Preparation and Hydro-Lipophilic Properties of Monosubstituted N -Aryl-4-hydroxyquinoline-3-carboxanilides †

: A series of twenty-two monosubstituted N -aryl-4-hydroxyquinoline-3-carboxanilides designed as dual anti-invasive agents was prepared and characterized. Lipophilicity signiﬁcantly affects biological activities of compounds and ADME properties; therefore, the lipo-hydrophilic properties of these 4-hydroxyquinoline-3-carboxanilides were investigated. All the derivatives were analyzed using reversed-phase high-performance liquid chromatography. The procedure was carried out under isocratic conditions with methanol as the organic modiﬁer in the mobile phase using an end-capped non-polar C18 stationary reversed-phase column. In this study, correlations between the logarithm of the capacity factor k and log P /Clog P values calculated using various methods are discussed, as well as the relationships between lipophilicity and chemical structure of the studied compounds.


Introduction
Many factors and parameters play an important role in the design and subsequent development of bioactive agents [1,2]. One of them is lipophilicity, which is among the most important of all investigated physicochemical properties, as it affects not only the ligand-target binding interaction, but also solubility and subsequent absorption (biological availability), binding to transporters, metabolism and excretion [3][4][5]. Lipophilicity is based on the distribution of a compound between two immiscible phases. It therefore represents the affinity of the compound to the lipophilic environment [6]. Lipophilicity can be expressed by the logarithm of the distribution coefficient log P or the distribution coefficient log D [5,7]. A number of methods have been developed to determine lipophilicity, which can be divided into experimental and computational [7,8]. The oldest and still frequently used experimental methods are chromatography, especially reversed-phase high-performance liquid chromatography (RP-HPLC) and reversed-phase thin-layer chromatography (RP-TLC), which can be used to determine a wide range of log P values [6,9,10].
Compounds that bind to multiple targets represent an innovative approach in designing anti-invasive compounds because they both prevent the emergence of resistant cells/pathogens and are able to destroy resistant cells/pathogens. Compounds based on quinoline scaffold (all azanaphthalenes) have a wide range of promising biological properties and can be considered privileged structures of multi-target agents [11][12][13]. Moreover, azanaphalene structures can be easily and rapidly synthesized, demonstrating the importance of these privileged structures. In addition, this simple scaffold has unique physicochemical properties and provides the possibility of a large number of modifications (through targeted-or diversity-oriented synthesis) and the preparation of many isomeric forms and bioisosteres. On the other hand, it is not easy to determine the exact mechanism of action of these compounds. For example, primaquine has celebrated more than 60 years of clinical application, but its mode of action has not been elucidated [14]. Hydroxyquinolines are known to be able to chelate not only iron (which is an essential nutrient), but also copper, manganese, magnesium, zinc and other vital metals [15]. Further research has led to the discovery that the mechanisms of action of these compounds are actually more complex. In addition to their bidentate properties causing metal chelation, substituted quinolines show different mechanisms of action, e.g., they inhibit mycobacterial gyrase, ATP synthase, FtsZ protein, glutathione S-transferase, enoyl-ACP reductase, decaprenylphosphoryl-β-D-ribose-2 -epimerase (DprE1) or FadD32 [16][17][18][19][20][21][22][23][24].

Results and Discussion
All studied compounds 1-8c were prepared according to Scheme 1 using modified microwave-assisted (MW) synthesis [29,30]. Briefly: in dry chlorobenzene, the carboxyl group was activated with phosphorus chloride, and then the resulting acyl chloride was aminolyzed with a ring-substituted aniline. All the crude target compounds (see Table 1) were recrystallized from ethanol. quinoline scaffold (all azanaphthalenes) have a wide range of promising biological properties and can be considered privileged structures of multi-target agents [11][12][13]. Moreover, azanaphalene structures can be easily and rapidly synthesized, demonstrating the importance of these privileged structures. In addition, this simple scaffold has unique physicochemical properties and provides the possibility of a large number of modifications (through targeted-or diversity-oriented synthesis) and the preparation of many isomeric forms and bioisosteres. On the other hand, it is not easy to determine the exact mechanism of action of these compounds. For example, primaquine has celebrated more than 60 years of clinical application, but its mode of action has not been elucidated [14]. Hydroxyquinolines are known to be able to chelate not only iron (which is an essential nutrient), but also copper, manganese, magnesium, zinc and other vital metals [15]. Further research has led to the discovery that the mechanisms of action of these compounds are actually more complex. In addition to their bidentate properties causing metal chelation, substituted quinolines show different mechanisms of action, e.g., they inhibit mycobacterial gyrase, ATP synthase, FtsZ protein, glutathione S-transferase, enoyl-ACP reductase, decaprenylphosphoryl-β-D-ribose-2′-epimerase (DprE1) or FadD32 [16][17][18][19][20][21][22][23][24]. Following on from previous ADMET studies dealing with (aza)naphthalenes [25][26][27][28][29][30][31][32][33][34][35][36][37][38], this contribution is devoted to the synthesis and structure-lipophilicity relationships of a series of monosubstituted anilides prepared from 4-hydroxyquinoline-3-carboxylic acid.

Results and Discussion
All studied compounds 1-8c were prepared according to Scheme 1 using modified microwave-assisted (MW) synthesis [29,30]. Briefly: in dry chlorobenzene, the carboxyl group was activated with phosphorus chloride, and then the resulting acyl chloride was aminolyzed with a ring-substituted aniline. All the crude target compounds (see Table 1) were recrystallized from ethanol. The lipophilicity of the studied compounds was determined using RP-HPLC as capacity factors k with subsequent calculation of log k. The retention times of individual compounds were determined under isocratic conditions with methanol as an organic modifier in the mobile phase using end-capped non-polar C18 stationary RP columns. In addition, the lipophilicities (log P/Clog P data) of all target anilides were calculated using two commercially available programs: ACD/Percepta ver. 2012, and ChemBioDraw Ultra 13.0. All results are shown in Table 1.
Log P and Clog P calculations in ChemBioDraw software are based on the fragment method, whereby the log P calculation algorithm in this software neglects the position of the substituents and therefore calculates the same log P values for the individual triplets of positional isomers (a/b/c). The values are shown only in Table 1 without other discussion. According to the Clog P algorithm, which also includes possible chemical interactions of the molecule, lipophilicity values were the same only for meta-and para-isomers. Thus, only the log P values calculated by ACD/Percepta are unique for each isomer except for the methyl-substituted derivatives 3a-c, where the software predicted log P = 4.50 for all three isomers.
Correlations between the experimentally determined values of log k and the predicted values of log P (ACD/Percepta) and Clog P (ChemBioDraw) are shown in Figures  1-3, with the ortho-, meta-and para-isomers separately illustrated for greater clarity and explanatory value. The lipophilicity of the studied compounds was determined using RP-HPLC as capacity factors k with subsequent calculation of log k. The retention times of individual compounds were determined under isocratic conditions with methanol as an organic modifier in the mobile phase using end-capped non-polar C18 stationary RP columns. In addition, the lipophilicities (log P/Clog P data) of all target anilides were calculated using two commercially available programs: ACD/Percepta ver. 2012, and ChemBioDraw Ultra 13.0. All results are shown in Table 1.
Log P and Clog P calculations in ChemBioDraw software are based on the fragment method, whereby the log P calculation algorithm in this software neglects the position of the substituents and therefore calculates the same log P values for the individual triplets of positional isomers (a/b/c). The values are shown only in Table 1 without other discussion. According to the Clog P algorithm, which also includes possible chemical interactions of the molecule, lipophilicity values were the same only for metaand paraisomers. Thus, only the log P values calculated by ACD/Percepta are unique for each isomer except for the methyl-substituted derivatives 3a-c, where the software predicted log P = 4.50 for all three isomers.
Correlations between the experimentally determined values of log k and the predicted values of log P (ACD/Percepta) and Clog P (ChemBioDraw) are shown in Figures 1-3, with the ortho-, metaand para-isomers separately illustrated for greater clarity and explanatory value.
As can be seen from the individual graphs, the correlations between the experimental and calculated values are quite poor, especially for ortho-isomers. The highest agreement is for meta-derivatives and data calculated using ACD/Percepta, where the correlation coefficient is r = 0.9531 (n = 7), see Figure 2A. As above-mentioned, the ortho-substituted derivatives gave the worst correlations ( Figure 1). In addition, in graphs 1a and 1b (Figure 1), substituents capable of forming hydrogen bonds and/or other weak interactions with the aqueous environment or interactions within the molecule or with neighboring molecules are indicated. The spatially close the amide group, the hydroxyl group at C (4) and the quinoline nitrogen are of great importance for the overall poor correlation. The lipophilicity of the studied compounds was determined using RP-HPLC as capacity factors k with subsequent calculation of log k. The retention times of individual compounds were determined under isocratic conditions with methanol as an organic modifier in the mobile phase using end-capped non-polar C18 stationary RP columns. In addition, the lipophilicities (log P/Clog P data) of all target anilides were calculated using two commercially available programs: ACD/Percepta ver. 2012, and ChemBioDraw Ultra 13.0. All results are shown in Table 1.
Log P and Clog P calculations in ChemBioDraw software are based on the fragment method, whereby the log P calculation algorithm in this software neglects the position of the substituents and therefore calculates the same log P values for the individual triplets of positional isomers (a/b/c). The values are shown only in Table 1 without other discussion. According to the Clog P algorithm, which also includes possible chemical interactions of the molecule, lipophilicity values were the same only for meta-and para-isomers. Thus, only the log P values calculated by ACD/Percepta are unique for each isomer except for the methyl-substituted derivatives 3a-c, where the software predicted log P = 4.50 for all three isomers.
Correlations between the experimentally determined values of log k and the predicted values of log P (ACD/Percepta) and Clog P (ChemBioDraw) are shown in Figures  1-3, with the ortho-, meta-and para-isomers separately illustrated for greater clarity and explanatory value.  3a, 4a, 5a, 6a, 7a, 8a.   As can be seen from the individual graphs, the correlations between the experimental and calculated values are quite poor, especially for ortho-isomers. The highest agreement is for meta-derivatives and data calculated using ACD/Percepta, where the correlation coefficient is r = 0.9531 (n = 7), see Figure 2A. As above-mentioned, the ortho-substituted derivatives gave the worst correlations (Figure 1). In addition, in graphs 1a and 1b ( Figure  1), substituents capable of forming hydrogen bonds and/or other weak interactions with the aqueous environment or interactions within the molecule or with neighboring molecules are indicated. The spatially close the amide group, the hydroxyl group at C(4) and the quinoline nitrogen are of great importance for the overall poor correlation.
According to the experimental values, it can be concluded that 4-hydroxy-N-(2-nitrophenyl)quinoline-3-carboxamide (8a) is the least lipophilic, and 4-hydroxy-N-[4-(trifluoromethyl)phenyl]quinoline-3-carboxamide (7c) is the most lipophilic. In general, ortho-derivatives have the lowest log k values. The exception is the methoxy substituents, where the ortho-isomer 2a is the most lipophilic of the three. The meta-and para-derivatives in the triad mostly have close log k values, except for N-(4-bromophenyl)-4-hydroxyquinoline-3-carboxamide (6c), where there is a large "jump" between the log k values for the paraand meta-isomers. The order of compounds arranged according to increasing log k values is shown in Figure 4.
Regarding all these observations, it should be summarized that for these highly functionalized quinoline derivatives, standard commercially available lipophilicity calculation programs are unable to provide relevant data due to the high incidence of intra-and intermolecular interactions.  As can be seen from the individual graphs, the correlations between the experimental and calculated values are quite poor, especially for ortho-isomers. The highest agreement is for meta-derivatives and data calculated using ACD/Percepta, where the correlation coefficient is r = 0.9531 (n = 7), see Figure 2A. As above-mentioned, the ortho-substituted derivatives gave the worst correlations (Figure 1). In addition, in graphs 1a and 1b ( Figure  1), substituents capable of forming hydrogen bonds and/or other weak interactions with the aqueous environment or interactions within the molecule or with neighboring molecules are indicated. The spatially close the amide group, the hydroxyl group at C(4) and the quinoline nitrogen are of great importance for the overall poor correlation.
According to the experimental values, it can be concluded that 4-hydroxy-N-(2-nitrophenyl)quinoline-3-carboxamide (8a) is the least lipophilic, and 4-hydroxy-N-[4-(trifluoromethyl)phenyl]quinoline-3-carboxamide (7c) is the most lipophilic. In general, ortho-derivatives have the lowest log k values. The exception is the methoxy substituents, where the ortho-isomer 2a is the most lipophilic of the three. The meta-and para-derivatives in the triad mostly have close log k values, except for N-(4-bromophenyl)-4-hydroxyquinoline-3-carboxamide (6c), where there is a large "jump" between the log k values for the paraand meta-isomers. The order of compounds arranged according to increasing log k values is shown in Figure 4.
Regarding all these observations, it should be summarized that for these highly functionalized quinoline derivatives, standard commercially available lipophilicity calculation programs are unable to provide relevant data due to the high incidence of intra-and intermolecular interactions.  According to the experimental values, it can be concluded that 4-hydroxy-N-(2nitrophenyl)quinoline-3-carboxamide (8a) is the least lipophilic, and 4-hydroxy-N- [4-(trifluoromethyl)phenyl]quinoline-3-carboxamide (7c) is the most lipophilic. In general, ortho-derivatives have the lowest log k values. The exception is the methoxy substituents, where the ortho-isomer 2a is the most lipophilic of the three. The metaand para-derivatives in the triad mostly have close log k values, except for N-(4-bromophenyl)-4-hydroxyquinoline-3-carboxamide (6c), where there is a large "jump" between the log k values for the paraand meta-isomers. The order of compounds arranged according to increasing log k values is shown in Figure 4.
Regarding all these observations, it should be summarized that for these highly functionalized quinoline derivatives, standard commercially available lipophilicity calculation programs are unable to provide relevant data due to the high incidence of intra-and intermolecular interactions.

General Methods
All reagents were purchased from Merck (Sigma-Aldrich, St. Louis, MO, USA) and Alfa (Alfa-Aesar, Ward Hill, MA, USA). Microwave-assisted reactions were performed using a StartSYNTH microwave lab station (Milestone, Sorisole, BG, Italy). The melting points were determined on a Kofler hot-plate apparatus HMK (Franz Kustner Nacht KG, Dresden, Germany) and are uncorrected. Infrared (IR) spectra were recorded on an ATR diamond iD7 for Nicolet™ Impact 410 Fourier-transform IR spectrometer (Thermo Scientific, West Palm Beach, FL, USA). The spectra were obtained through the accumulation of 64 scans with a 2 cm −1 resolution in the region of 4000-650 cm −1 . All 1 H-and 13 C-NMR spectra were recorded on a JEOL ECZR 400 MHz NMR spectrometer (400 MHz for 1 H and 100 MHz for 13 C, Jeol, Tokyo, Japan) in dimethyl sulfoxide-d6 (DMSO-d6). 1 H and 13 C chemical shifts (δ) are reported in ppm. High-resolution mass spectra were measured using a high-performance liquid chromatograph Dionex UltiMate ® 3000 (Thermo Scientific, West Palm Beach, FL, USA) coupled with an LTQ Orbitrap XL TM Hybrid Ion Trap-Orbitrap Fourier-transform mass spectrometer (Thermo Scientific) equipped with a HESI II (heated electrospray ionization) source in the positive mode.

Synthesis
General Procedure for Synthesis of Carboxamides 1-8c 4-Hydroxyquinoline-3-carboxylic acid (0.5 g, 2.64 mM) was suspended in dry chlorobenzene (25 mL) at ambient temperature and phosphorus trichloride (0.12 mL, 1.32 mM, 0.5 eq.), and the corresponding substituted aniline (2.64 mM, 1 eq.) was added dropwise. The reaction mixture was transferred to the microwave reactor, where the synthesis was performed (1st phase: 10 min, 100 °C; 2nd phase: 15 min, 120 °C; 3rd phase: 20 min, 130 °C, 500 W). Then, the mixture was cooled to 60 °C, and the solvent was removed to dryness under reduced pressure. The residue was washed with hydrochloric acid and water. The crude product was recrystallized from diluted EtOH. All the studied compounds are presented in Table 1.

General Methods
All reagents were purchased from Merck (Sigma-Aldrich, St. Louis, MO, USA) and Alfa (Alfa-Aesar, Ward Hill, MA, USA). Microwave-assisted reactions were performed using a StartSYNTH microwave lab station (Milestone, Sorisole, BG, Italy). The melting points were determined on a Kofler hot-plate apparatus HMK (Franz Kustner Nacht KG, Dresden, Germany) and are uncorrected. Infrared (IR) spectra were recorded on an ATR diamond iD7 for Nicolet™ Impact 410 Fourier-transform IR spectrometer (Thermo Scientific, West Palm Beach, FL, USA). The spectra were obtained through the accumulation of 64 scans with a 2 cm −1 resolution in the region of 4000-650 cm −1 . All 1 H-and 13 C-NMR spectra were recorded on a JEOL ECZR 400 MHz NMR spectrometer (400 MHz for 1 H and 100 MHz for 13 C, Jeol, Tokyo, Japan) in dimethyl sulfoxide-d 6 (DMSO-d 6 ). 1 H and 13 C chemical shifts (δ) are reported in ppm. High-resolution mass spectra were measured using a high-performance liquid chromatograph Dionex UltiMate ® 3000 (Thermo Scientific, West Palm Beach, FL, USA) coupled with an LTQ Orbitrap XL TM Hybrid Ion Trap-Orbitrap Fourier-transform mass spectrometer (Thermo Scientific) equipped with a HESI II (heated electrospray ionization) source in the positive mode.

Synthesis
General Procedure for Synthesis of Carboxamides 1-8c 4-Hydroxyquinoline-3-carboxylic acid (0.5 g, 2.64 mM) was suspended in dry chlorobenzene (25 mL) at ambient temperature and phosphorus trichloride (0.12 mL, 1.32 mM, 0.5 eq.), and the corresponding substituted aniline (2.64 mM, 1 eq.) was added dropwise. The reaction mixture was transferred to the microwave reactor, where the synthesis was performed (1st phase: 10 min, 100 • C; 2nd phase: 15 min, 120 • C; 3rd phase: 20 min, 130 • C, 500 W). Then, the mixture was cooled to 60 • C, and the solvent was removed to dryness under reduced pressure. The residue was washed with hydrochloric acid and water. The crude product was recrystallized from diluted EtOH. All the studied compounds are presented in Table 1.   13