bacteria (Staphylococcus aureus and Listeria monocytogenes) was evaluated as described in the materials and methods section, and the results are displayed in Table 1. Only compounds 12 and 14 showed antibacterial activity when tested at 500 μM. Compound 12 was active against E. coli, S. aureus, and C. freundii but inactive against L. monocytogenes, whereas compound 14 was only active against L. monocytogenes. Table 1. The antibiotic activity of the active synthesized pyrazole ligands was determined using the broth macro dilution assay and the phenol red indicator.  Infectious diseases are the main threat, especially in developing countries [21], and include listeriosis [4,[22][23][24] caused by L. monocytogenes, septicemia, and meningitis caused by E. coli [8,25], bloodstream infections, and meninges caused by C. freundii [26][27][28][29]. These diseases commonly affect healthy, sensitive individuals, such as older people, pregnant women, and the immunosuppressed [11]. In addition, however, candidiasis and related fungal bloodstream infections are caused by Saccharomyces cerevisiae [30], Candida albicans, and Candida glabrata [31].

Antibacterial and Antifungal Activities
The antibacterial potential of the compounds against two Gram-negative bacterial strains (Escherichia coli and Citrobacter freundii) and two Gram-positive bacteria (Staphylococcus aureus and Listeria monocytogenes) was evaluated as described in the materials and methods section, and the results are displayed in Table 1. Only compounds 12 and 14 showed antibacterial activity when tested at 500 µM. Compound 12 was active against E. coli, S. aureus, and C. freundii but inactive against L. monocytogenes, whereas compound 14 was only active against L. monocytogenes. The MICs of compounds 12 and 14 were then determined as described in the material and methods ( Table 2). The MIC of compound 12 was 134.9 mg/L against E. coli, 168.7 mg/L against S. aureus, and 168.7 mg/L against C. freundii. For compound 14, the MIC against L. monocytogenes was 134.6 mg/L. Interestingly, the determination of the MBC of these compounds showed that they are bactericidal, as demonstrated by the ratio of MBC/MIC ≤ 2 (Table 2). Regarding the antifungal activity, all the compounds were tested for toxicity against Saccharomyces cerevisiae and two species of Candida, Candida glabrata and Candida albicans, as described in Section 3.3.3. All compounds showed no antifungal activity against all three strains used. Together with the antibacterial activity analysis, these results suggest that compounds 1 to 15 lack antifungal activity, and only compounds 12 and 14 act specifically as antibacterial agents.

MEP Analysis of the Compounds 12, 14, Ampicillin, and Cefotaxime
Molecular electrostatic potential (MEP) maps of compounds 12, 14, Ampicillin, and cefotaxime have been generated ( Figure 2) to determine and predict the reactive sites (nucleophilic or electrophilic) on the molecular system of the studied compounds. and methods ( Table 2). The MIC of compound 12 was 134.9 mg/L against E. coli, 168.7 mg/L against S. aureus, and 168.7 mg/L against C. freundii. For compound 14, the MIC against L. monocytogenes was 134.6 mg/L. Interestingly, the determination of the MBC of these compounds showed that they are bactericidal, as demonstrated by the ratio of MBC/MIC ≤ 2 ( Table 2). Regarding the antifungal activity, all the compounds were tested for toxicity against Saccharomyces cerevisiae and two species of Candida, Candida glabrata and Candida albicans, as described in Section 3.3.3. All compounds showed no antifungal activity against all three strains used. Together with the antibacterial activity analysis, these results suggest that compounds 1 to 15 lack antifungal activity, and only compounds 12 and 14 act specifically as antibacterial agents.

MEP Analysis of the Compounds 12, 14, Ampicillin, and Cefotaxime
Molecular electrostatic potential (MEP) maps of compounds 12,14, Ampicillin, and cefotaxime have been generated ( Figure 2) to determine and predict the reactive sites (nucleophilic or electrophilic) on the molecular system of the studied compounds.  As presented in Figure 2, the positive electrostatic potential areas (in blue) are concentrated over the hydroxyl group of the drugs cefotaxime and Ampicillin with 1.570 and 1.801 eV (at an iso value of 0.0004 electrons/Å 3 ). For compound 14, the highest value was about 1.790 eV (located on the hydroxyl group substituent on the phenyl ring), while a low value of 0.040 eV was estimated for compound 12. The negative electrostatic potential areas are located over the carbonyl group of the acidic function of the drugs cefotaxime and Ampicillin, with values of −1.028 and −0.946 eV, respectively, with a higher value for the compound 12, with the value of −1.521 eV. In comparison, it was −0.832 eV for compound 14.
To sum up, compound 12 only has a higher negative charge value over the carbonyl. In contrast, compound 14, similar to the antibiotics cefotaxime and Ampicillin, has lower negative-positive (δ + -δ − ) charge values over the hydroxyl and the acidic function. These results, which agree with the experimental results, give us information about the possible sites for binding modes to the biological targets that need molecular docking investigations.
In general, and in the light of these observations, we can postulate that the substitution by groups with different electronic effects (withdrawing-donating) is more beneficial in improving the antibacterial potency than the substitution by one group (Figure 3). Figure 2, the positive electrostatic potential areas (in blue) are concentrated over the hydroxyl group of the drugs cefotaxime and Ampicillin with 1.570 and 1.801 eV (at an iso value of 0.0004 electrons/Å 3 ). For compound 14, the highest value was about 1.790 eV (located on the hydroxyl group substituent on the phenyl ring), while a low value of 0.040 eV was estimated for compound 12.

As presented in
The negative electrostatic potential areas are located over the carbonyl group of the acidic function of the drugs cefotaxime and Ampicillin, with values of −1.028 and −0.946 eV, respectively, with a higher value for the compound 12, with the value of −1.521 eV. In comparison, it was −0.832 eV for compound 14.
To sum up, compound 12 only has a higher negative charge value over the carbonyl. In contrast, compound 14, similar to the antibiotics cefotaxime and Ampicillin, has lower negative-positive (δ + -δ − ) charge values over the hydroxyl and the acidic function. These results, which agree with the experimental results, give us information about the possible sites for binding modes to the biological targets that need molecular docking investigations.
In general, and in the light of these observations, we can postulate that the substitution by groups with different electronic effects (withdrawing-donating) is more beneficial in improving the antibacterial potency than the substitution by one group (Figure 3). Regarding the R1 (substituent on the phenyl ring) and R2 (substituent at positions 3 and 5 of the pyrazole moiety) substituents, the inductive and mesomeric effect study revealed that the presence of the formyl (CHO) group (electron-withdrawing effect (-M)) on the phenyl ring and the methyl (CH3) groups (electron-donating effect (+I)) on the pyrazole moieties at positions 3 and 5 (12) is highly favorable for the inhibitory potency against E. coli, S. aureus, and C. freundii strains. In contrast, the presence of the hydroxyl (OH) group (electron-donating effect (+M)) on the phenyl ring with non-substituted pyrazole moieties (14) resulted in selective antibacterial activity against L. monocytogenes.

ADME Predictions
For the ADME predictions, the physicochemical properties (MW: molecular weight expressed in Daltons; logP: octanol/water partition coefficient characterizing Lipophilicity; HDO: number of hydrogen bond donors; HAC: number of hydrogen bond acceptors; NRO: number of rotatable bonds; TPSA: total polar surface area) were calculated and are presented in Table 3 for the compounds 12 and 14 and the drugs streptomycin, Ampicillin, and cefotaxime as references. Regarding the R 1 (substituent on the phenyl ring) and R 2 (substituent at positions 3 and 5 of the pyrazole moiety) substituents, the inductive and mesomeric effect study revealed that the presence of the formyl (CHO) group (electron-withdrawing effect (-M)) on the phenyl ring and the methyl (CH 3 ) groups (electron-donating effect (+I)) on the pyrazole moieties at positions 3 and 5 (12) is highly favorable for the inhibitory potency against E. coli, S. aureus, and C. freundii strains. In contrast, the presence of the hydroxyl (OH) group (electron-donating effect (+M)) on the phenyl ring with non-substituted pyrazole moieties (14) resulted in selective antibacterial activity against L. monocytogenes.

ADME and Toxicity Predictions ADME Predictions
For the ADME predictions, the physicochemical properties (MW: molecular weight expressed in Daltons; logP: octanol/water partition coefficient characterizing Lipophilicity; HDO: number of hydrogen bond donors; HAC: number of hydrogen bond acceptors; NRO: number of rotatable bonds; TPSA: total polar surface area) were calculated and are presented in Table 3 for the compounds 12 and 14 and the drugs streptomycin, Ampicillin, and cefotaxime as references.  These results make compound 12 a better antibacterial candidate than cefotaxime with the same selective multitarget activity, but further toxicity predictions are required to validate these propositions.

Molecular Docking and Virtual Screening Studies
Ligand-protein docking simulations were carried out to determine the binding mode of the studied compounds with the catalytic sites of the selected receptors. Flexibility was allowed in all the rotatable bonds of the ligand; the protein was used as a rigid structure.
The three-dimensional structure of New Delhi metallo-β-lactamase (NDM1) [17,58,59] is represented in Figure 4, where two sequences, A and B, were co-crystalized with Ampicillin; so in this part, we studied the binding affinity of the compounds 12 and 14 compared to Ampicillin.  As previously mentioned in the docking study of the transpeptidase inhibition study, the NDM1 hydrolysis followed the same parameters, and the binding affinity results of both active sites are collected in Table 4.
From Table 5, compound 12 had a better affinity than compound 14 in both selected The protein preparation was performed by removing all the water, zinc, and OH molecules.
As previously mentioned in the docking study of the transpeptidase inhibition study, the NDM1 hydrolysis followed the same parameters, and the binding affinity results of both active sites are collected in Table 4.
From Table 5, compound 12 had a better affinity than compound 14 in both selected active sites, NDM1 A and B, with a binding affinity of −6.0075 and −6.6776 Kcal/mol, respectively. In addition, compared to Ampicillin, with −6.9737 and −6.7344 Kcal/mol, compound 12 was more readily hydrolyzed than compound 14. First, compound 12 has two H-acceptors, bond 3.08 and 2.90 Å, respectively, between the carbonyl and the nitrogen ND1 and NE2 of His122 and His 189, as shown in Table 5 and Figure 5. First, compound 12 has two H-acceptors, bond 3.08 and 2.90 Å, respectively, between the carbonyl and the nitrogen ND1 and NE2 of His122 and His 189, as shown in Table 5 and Figure 5. Contrary to compound 12, compound 14, as represented in Figure 6, has only weak van der Waals bonds with Asn 220 and His 122, with a distance range of 3.60-3.90 Å, making this compound more stable against NDM1 hydrolysis. As penicillin β-lactam antibiotics reported in the literature, Ampicillin seems to in- Contrary to compound 12, compound 14, as represented in Figure 6, has only weak van der Waals bonds with Asn 220 and His 122, with a distance range of 3.60-3.90 Å, making this compound more stable against NDM1 hydrolysis.
As penicillin β-lactam antibiotics reported in the literature, Ampicillin seems to interact with more residues at the site (A) than at the site (B) of NDM1 protein, as represented in Table 5 and Figure 7. In the active pocket A, the compound forms seven bonds: one H-donor and one H-acceptor (of 2.90 and 3.55 Å, respectively) with Asp124 residue, two H-acceptor bonds with Lys211 amino acid with distances equal to 2.98 and 3.37 Å, one H-pi interaction (4.07 Å) between the carbon (C16 12) atom and 5-ring of His250, and two pi-H interactions between the 6-ring and the nitrogen and carbon atoms of Gln123 (4.08 and 3.78 Å). In contrast, Ampicillin was found to bind into the active pocket B of NDM1 only with three interactions: two by H-donor and H-acceptor bonds with Asp124 (2.99 and 3.55 Å) and one by an H-acceptor bond with Lys211 with a distance of 2.88 Å involving its oxygen atom (O1 1).  Contrary to compound 12, compound 14, as represented in Figure 6, has only weak van der Waals bonds with Asn 220 and His 122, with a distance range of 3.60-3.90 Å, making this compound more stable against NDM1 hydrolysis. As penicillin β-lactam antibiotics reported in the literature, Ampicillin seems to interact with more residues at the site (A) than at the site (B) of NDM1 protein, as represented in Table 5 and Figure 7. In the active pocket A, the compound forms seven bonds: one H-donor and one H-acceptor (of 2.90 and 3.55 Å, respectively) with Asp124 residue, two H-acceptor bonds with Lys211 amino acid with distances equal to 2.98 and 3.37 Å, one H-pi interaction (4.07 Å) between the carbon (C16 12) atom and 5-ring of His250, and two pi-H interactions between the 6-ring and the nitrogen and carbon atoms of Gln123 (4.08 and 3.78 Å). In contrast, Ampicillin was found to bind into the active pocket B of NDM1 only with three interactions: two by H-donor and H-acceptor bonds with Asp124 (2.99 and 3.55 Å) and one by an H-acceptor bond with Lys211 with a distance of 2.88 A° involving its oxygen atom (O1 1).  As penicillin β-lactam antibiotics reported in the literature, Ampicillin seems to interact with more residues at the site (A) than at the site (B) of NDM1 protein, as represented in Table 5 and Figure 7. In the active pocket A, the compound forms seven bonds: one H-donor and one H-acceptor (of 2.90 and 3.55 Å, respectively) with Asp124 residue, two H-acceptor bonds with Lys211 amino acid with distances equal to 2.98 and 3.37 Å, one H-pi interaction (4.07 Å) between the carbon (C16 12) atom and 5-ring of His250, and two pi-H interactions between the 6-ring and the nitrogen and carbon atoms of Gln123 (4.08 and 3.78 Å). In contrast, Ampicillin was found to bind into the active pocket B of NDM1 only with three interactions: two by H-donor and H-acceptor bonds with Asp124 (2.99 and 3.55 Å) and one by an H-acceptor bond with Lys211 with a distance of 2.88 A° involving its oxygen atom (O1 1). From the docking study results, compound 12 is more sensitive to NDM1 hydrolysis, which inactivates its antibacterial activity against Listeria monocytogenes due to the common carbonyl function in all β-lactam antibiotics. Although otherwise, compound 14 has listericidal activity due to its weak binding with the selected NDM1, its specific mechanism needs more computational studies and biological assays for prediction. From the docking study results, compound 12 is more sensitive to NDM1 hydrolysis, which inactivates its antibacterial activity against Listeria monocytogenes due to the common carbonyl function in all β-lactam antibiotics. Although otherwise, compound 14 has listericidal activity due to its weak binding with the selected NDM1, its specific mechanism needs more computational studies and biological assays for prediction.

Blind Docking/Virtual Screening against the NDM-1 β-lactamase (NDM1) Protein
As presented in Figure 8, the B-chain has considered the active site with all the docking poses. At the same time, there is good alignment between the ligand 12 and ampicillin docking poses with smooth variation, while ligand 14 is so far in a different site.  Table 6 shows that the mode of binding interaction is the same as that of LYS216, which is bound with the nitrogen of pyrazole for the ligand 12 and the oxygen for Ampicillin with the same binding affinity of −7.1 Kcal/mol. On the other hand, ligand 14 has a lower −7.0 Kcal/mol value with amino acids such as ILE203 and LYS242. Table 6. Binding affinity and L-AA interaction of the ligands 12 and 14 and ampicillin with the  Table 6 shows that the mode of binding interaction is the same as that of LYS216, which is bound with the nitrogen of pyrazole for the ligand 12 and the oxygen for Ampicillin with the same binding affinity of −7.1 Kcal/mol. On the other hand, ligand 14 has a lower −7.0 Kcal/mol value with amino acids such as ILE203 and LYS242. Table 6. Binding affinity and L-AA interaction of the ligands 12 and 14 and ampicillin with the chosen target, NDM1.

Compound
Binding Affinity in kcal/mol Interaction L-AA (Hydrogen Bonds Only)

Analytical Procedures
A Bruker DPX 800 MHz Spectrometer recorded the 1 HNMR (500 MHz, DMSO-d6) and 13 CNMR (125 MHz, chloroform-d) spectra. Chemical shift (δ) values were stated in parts per million (ppm) using internal standard tetramethylsilane, according to the D 2 O exchange. Chemical shift (d) values were stated in parts per million (ppm) using internal standard tetramethylsilane. The D 2 O exchange confirmed the exchangeable protons (OH and NH). The FTIR analyses were performed using an FTIR 8400S spectrophotometer recorded in KBr pellets.
Many different pyrazole derivatives were synthesized and indexed.
Pharmaceuticals 2022, 15, x FOR PEER REVIEW 9 standard tetramethylsilane. The D2O exchange confirmed the exchangeable protons and NH). The FTIR analyses were performed using an FTIR 8400S spectrophotometer orded in KBr pellets. Many different pyrazole derivatives were synthesized and indexed.

Antibacterial Assay
The microdilution method with phenol red [88] evaluated the antibacterial effect against four bacterial strains: Listeria monocytogenes, Escherichia coli, Staphylococcus aureus, and Citrobacter freundii. First, the bacterial isolate was cultivated in liquid Luria-Bertani medium (LB) overnight at 37 • C under aeration. Then, a suspension containing 10 6 CFU/mL of bacteria cells was prepared. Next, an aliquot from this bacterial suspension was added to test tubes containing phenol red medium and the compound to be tested. After an incubation of 24 h at 37 • C, the color of the culture remains red in the absence of growth, indicating that the tested compound has antibacterial activity against the tested strain. However, if there is bacterial growth, the culture becomes yellow due to the acidification of the medium and indicating that the tested compound lacks antibacterial activity against the tested strain. All the experiments were repeated twice for each drug, including the antibiotic streptomycin as a positive control, and means were calculated.

Determination of the Minimum Inhibitory Concentration (MIC) and the Minimum Bactericidal Concentration (MBC)
The MIC (the lowest drug concentration that inhibits bacterial growth after incubation at 37 • C for 24 h) and the MBC (the lowest drug concentration that kills 99% of bacteria after 24 h of incubation) were obtained as described in the literature [89,90].

Antifungal Assay
The prepared ligands were evaluated for their antifungal activity using liquid cell culture against Saccharomyces cerevisiae and Candida species: Candida glabrata and Candida albicans. The growth rate of fungal cells in liquid culture was monitored by absorbance measurements at 600 nm (OD 600 ) using a V-1200 spectrophotometer (Shanghai Mapada Instruments Co., Ltd., Shanghai, China). The antifungal activity was described (Bouchal et al., 2019 [88]). Briefly, cells were cultured in the presence and absence of 500 µM of each compound for 24 h, and OD600 measurements were then used to monitor the growth rate. Growth in the presence of a compound was expressed as a percentage relative to the untreated control. All experiments were repeated at least twice, and means were calculated.

DFT, Molecular Ligand-Protein Docking, and Virtual Screening Studies
The chemical structures of the studied molecules were sketched using Gaussview 6.0, then optimized using the DFT/B3LYP method with 6-31G(d,p) basis sets in the Gaussian 09W software [114]. The docking study performed with New Delhi metalloβ-lactamase was conducted in two different active sites of the crystal structure of NDM-1 at pH 5.5 (Bis-Tris) in a complex with hydrolyzed Ampicillin (PDB: 5ZGE) with a resolution of 1.00 Å.
Blind docking/virtual screening was considered to target the previous protein (5ZGE), which was prepared in Autodock 4 [115] default parameters, and the whole protein was used for the grid box (Table S1), with Perl as a launcher for all the ligands in Autodock Vina [116][117][118][119][120].

Conclusions
Fifteen compounds based on pyrazole derivatives were prepared with good yield and characterized using different physicochemical analyses, such as FTIR, UV-Visible, 1 H and 13 C NMR, and MS. These compounds were evaluated for their antifungal and antibacterial activities against several fungal and bacterial strains. None of the compounds had antifungal activity. Interestingly, compounds 12 and 14 displayed intense antibacterial activity. In addition, compound 12 presented excellent antibacterial activity against E. coli, S. aureus, and C. freundii, with inactivity against L. monocytogenes and cephalosporins antibiotics. In contrast, compound 14 showed tremendous antibacterial potential against L. monocytogenes, with no effect against the other bacterial strains.
Compound 14's listericidal activity could be due to the presence of the hydroxyl as a substituent on the phenyl ring, an electron donor group (+M) that causes oxidative stress to the bacterial strain the production of hydroxyl radicals. In contrast, compound 12 has carbonyl as an electron-withdrawing group (-M), methyl as an electron donor group (+I), and a bulky substituent. Furthermore, from the MEP surface analysis, compound 12 only has a higher negative charge value over the carbonyl. In contrast, compound 14, similar to the antibiotics cefotaxime and Ampicillin, has a close negative-positive (δ + -δ − ) higher charge value over the hydroxyl and the acidic function.
From ADME and toxicity predictions, compounds 12 and 14 have no violations of Lipinski's rule of five, better than streptomycin, with three violations, and cefotaxime, with one offense. In contrast, compounds 12 and 14 have lower predicted LD 50 than Ampicillin and cefotaxime, with less toxicity (class 4) than streptomycin (class 3), even though they are both active as carcinogens with mutagenic activity for compound 14 and have no binding probability with all the toxicity targets better than Ampicillin and cefotaxime, which have probable binding with toxicity targets.
From the docking results, compound 12 has a better affinity with both active protein sites, which have an H-acceptor bond, than compound 14, with ligand exposure. However, Ampicillin and cefotaxime still have the best values, with more H-donors and H-acceptors; otherwise, compound 12 is hydrolyzed by NDM1 hydrolysis and inactivates its antibacterial activity against Listeria monocytogenes due to the presence of the carbonyl function, which is common in all β-lactam antibiotics. On the other hand, compound 14 has listericidal activity with a specific mechanism that needs more computational studies and biological assays for prediction; from blind docking/virtual screening studies, the mode of binding interaction is the same as that of LYS216, which is bound with the nitrogen of pyrazole for the ligand 12 and the oxygen for Ampicillin with the same binding affinity of −7.1 Kcal/mol. On the other hand, ligand 14 has a lower −7.0 Kcal/mol value with amino acids such as ILE203 and LYS242.