Benzoquinoline Derivatives: A Straightforward and Efficient Route to Antibacterial and Antifungal Agents

We report here the design, synthesis, experimental and in silico evaluation of the antibacterial and antifungal activity of some new benzo[f]quinoline derivatives. Two classes of benzo[f]quinolinium derivatives—(benzo[f]quinolinium salts (BQS) and pyrrolobenzo[f]quinolinium cycloadducts (PBQC)—were designed and obtained in two steps via a direct and facile procedure: quaternization followed by a cycloaddition reaction. The synthesized compounds were characterized by elemental and spectral analysis (FT-IR, 1H-NMR, 13C-NMR). The antimicrobial assay reveals that the BQS salts have an excellent quasi-nonselective antifungal activity against the fungus Candida albicans (some of them higher that the control drug nystatin) and very good antibacterial activity against the Gram positive bacterium Staphylococcus aureus. The PBQC compounds are inactive. Analysis of the biological data reveals interesting SAR correlations in the benzo[f]quinolinium series of compounds. The in silico studies furnished important data concerning the pharmacodynamics, pharmacokinetics and ADMET parameters of the BQS salts. Studies of the interaction of each BQS salt 3a–o with ATP synthase in the formed complex, reveal that salts 3j, 3i, and 3n have the best fit in a complex with ATP synthase. Study of the interaction of each BQS salt 3a-o with TOPO II in the formed complex reveals that salts 3j and 3n have the best-fit in complex with TOPO II. The in silico ADMET studies reveal that the BQS salts have excellent drug-like properties, including a low toxicity profile. Overall, the experimental and in silico studies indicate that compounds 3e and 3f (from the aliphatic series), respectively, and 3i, 3j and 3n (from the aromatic series), are promising leading drug candidates.


Introduction
According to the WHO, infectious diseases, especially those caused by bacterial (Gram positive and Gram negative) and fungal microorganisms, have become a serious threat to the global health system, being responsible for 22% of all deaths and 27% of disabilityadjusted life years worldwide [1]. Antibiotics play a key central role in antimicrobial therapy, and are one of the most effective and successful weapons against different microorganisms. However, the overuse and misuse of antibiotics have led to widespread drug resistance (DR), multi-drug resistance (MDR) and extensive-drug-resistance (EDR), creating an urgent need for new antimicrobial agents [1][2][3][4].

Desingn and Chemistry
Taking into consideration the abovementioned data, we decided to combine the pharmacophoric antimicrobial capabilities (Scheme 1) of quinoline and its benzo derivatives [26] and also to convert them in salts, in view of the fact that salts usually have better antimicrobial activity and also better pharmacokinetic properties; we also had in mind increasing the number of fused cycles from three in benzo[f]quinoline to four and to see their influence in regards to the antimicrobial activity. In equal measure we were interested in seeing the influence on antimicrobial activity of the substituents on the quaternized nitrogen, taking into account two series an aliphatic and aromatic one. As a result, two series of benzo[f]quinoline derivatives was designed and synthesized: BQS salts and PBQC cycloadducts (Scheme 1).

Desingn and Chemistry
Taking into consideration the abovementioned data, we decided to combine pharmacophoric antimicrobial capabilities (Scheme 1) of quinoline and its benzo deri tives [26] and also to convert them in salts, in view of the fact that salts usually have be antimicrobial activity and also better pharmacokinetic properties; we also had in m increasing the number of fused cycles from three in benzo[f]quinoline to four and to their influence in regards to the antimicrobial activity. In equal measure we were in ested in seeing the influence on antimicrobial activity of the substituents on the qua nized nitrogen, taking into account two series an aliphatic and aromatic one. As a res two series of benzo[f]quinoline derivatives was designed and synthesized: BQS salts a PBQC cycloadducts (Scheme 1). Scheme 1. Design in the class of benzoquinolines derivatives with antimicrobial activity.
In order to synthesize our fused quinoline derivatives we used a direct and facile t step procedure: quaternization of nitrogen heterocycles followed by a cycloaddition re tion. Using an adaptation of the setup procedure from the literature [27][28][29], in the f step, we performed a quaternization reaction of benzo[f]quinoline (1) with variously a vated α-halocarbonyl compounds 2a-o (such as: 1-bromo-alkyl-2-one, 2-iodoacetam (un)substituted phenacyl bromides), when the corresponding benzo[f]quinolinium q ternary salts BQS 3a-o were obtained (Scheme 2). The next step consists in a Hüisgen Scheme 1. Design in the class of benzoquinolines derivatives with antimicrobial activity.
In order to synthesize our fused quinoline derivatives we used a direct and facile two step procedure: quaternization of nitrogen heterocycles followed by a cycloaddition reaction. Using an adaptation of the setup procedure from the literature [27][28][29], in the first step, we performed a quaternization reaction of benzo[f]quinoline (1) with variously activated α-halocarbonyl compounds 2a-o (such as: 1-bromo-alkyl-2-one, 2-iodoacetamide, (un)substituted phenacyl bromides), when the corresponding benzo[f]quinolinium qua- ternary salts BQS 3a-o were obtained (Scheme 2). The next step consists in a Hüisgen [3 + 2] dipolar cycloaddition of the benzo[f]quinolinium ylides (generated in situ from the corresponding benzo[f]quinolinium quaternary salts BQS 3a-o, in alkaline medium) to alkyne dipolarophiles (non-symmetrically or symmetrically substituted Z-alkynes, methyl propiolate and dimethyl acetylenedicarboxylate, DMAD), when the corresponding fused pyrrolobenzo[f]quinolinium cycloadducts PBQC 4, were obtained (Scheme 2). Initially, the cycloaddition reactions were performed in the case of salts 3e (aliphatic) and 3g (aromatic), and the obtained cycloadducts PBQC (4e1, 4e2, 4g1, and 4g2) was subject to antimicrobial assay. Because the antimicrobial activity of the obtained cycloadducts PBQC 4e and 4g, was negligible, we did not continue our studies in this direction. The structures of compounds were proved by elemental and spectral analysis (FT-1 H-NMR, 13 C-NMR, and two-dimensional experiments 2D-COSY, HMQC, HMBC). T main data furnished by FT-IR and NMR spectral analysis are listed in Table 1. In the F IR spectra of BQS salts 3a-o, the most important signals are those one of carbonyl grou In the aromatic BQS salts 3g-o, the signals corresponding to carbonyl ketone group situated between 1717 cm −1 (R = NO2, 3l) and 1673 cm −1 (R = OCH3, 3i), in accordance w the electronic effects of substituents on the 4-position of a phenyl ring (electron-withdra ing for NO2 and electron-donating for OCH3). In the aliphatic series (BQS salts 3a-f), signals corresponding to carbonyl group are situated between 1626 cm −1 (amide CO, and 1743 cm −1 (carboxymethyl CO, 3b), in accordance with the structures of the co pounds (amide and aliphatic ester). In the 1 H-NMR spectra the most important signals those of the H-2, H-4, and H-11 (methylene (-CH2, -) hydrogen) atoms. As it can be se in Table 1 the H-11 methylene hydrogen atoms appear at an unusual very high chemi shift, around 6.50 ppm in the aliphatic series and around 7.00 ppm in the aromatic ser This is in accordance with the powerful electron-withdrawing effect exerted by the po tive α-endocyclic nitrogen atom of the benzo[f]quinoline ring. The aromatic/aliphatic d ference of about 0.50 ppm is because of the supplementary electron-withdrawing eff exerted by the aromatic phenyl ring. The H-2 and H-4 protons from the pyridine r appear at high chemical shift, around 10.25 ppm H-4 (positive γ-endocyclic nitrogen) a around 9.50 ppm H-2 (positive α-endocyclic nitrogen). The structures of compounds were proved by elemental and spectral analysis (FT-IR, 1 H-NMR, 13 C-NMR, and two-dimensional experiments 2D-COSY, HMQC, HMBC). The main data furnished by FT-IR and NMR spectral analysis are listed in Table 1. In the FT-IR spectra of BQS salts 3a-o, the most important signals are those one of carbonyl groups. In the aromatic BQS salts 3g-o, the signals corresponding to carbonyl ketone group are situated between 1717 cm −1 (R = NO 2 , 3l) and 1673 cm −1 (R = OCH 3 , 3i), in accordance with the electronic effects of substituents on the 4-position of a phenyl ring (electronwithdrawing for NO 2 and electron-donating for OCH 3 ). In the aliphatic series (BQS salts 3a-f), the signals corresponding to carbonyl group are situated between 1626 cm −1 (amide CO, 3a) and 1743 cm −1 (carboxymethyl CO, 3b), in accordance with the structures of the compounds (amide and aliphatic ester). In the 1 H-NMR spectra the most important signals are those of the H-2, H-4, and H-11 (methylene (-CH 2 , -) hydrogen) atoms. As it can be seen in Table 1 the H-11 methylene hydrogen atoms appear at an unusual very high chemical shift, around 6.50 ppm in the aliphatic series and around 7.00 ppm in the aromatic series. This is in accordance with the powerful electron-withdrawing effect exerted by the positive α-endocyclic nitrogen atom of the benzo[f]quinoline ring. The aromatic/aliphatic difference of about 0.50 ppm is because of the supplementary electron-withdrawing effect exerted by the aromatic phenyl ring. The H-2 and H-4 protons from the pyridine ring appear at high chemical shift, around 10.25 ppm H-4 (positive γ-endocyclic nitrogen) and around 9.50 ppm H-2 (positive α-endocyclic nitrogen). The 13 C-NMR spectra of BQS salts 3a-o also confirm the structurse and are in accordance with the FT-IR and 1 H-NMR data presented above. The carbon from the ketone CO group (C-12) appears around 190 ppm in the aryl-alkyl BQS salts 3g-o; around 166 ppm in the amide and carboxylic BQS salts 3a-c, and around 205 ppm in the in the alkyl-alkyl BQS salts 3d-f. The C-11 methylene carbon atoms appear at high chemical shifts, between 66.0-59.3 ppm, in accordance with the electron-withdrawing effect of the positive α-endocyclic nitrogen atom from benzo[f]quinoline ring and with the structure of the ketone carbonyl group (alkyl or aryl or amide/aliphatic ester). The C-2 and C-4 atoms appear around 148 ppm and 143 ppm, respectively, in accordance with their position in the pyridine ring (α-or γendocyclic carbons). The remaining signals in the FT-IR, 1 H-NMR and 13 C-NMR spectra are also in accordance with the proposed structures.

Antimicrobial Assay
The sensitivity of the microorganisms to the substances under investigation, salts BQS and cycloadducts PBQC, was assessed based on the diameter of the inhibition zone, using the Kirby-Bauer agar disk diffusion method, adopted by the Clinical & Laboratory Standards Institute (CLSI M07-A11, 2018) [30]. The method uses Mueller Hinton nutrient agar medium for antibacterial tests and Sabouraud nutrient agar medium for antifungal tests. The disk diffusion test provides a number of advantages, such as simplicity of the method, low cost, large number of organisms and antimicrobial agents that can be tested, and ease of interpreting the resulting data [30,31]. The in vitro antibacterial activity was evaluated against the Gram-positive bacterium Staphylococcus aureus ATCC 25923 and the Gram-negative Escherichia coli ATCC 25922. The in vitro antifungal activity was evaluated against the fungus Candida albicans ATCC 10231. Penicillin (10 IU), carbenicillin (100 µg/mL) and nystatin (500,000 IU) were used as positive control (C+) for Staphylococcus aureus, Escherichia coli and Candida albicans, respectively. As negative control (C−) sterile filter paper disks (with no antimicrobial compounds) were used. The obtained results were expressed as diameters of inhibition zones (mm). The larger the diameter of the inhibition zones is, the more active the compounds are as antimicrobials and antifungals. The obtained results are listed in Table 2 (and Figures S1-S3, see the Supporting Material for details).
All values represented in the table are average of results of five separately conducted experiments. Bold and underline means very active, while underline means active. a Diameter of inhibition zone (mm); X ± SD, mean of five measurements ± standard deviation. C+: Penicillin 10 IU for Staphylococcus aureus, carbenicillin 100 µg/mL for Escherichia coli and nystatin 500,000 IU for Candida albicans.
The data presented in Table 2 and Figures S1-S3 reveal interesting data concerning the biological properties of salts BQS and cycloadducts PBQC. A first observation is the fact that the tested cycloadducts PBQC did not present any antibacterial and antifungal activity and for this reason we did not continue the assays for this category of benzo[f]quinoline derivatives. On the other hand, the tested BQS salts demonstrated a certain antibacterial and antifungal activity against the tested strains. Practically all the tested the BQS salts manifest a powerful activity against the fungus Candida albicans, with eight compounds being very active with a diameter of inhibition zone up to 20 mm (with two compounds, 3e and 3f, being remarkably active with inhibition zone diameters in the range of 30 mm) and higher that the control drug nystatin. The remaining other BQS salts are also active against C. albicans, having a diameter of inhibition zone between 15 mm (for 3k) and 19 mm (for 3m). Against the Gram positive Staphylococcus aureus bacterial strain, five BQS salts are very active, having an inhibition zone diameter of up to 20 mm (3c, 3f, 3h, 3i, 3n), while the remaining other BQS salts are active (with an inhibition zone diameter between 15 mm (for 3j, 3k, 3l) and 19.5 mm (for 3b)]. Against Gram negative Escherichia coli bacteria, the BQS salts have a moderate activity, six BQS salts being active, with a diameter of inhibition zone between 15.5 mm (for 3f) and 18.5 mm (for 3c).
In the next step of the antimicrobial assays, the minimum inhibitory concentration (MIC) of the ten most active BQS salts (namely 3b-e, 3g-k, 3n) were determined, using the standardized broth microdilution assay procedure [32][33][34][35]. The resulting MIC value is defined as the lowest concentration of the antimicrobial BQS salts under investigation, which prevents visible growth of the tested microorganism. The obtained results are listed in Table 3. Compared with the control drug, the data from Table 3 reveal that BQS salt 3i is active to a low concentration, having a MIC of 30.4 × 10 −4 µg/mL in the case of Staphylococcus aureus, 15.2 × 10 −4 µg/mL in the case of Escherichia coli and 575 × 10 −4 µg/mL in the case of Candida albicans. Significant results were also obtained for the BQS salts 3n (with a MIC of 975 × 10 −4 µg/mL for S. aureus and 0.195 µg/mL for E. coli and C. albicans), 3h and 3g (with a MIC in the range of 0.195 µg/mL for all germs). The diameter of inhibition zone data presented in Table 2, MIC data in Table 3  Docking was validated by redocking the ligands present in PDB target For this step ligands were cut from the crystallographic obtained structures c SDF files, minimized and charges corrected. Ligands were then docked on to and PDB structures. Results are depicted in Figure 2. Docking was validated by redocking the ligands present in PDB targ For this step ligands were cut from the crystallographic obtained structures SDF files, minimized and charges corrected. Ligands were then docked on t and PDB structures. Results are depicted in Figure 2.   3b  3c  3e  3d  3f  3g  3h  3i  3j  3k  3l  3m  3n  3o Total energy of the receptor-ligand complex (eV) BQS salts from 3a to 3o Docking was validated by redocking the ligands present in PDB target structures. For this step ligands were cut from the crystallographic obtained structures converted to SDF files, minimized and charges corrected. Ligands were then docked on to the free ligand PDB structures. Results are depicted in Figure 2.

Molecular Docking
During the last decades molecular docking became an important tool in medicinal chemistry, furnishing relevant information concerning binding sites, binding energy and Absorption, Distribution, Metabolism, Excretion, Toxicity (ADMET) properties. As a result, we decided to perform a computational (in silico) study concerning these parameters for the salts BQS 3. The chemical space was also screened for similar compounds with remarkable bioactivity and ADMET and drug-like properties. Bioactivity is studied with the aid of a computational build = receptor-ligand system. The complex is characterized by its energy, hydrogen bonds, and steric constraints. The targets (binding sites) used for this class of compounds were retrieved from the literature (mainly the work of Hunter et al. [36] and Xavier et al. [37]) and using an online service that suggests ligand predilection for a specific target [38] and are ATP synthase and topoisomerase II (TOPO II). Target structures were retrieved from the PDB database: 6WLZ for ATP synthase [39] and 3KSB for TOPO II [40]. The PDB structures were energetically minimized, charges corrected, names corrected, potential energy recomputed. Cofactors and Aa chains were kept, and ligands and water molecules were removed. BQS 3 salts were introduced computationally as SDF. Files that were energetically minimized and charges corrected. Binding site coordinates (Å) were retrieved from the literature and from an algorithm based on expanded Van der Walls charges [41]. The maximum number of cavities which was set to be detected was 5, corresponding to the binding site molecular theory [42]. The cavity with the most significant volume (Å 3 ) was chosen for each target (see Supporting Materials for details).
First, the two sets of compounds were docked against the two molecular targets. MOE 2009 software and its methodology were used in the docking procedure [43][44][45]. Cartesian coordinates for the two binding sites are as follows: ATP synthase: x45.94 Å, y46.91 Å, z198.20 Å; TOPO II: x-16.33 Å, y43.11 Å, z-34.50 Å. For computational and action mechanism reasons, only protein corona of ATP synthase was used in building the in-silico system (Figure 1).
Docking was validated by redocking the ligands present in PDB target structures. For this step ligands were cut from the crystallographic obtained structures converted to SDF files, minimized and charges corrected. Ligands were then docked on to the free ligand PDB structures. Results are depicted in Figure 2.  In Figure 5 is presented the complex with ATP synthase of BQS salt 3f. In the binding site pocket we may notice two powerful H-π interactions between the two benzene rings from the benzo[f]quinoline moiety and the aminoacid Glu 62. Another H-π interaction is formed between Ala 356 and the benzene from the center of the aromatic core. Also, Ala 356 has interresidue contacts with Tyr 353 and Arg 357. A benzene carbon donates electrons to Phe 60, which has an inter-residue contact with Pro 227. All these interactions are stabilizing the BQS salt 3f-ATP synthase complex. The hydrogen bond energy between ligand and binding pocket aminoacids (eV) BQS salts from 3a to 3o In Figure 5 is presented the complex with ATP synthase of BQS salt 3f. In the binding site pocket we may notice two powerful H-π interactions between the two benzene rings from the benzo[f]quinoline moiety and the aminoacid Glu 62. Another H-π interaction is formed between Ala 356 and the benzene from the center of the aromatic core. Also, Ala 356 has interresidue contacts with Tyr 353 and Arg 357. A benzene carbon donates electrons to Phe 60, which has an inter-residue contact with Pro 227. All these interactions are stabilizing the BQS salt 3f-ATP synthase complex. site pocket we may notice two powerful H-π interactions between the two benzene ring from the benzo[f]quinoline moiety and the aminoacid Glu 62. Another H-π interaction formed between Ala 356 and the benzene from the center of the aromatic core. Also, A 356 has interresidue contacts with Tyr 353 and Arg 357. A benzene carbon donates ele trons to Phe 60, which has an inter-residue contact with Pro 227. All these interactions a stabilizing the BQS salt 3f-ATP synthase complex. In the next step, a similar protocol was followed to study the interactions of BQS sal with TOPO II. Figures 6 and 7, describe the interaction of each BQS 3a-o salt with TOP II in the formed complex, in terms of total energies of the complexes (ET) and hydroge bond energy (EH). In the next step, a similar protocol was followed to study the interactions of BQS salts with TOPO II. Figures 6 and 7, describe the interaction of each BQS 3a-o salt with TOPO II in the formed complex, in terms of total energies of the complexes (E T ) and hydrogen bond energy (E H ).   ; compounds 3a, 3c, 3e, 3f, 3g, 3h, 3i, 3n and 3o don't form hydrogen bonds.
From the obtained data we may notice that salts 3j and 3n have the best-fit in complex with TOPO II (in Figure 8 the complex of salt 3j is presented). In the binding pocket of TOPO II, a hydrogen bond is formed between the nitrogen atom of the benzo[f]quinoline ring and aminoacid ASP-510, stabilizing the salt BQS 3j-TOPO II complex. We also noticed that TYR 118 and His 76 interact with the aromatic rings. Also, a π bond is observed between His 76 (aromatic) and the positive nitrogen atom.   ; compounds 3a, 3c, 3e, 3f, 3g, 3h, 3i, 3n and 3o don't form hydrogen bonds.
From the obtained data we may notice that salts 3j and 3n have the best-fit in complex with TOPO II (in Figure 8 the complex of salt 3j is presented). In the binding pocket of TOPO II, a hydrogen bond is formed between the nitrogen atom of the benzo[f]quinoline ring and aminoacid ASP-510, stabilizing the salt BQS 3j-TOPO II complex. We also noticed that TYR 118 and His 76 interact with the aromatic rings. Also, a π bond is observed between His 76 (aromatic) and the positive nitrogen atom.  3a  3b  3c  3d  3e  3f  3g  3h  3i  3j  3k  3l  3m  3n  3o Hydrogen bond energy between ligand and binding pocket aminoacids (eV) BQS salts from 3a to 3o   Ciprofloxacin, the control molecule, when docked with ATP synthase, shows a total energy of the complex of −2.59 eV and a hydrogen bond energy of −8.31 eV. In a complex with TOPO II, ciprofloxacin shows a total energy of −2.56 eV and a hydrogen bonding energy of −2.037 eV (Figure 9). Ciprofloxacin, the control molecule, when docked with ATP synthase, shows a total energy of the complex of −2.59 eV and a hydrogen bond energy of −8.31 eV. In a complex with TOPO II, ciprofloxacin shows a total energy of −2.56 eV and a hydrogen bonding energy of −2.037 eV (Figure 9). The most active antimicrobial compounds 3i and 3n show a similar behaviour in the ATP synthase and TOPO II binding pockets as compound 3j (Figure 10). The most active antimicrobial compounds 3i and 3n show a similar behaviour in the ATP synthase and TOPO II binding pockets as compound 3j ( Figure 10).
Lastly, regarding our computational docking study, compound 3j displayed the best interaction energies. However, experimentally compounds 3i and 3n have the best bioactivities, while 3j has only moderate antimicrobial activity. In comparison with cyprofloxacin where the aromatic centres seem to play no role, in the case of BQS those aromatic centres are crucial in the interaction with the active sites. In the next step, the ADME properties of BQS salts 3a-o were studied (Table 4).  Lastly, regarding our computational docking study, compound 3j displayed the best interaction energies. However, experimentally compounds 3i and 3n have the best bioactivities, while 3j has only moderate antimicrobial activity. In comparison with cyprofloxacin where the aromatic centres seem to play no role, in the case of BQS those aromatic centres are crucial in the interaction with the active sites. In the next step, the ADME properties of BQS salts 3a-o were studied (Table 4). Analysis of the data from Table 4 reveals interesting data concerning the BQS salts' ADME properties. All salt derivatives show good aqueous solubility. This is indicated by two descriptors: QP logS (which describes the general aqueous solubility) and ClQP logS (which describes the conformation-independent aqueous solubility). The optimal values of these two descriptors are in the range between −6.5 and 0.5. The data from Table 4 indicate for our BQS salt 3a has good solubility. Three compounds (3j, 3m, and 3n) are exceptions being slightly out of limit.
The QP log HERG descriptor is a computational equivalent of zebrafish model toxicity in respect to the mode of action (MOA). This descriptor is linked with the potential K channel-blocking effect due to the electrophilic nature of drugs. The concerned values are below −5. As observed from Table 4 the majority of the synthesized compounds have predictable, convenient values, which means that our BQS quaternary salts 3a-o have low toxicity profiles. The reduced toxicity it is also confirmed by the Lipinski's rule of  Table 4 we can notice that only compounds 3j, 3m, and 3n present a violation of Lipinski's rule of 5 and Jorgensen's rule of 3, respectively. According with Lipinski's rule of five, the maximum number of violations for a computed compound is five (these five rules are: mol_MW < 500; QP log Po/w < 5; donor HB < 5; accpt. HB < 10) while for Jorgensen's rule of three, the maximum number of violations for a computed compound is three (QP log s > −5.7; QP PCaco > 22 nm/s; primary metabolites < 7). As a result, all our BQS salts 3a-o have excellent drug-like properties. However, in vitro testing should be performed in order to confirm these theoretical results.
The QP PCaco descriptor predicts permeability at the gut drug barrier. Predictions are for non-active transport. Values below 25 describe a low permeability while values higher than 500 are characteristic of excellent permeability. All described compounds show poor predicted gut barrier permeability.
QP Log BB descriptor predicts brain-blood partition coefficient for oral delivery. An optimal interval is between descriptor values of −3.0 to 1.2. All compounds are off the limit, meaning that potentially they may have poor blood-brain partition coefficients when administered orally.
QP PMDCK is a descriptor designed to simulate MDCK cell permeability, considered to be a good mimic of the blood-brain barrier. Values below 25 describe a poor permeability while values higher than 500 describe an excellent permeability. The data from Table 4 indicate that all BQS salts 3a-o has excellent blood-brain barrier permeability.
QP log KP is a descriptor that predicts skin permeability, with an optimal interval between −8.0 and −1.0. The data from Table 4 indicate that the compounds might not have good skin absorption.
Predicted chemical reactivity (descriptor Metab.) shows very high values for our BQS salts, suggesting a negative behavior as leads (the optimal values for Metab. descriptor are range between 1 to −8).
Interactions of compounds with serum albumin were explored computationally using the QP log Khsa descriptor. The optimal values for QP log Khsa descriptor are in the range between −1.5 to 15. The data from Table 4 indicate that all BQS salts 3a-o have a high potential of interacting with human serum albumin.
%Ab-sorbtion is a descriptor that predicts human oral absorption on a 0 to 100% scale. Values below 25% describe a poor oral absorption while value higher than 80% describe an excellent oral absorption. The data from Table 4 indicate that all BQS salts 3a-o have values higher than 80%, which suggest excellent oral absorption.

General Information
Reagents and solvents were purchased from commercial sources and used without further purification. The melting points (uncorrected) of compounds were recorded in open capillary tubes using a MEL-TEMP Electrothermal apparatus (Barnstead International, Dubuque, IA, USA). The nuclear magnetic resonance spectra were recorded on an Avance III 500 MHz spectrometer (Bruker, Vienna, Austria) operating at 500 MHz for 1 H and 125 MHz for 13 C. Chemical shifts were reported in delta (δ) units (ppm), relative to the residual peak of solvent (ref: DMSO, 1 H: 2.50 ppm; 13 C: 39.52 ppm), and coupling constants (J) in Hz. In the NMR spectra the multiplicity of signals was indicated using the abbreviations s = singlet, bs = broad singlet, d = doublet, ad = apparent doublet, add = apparent doublet of doublets, t = triplet, at = apparent triplet, td = triplet of doublets, atd = apparent triplet of doublets, q = quartet, m = multiplet. The IR spectra were recorded using a VERTEX 70 FTIR spectrometer (Bruker, Vienna, Austria) equipped with an ATR module. Thin layer chromatography (TLC) was performed on commercial silica gel plates (silica gel 60 F 254 plates, Merck, Darmstadt, Germany), the visualization being done using a UV lamp (λ max = 254 or 365 nm). The microanalysis results were in satisfactory agreement (C, ±0.15; all the tested compounds, the concentration used was 25 mg/mL. Following incubation at the optimal temperatures for bacteria and fungi, of 37 • C and 28 • C, respectively, for 24 h (bacteria) and 72 h (fungi), the diameters of the inhibition zones were measured using a ruler. The controls were prepared in the same growth conditions (i.e., C+: sterile filter paper disks impregnated with antibiotics inducing sensitivity in the organisms under investigation, namely penicillin 10 IU for Staphylococcus aureus, carbenicillin 100 µg/mL for Escherichia coli and nystatin 500,000 IU for Candida albicans, and C−: sterile filter paper disks with no antimicrobial compounds).

Broth Microdilution Method for Determining the Minimum Inhibitory Concentration (MIC)
The working technique involves the use of a 96-well microtiter plate (microdilution). In each well of the plate, 80 µL of growth medium MH, 10 µL of microbial inoculum (Staphylococcus aureus ATCC 25923, Escherichia coli ATCC 25922, or Candida albicans ATCC 10231) prepared in the same manner as in the diffusion test (i.e., by diluting the standardized microbial suspension adjusted to a 0.5 McFarland standard), and 100 µL of antimicrobial substance to be tested were transferred by pipetting, in different concentrations. For this purpose, double dilutions of the antimicrobial agent were made in the DMSO 3%, starting with the 25 µg/mL dilution (e.g., 12.5 µg/mL, 6.25 µg/mL, 3.12 µg/mL, 1.56 µg/mL, 0.78 µg/mL and so on). For each tested microorganism, a positive control C+ (containing 80 µL of MH growth medium, 10 µL of diluted microbial culture, 100 µL successive double dilutions of antibiotic) and a negative one C− (containing 80 µL of MH growth medium and 10 µL of diluted microbial culture) were prepared. Following the incubation of the microplates at 37 • C for 24 h (for Staphylococcus aureus ATCC 25923 and Escherichia coli ATCC 25922) and at 28 • C for 72 h (for Candida albicans ATCC 10231), 10 µL of resazurin were added in each well. The samples were incubated once again at the optimal temperature for each microorganism for one hour. The colour of the indicator turned from purple to pink. Resazurin is a colorimetric indicator for cell viability widely applied for monitoring cell proliferation. The redox dye, resazurin, enters the cytosol in the oxidized form (purple-blue) and is converted to the reduced form, resorufin (pink).

Conclusions
We report herein the design, synthesis, experimental and in silico evaluation of the antibacterial and antifungal activity of some new benzo[f]quinoline derivatives. Two classes of benzo[f]quinolinium derivatives (salts BQS and cycloadducts PBQC) were designed and obtained via a direct and facile two step procedure: quaternization followed by a cycloaddition reaction. The synthesized compounds were characterized by elemental and spectral analysis (FT-IR, 1 H-NMR, 13 C-NMR). The antifungal assays revealed that the BQS salts have an excellent quasi-nonselective antifungal activity against the fungus Candida albicans, some of them higher that the control drug nystatin. The antibacterial assay revealed that the BQS salts have a very good antibacterial activity against the Gram positive germ Staphylococcus aureus while the activity against the Gram negative germ Escherichia coli is negligible. The cycloadducts PBQC 4 are inactive. Analysis of the biological data reveals some interesting SAR correlations between the structures and their antimicrobial activity. The in silico studies furnished important data concerning the pharmacodynamics, pharmacokinetics and ADMET parameters of the BQS salts. Study of the de interaction of each BQS salt 3a-o with ATP synthase in the formed complex, reveal that salts 3j, 3i, and 3n have the best-fit in complex with ATP synthase. Study of the de interaction of each BQS salt 3a-o with TOPO II in the formed complex, revealed that salts 3j and 3n have the best-fit in complex with TOPO II. The in silico ADMET studies reveal that the BQS salts have excellent drug-like properties, low toxicity profiles, excellent blood-brain barrier permeability, an excellent oral absorption and a good solubility. Overall, the experimental and in silico studies indicated that compounds 3e and 3f (from the aliphatic series) and 3i, 3j and 3n (from the aromatic series), are promising leading drug candidates.
Author Contributions: Design, conception and writing were performed by G.Z. and I.I.M. Biological assay was performed by S.D. Molecular docking was performed by C.N.L. Synthesis, structure elucidation, biological data analysis and molecular docking interpretations were performed by all authors, which also reviewed and approved the final version. All authors have read and agreed to the published version of the manuscript.