Structure-Based Design and in Silico Screening of Virtual Combinatorial Library of Benzamides Inhibiting 2-trans Enoyl-Acyl Carrier Protein Reductase of Mycobacterium tuberculosis with Favorable Predicted Pharmacokinetic Profiles

Background: During the previous decade a new class of benzamide-based inhibitors of 2-trans enoyl-acyl carrier protein reductase (InhA) of Mycobacterium tuberculosis (Mt) with unusual binding mode have emerged. Here we report in silico design and evaluation of novel benzamide InhA-Mt inhibitors with favorable predicted pharmacokinetic profiles. Methods: By using in situ modifications of the crystal structure of N-benzyl-4-((heteroaryl)methyl) benzamide (BHMB)-InhA complex (PDB entry 4QXM), 3D models of InhA-BHMBx complexes were prepared for a training set of 19 BHMBs with experimentally determined inhibitory potencies (half-maximal inhibitory concentrations IC50exp). In the search for active conformation of the BHMB1-19, linear QSAR model was prepared, which correlated computed gas phase enthalpies of formation (∆∆HMM) of InhA-BHMBx complexes with the IC50exp. Further, taking into account the solvent effect and entropy changes upon ligand, binding resulted in a superior QSAR model correlating computed complexation Gibbs free energies (∆∆Gcom). The successive pharmacophore model (PH4) generated from the active conformations of BHMBs served as a virtual screening tool of novel analogs included in a virtual combinatorial library (VCL) of compounds containing benzamide scaffolds. The VCL filtered by Lipinski’s rule-of-five was screened by the PH4 model to identify new BHMB analogs. Results: Gas phase QSAR model: −log10(IC50exp) = pIC50exp = −0.2465 × ∆∆HMM + 7.95503, R2 = 0.94; superior aqueous phase QSAR model: pIC50exp = −0.2370 × ∆∆Gcom + 7.8783, R2 = 0.97 and PH4 pharmacophore model: pIC50exp = 1.0013 × pIC50exp − 0.0085, R2 = 0.95. The VCL of more than 114 thousand BHMBs was filtered down to 73,565 analogs Lipinski’s rule. The five-point PH4 screening retained 90 new and potent BHMBs with predicted inhibitory potencies IC50pre up to 65 times lower than that of BHMB1 (IC50exp = 20 nM). Predicted pharmacokinetic profile of the new analogs showed enhanced cell membrane permeability and high human oral absorption compared to current anti-tuberculotics. Conclusions: Combined use of QSAR models that considered binding of the BHMBs to InhA, pharmacophore model, and ADME properties helped to recognize bound active conformation of the benzamide inhibitors, permitted in silico screening of VCL of compounds sharing benzamide scaffold and identification of new analogs with predicted high inhibitory potencies and favorable pharmacokinetic profiles.


Introduction
"United to end tuberculosis: an urgent global response to a global epidemic" is the title of the declaration adopted by the UN General Assembly on the fight against tuberculosis (TB), which reaffirmed the commitment to end the tuberculosis epidemic globally by 2030 [1]. The General Assembly acknowledged that the Millennium Development Goals [2] and associated strategies helped to reverse the trend of the tuberculosis epidemic and until 2016 reduced the tuberculosis mortality by 37%. Nonetheless, the current WHO year report [3,4] revealed alarming statistics showing that throughout the world about 3 TB deaths occurred and about 19 persons developed TB every minute. On the other hand, drug development against Mycobacterium tuberculosis (Mt), despite the increased occurrence of multiple drug resistant (MDR-TB) and extensively drug resistant (XDR-TB) strains, has reached a noticeable progress on inhibitor design against 2-trans enoyl-acyl carrier protein reductase (InhA), the most frequently addressed validated mycobacterial drug target. The division of the InhA substrate binding site into three structural subsites (site I, catalytic; site II, hydrophobic; site III, hydrophilic) and the determination of the catalytic residue Tyr158 conformation ('in' or 'out'), have paved the way for the improvement in structure-based design and development of novel InhA inhibitors [5]. Another important observation is connected with the presence of a previously unnoticed interaction pocket formed by the side chains of Phe41 and Arg43 at the InhA active site displaying Tyr158 'out' conformation [6]. Promising drug candidates, which target the isoniazid-resistant Mt, are direct InhA inhibitors that do not require KatG (Mycobacterium tuberculosis catalase-peroxidase) activation [7]. Recent useful structural information involving key binding site residues identified by site-directed mutations of the InhA gene revealed that these residues (except Ser94 and Tyr158) interact with the ligand mostly through hydrophobic contacts [8]. The long list of known InhA inhibitors may be divided into, on the one hand, class 1 scaffolds: triclosan derivatives (TCL) [9], diphenyl ether [10,11], pyrrolidine carboxamide (PCAM) [12], and aryl amide derivatives [13] with Tyr158 'in' conformation and typical stacking interaction with the Phe97 residue. On the other hand, class 2 scaffolds include methyl-thiazole derivatives [5], pyrazoles [14], benzamides [15] with Tyr158 'out' conformation and interaction with the Phe41 and Arg43 pocket instead of the stacking with Phe97. The 3D-QSAR pharmacophores (PH4) for InhA inhibition are available for class 1 TCL and PCAM inhibitors only [16,17] but not for the class 2 compounds. Figure 1A,B show various numbers of hydrophobic features (HYD) for the PH4 of TCL and PCAM. The third HYD feature of TCL PH4 suggests that a bulky group can fill large hydrophobic pocket (LHP, site II) delimited by residues Met155, Pro193, Ile215, Leu217, Leu218, and Trp222 as a major structural requirement for efficient InhA inhibition [18]. Indeed, the best substitutions on candidates with the Triclosan scaffold direct a nonpolar group containing an ethyl linker capped by phenyl (IC 50 exp = 21 nM) or pentyl group (IC 50 exp = 11 nM with removal of all Cl atoms) to this LHP. 3FNH [21], five key interactions with InhA: HB-Tyr158, π-π-NAD and hydrophobic contacts). (B) PH4 for pyrrolidine carboxamide (PCAM) derivatives displaying 2 HYD (light blue) and the mapping of the most active derivative synthesized (IC 50 = 390 nM [17], PDB: 4U0J [12], main interactions with InhA: HB-Tyr158, HB-NAD). (C) PH4 for the active site of InhA, depicted in ribbon, (PDB: 4DRE [22]) with one acceptor in green color, one donor in purple color, and 4 HYD features (cyan), for clarity the acceptor and donor spheres were removed. The LHP is enclosed by 2 residues in yellow color and labeled in green surrounding two HYDs.
The main objective of this work was to design novel potent N-benzyl-4-((heteroaryl)methyl) benzamides (BHMBs) based on a series of 19 (training set) plus 6 (validation set) nanomolar inhibitors with observed inhibitory potencies as low as IC 50 exp = 20 nM [23]. Starting with in situ modification of the crystal structure of InhA-BHMB 2 complex (PDB: 4QXM), we have elaborated a QSAR model which correlated Gibbs free energies of InhA-BHMB x complex formation with the potencies IC 50 exp and determined the active conformation of BHMBs bound at the active site of InhA of Mt (MM-PB complexation approach). Based on this active conformation we have formulated 3D QSAR pharmacophore of InhA inhibition (PH4). Large virtual library of compounds sharing the BHMB scaffold has been generated and in silico screened with the PH4. The screening yielded virtual hits that exhibited predicted inhibitory potencies IC 50 pre more than 60 times lower than the most active training set compound BHMB1. Several of the identified putative inhibitors displayed favorable ADME profiles. Moreover, a series of drugs currently used in clinical practice, which include the benzamide scaffold in their molecular structure, were assessed with our new PH4 for InhA inhibition.  [16], PDB: 3FNH [21], five key interactions with InhA: HB-Tyr158, π-π-NAD and hydrophobic contacts). (B) PH4 for pyrrolidine carboxamide (PCAM) derivatives displaying 2 HYD (light blue) and the mapping of the most active derivative synthesized (IC 50 = 390 nM [17], PDB: 4U0J [12], main interactions with InhA: HB-Tyr158, HB-NAD). (C) PH4 for the active site of InhA, depicted in ribbon, (PDB: 4DRE [22]) with one acceptor in green color, one donor in purple color, and 4 HYD features (cyan), for clarity the acceptor and donor spheres were removed. The LHP is enclosed by 2 residues in yellow color and labeled in green surrounding two HYDs.
The main objective of this work was to design novel potent N-benzyl-4-((heteroaryl)methyl) benzamides (BHMBs) based on a series of 19 (training set) plus 6 (validation set) nanomolar inhibitors with observed inhibitory potencies as low as IC 50 exp = 20 nM [23]. Starting with in situ modification of the crystal structure of InhA-BHMB 2 complex (PDB: 4QXM), we have elaborated a QSAR model which correlated Gibbs free energies of InhA-BHMB x complex formation with the potencies IC 50 exp and determined the active conformation of BHMBs bound at the active site of InhA of Mt (MM-PB complexation approach). Based on this active conformation we have formulated 3D QSAR pharmacophore of InhA inhibition (PH4). Large virtual library of compounds sharing the BHMB scaffold has been generated and in silico screened with the PH4. The screening yielded virtual hits that exhibited predicted inhibitory potencies IC 50 pre more than 60 times lower than the most active training set compound BHMB1. Several of the identified putative inhibitors displayed favorable ADME profiles. Moreover, a series of drugs currently used in clinical practice, which include the benzamide scaffold in their molecular structure, were assessed with our new PH4 for InhA inhibition. Top five approved drugs identified by the PH4 screening exhibited predicted potencies IC 50 pre ranging from 1.7 to 60 nM.  -19) and validation set (BHBV1-6) of InhA inhibitors [23] used in the preparation of QSAR models of inhibitor binding. The R 1 and R 2 groups are numbered in the first part of the Top five approved drugs identified by the PH4 screening exhibited predicted potencies IC 50 pre ranging from 1.7 to 60 nM.

Training and Validation Sets
The training set of 19 BHMBs and validation set of another 6 analogs (Table 1) were selected from a homogeneous series of InhA inhibitors with known experimentally determined inhibitory activities originating from a single laboratory [23]. The whole series was obtained by variations at two positions R 1 and R 2 of the phenyl ring and amide group as shown in Table 1. The experimental half-maximal inhibitory concentrations (20 ≤ IC 50 exp ≤ 5930 nM) [23] cover a sufficiently wide concentration range for building of a reliable QSAR model. The ratio between the sizes of training and validation sets remains a critical point of correct classification but is limited by the count of the set of homologous compounds available from the literature [24]. Table 1. Set (BHBM1- 19) and validation set (BHBV1-6) of InhA inhibitors [23] used in the preparation of QSAR models of inhibitor binding. The R 1 and R 2 groups are numbered in the first part of the Top five approved drugs identified by the PH4 screening exhibited predicted potencies IC50 pre ranging from 1.7 to 60 nM.

Training and Validation Sets
The training set of 19 BHMBs and validation set of another 6 analogs (Table 1) were selected from a homogeneous series of InhA inhibitors with known experimentally determined inhibitory activities originating from a single laboratory [23]. The whole series was obtained by variations at two positions R1 and R2 of the phenyl ring and amide group as shown in Table 1. The experimental half-maximal inhibitory concentrations (20 ≤ IC50 exp ≤ 5930 nM) [23] cover a sufficiently wide concentration range for building of a reliable QSAR model. The ratio between the sizes of training and validation sets remains a critical point of correct classification but is limited by the count of the set of homologous compounds available from the literature [24]. Table 1. Set (BHBM1- 19) and validation set (BHBV1-6) of InhA inhibitors [23] used in the preparation of QSAR models of inhibitor binding. The R1 and R2 groups are numbered in the first part of the Top five approved drugs identified by the PH4 screening exhibited predicted potencies IC50 pre ranging from 1.7 to 60 nM.

Training and Validation Sets
The training set of 19 BHMBs and validation set of another 6 analogs (Table 1) were selected from a homogeneous series of InhA inhibitors with known experimentally determined inhibitory activities originating from a single laboratory [23]. The whole series was obtained by variations at two positions R1 and R2 of the phenyl ring and amide group as shown in Table 1. The experimental half-maximal inhibitory concentrations (20 ≤ IC50 exp ≤ 5930 nM) [23] cover a sufficiently wide concentration range for building of a reliable QSAR model. The ratio between the sizes of training and validation sets remains a critical point of correct classification but is limited by the count of the set of homologous compounds available from the literature [24]. Table 1. Set (BHBM1- 19) and validation set (BHBV1-6) of InhA inhibitors [23] used in the preparation of QSAR models of inhibitor binding. The R1 and R2 groups are numbered in the first part of the Top five approved drugs identified by the PH4 screening exhibited predicted potencies IC50 pre ranging from 1.7 to 60 nM.

Training and Validation Sets
The training set of 19 BHMBs and validation set of another 6 analogs (Table 1) were selected from a homogeneous series of InhA inhibitors with known experimentally determined inhibitory activities originating from a single laboratory [23]. The whole series was obtained by variations at two positions R1 and R2 of the phenyl ring and amide group as shown in Table 1. The experimental half-maximal inhibitory concentrations (20 ≤ IC50 exp ≤ 5930 nM) [23] cover a sufficiently wide concentration range for building of a reliable QSAR model. The ratio between the sizes of training and validation sets remains a critical point of correct classification but is limited by the count of the set of homologous compounds available from the literature [24]. Table 1. Set (BHBM1- 19) and validation set (BHBV1-6) of InhA inhibitors [23] used in the preparation of QSAR models of inhibitor binding. The R1 and R2 groups are numbered in the first part of the Top five approved drugs identified by the PH4 screening exhibited predicted potencies IC50 pre ranging from 1.7 to 60 nM.

Training and Validation Sets
The training set of 19 BHMBs and validation set of another 6 analogs (Table 1) were selected from a homogeneous series of InhA inhibitors with known experimentally determined inhibitory activities originating from a single laboratory [23]. The whole series was obtained by variations at two positions R1 and R2 of the phenyl ring and amide group as shown in Table 1. The experimental half-maximal inhibitory concentrations (20 ≤ IC50 exp ≤ 5930 nM) [23] cover a sufficiently wide concentration range for building of a reliable QSAR model. The ratio between the sizes of training and validation sets remains a critical point of correct classification but is limited by the count of the set of homologous compounds available from the literature [24]. Table 1. Set (BHBM1- 19) and validation set (BHBV1-6) of InhA inhibitors [23] used in the preparation of QSAR models of inhibitor binding. The R1 and R2 groups are numbered in the first part of the Top five approved drugs identified by the PH4 screening exhibited predicted potencies IC50 pre ranging from 1.7 to 60 nM.

Training and Validation Sets
The training set of 19 BHMBs and validation set of another 6 analogs (Table 1) were selected from a homogeneous series of InhA inhibitors with known experimentally determined inhibitory activities originating from a single laboratory [23]. The whole series was obtained by variations at two positions R1 and R2 of the phenyl ring and amide group as shown in Table 1. The experimental half-maximal inhibitory concentrations (20 ≤ IC50 exp ≤ 5930 nM) [23] cover a sufficiently wide concentration range for building of a reliable QSAR model. The ratio between the sizes of training and validation sets remains a critical point of correct classification but is limited by the count of the set of homologous compounds available from the literature [24]. Top five approved drugs identified by the PH4 screening exhibited predicted potencies IC50 pre ranging from 1.7 to 60 nM.

Training and Validation Sets
The training set of 19 BHMBs and validation set of another 6 analogs (Table 1) were selected from a homogeneous series of InhA inhibitors with known experimentally determined inhibitory activities originating from a single laboratory [23]. The whole series was obtained by variations at two positions R1 and R2 of the phenyl ring and amide group as shown in Table 1. The experimental half-maximal inhibitory concentrations (20 ≤ IC50 exp ≤ 5930 nM) [23] cover a sufficiently wide concentration range for building of a reliable QSAR model. The ratio between the sizes of training and validation sets remains a critical point of correct classification but is limited by the count of the set of homologous compounds available from the literature [24]. Top five approved drugs identified by the PH4 screening exhibited predicted potencies IC50 pre ranging from 1.7 to 60 nM.

Training and Validation Sets
The training set of 19 BHMBs and validation set of another 6 analogs (Table 1) were selected from a homogeneous series of InhA inhibitors with known experimentally determined inhibitory activities originating from a single laboratory [23]. The whole series was obtained by variations at two positions R1 and R2 of the phenyl ring and amide group as shown in Table 1. The experimental half-maximal inhibitory concentrations (20 ≤ IC50 exp ≤ 5930 nM) [23] cover a sufficiently wide concentration range for building of a reliable QSAR model. The ratio between the sizes of training and validation sets remains a critical point of correct classification but is limited by the count of the set of homologous compounds available from the literature [24]. Top five approved drugs identified by the PH4 screening exhibited predicted potencies IC50 pre ranging from 1.7 to 60 nM.

Training and Validation Sets
The training set of 19 BHMBs and validation set of another 6 analogs (Table 1) were selected from a homogeneous series of InhA inhibitors with known experimentally determined inhibitory activities originating from a single laboratory [23]. The whole series was obtained by variations at two positions R1 and R2 of the phenyl ring and amide group as shown in Table 1. The experimental half-maximal inhibitory concentrations (20 ≤ IC50 exp ≤ 5930 nM) [23] cover a sufficiently wide concentration range for building of a reliable QSAR model. The ratio between the sizes of training and validation sets remains a critical point of correct classification but is limited by the count of the set of homologous compounds available from the literature [24]. Top five approved drugs identified by the PH4 screening exhibited predicted potencies IC50 pre ranging from 1.7 to 60 nM.

Training and Validation Sets
The training set of 19 BHMBs and validation set of another 6 analogs (Table 1) were selected from a homogeneous series of InhA inhibitors with known experimentally determined inhibitory activities originating from a single laboratory [23]. The whole series was obtained by variations at two positions R1 and R2 of the phenyl ring and amide group as shown in Table 1. The experimental half-maximal inhibitory concentrations (20 ≤ IC50 exp ≤ 5930 nM) [23] cover a sufficiently wide concentration range for building of a reliable QSAR model. The ratio between the sizes of training and validation sets remains a critical point of correct classification but is limited by the count of the set of homologous compounds available from the literature [24]. Top five approved drugs identified by the PH4 screening exhibited predicted potencies IC50 pre ranging from 1.7 to 60 nM.

Training and Validation Sets
The training set of 19 BHMBs and validation set of another 6 analogs (Table 1) were selected from a homogeneous series of InhA inhibitors with known experimentally determined inhibitory activities originating from a single laboratory [23]. The whole series was obtained by variations at two positions R1 and R2 of the phenyl ring and amide group as shown in Table 1. The experimental half-maximal inhibitory concentrations (20 ≤ IC50 exp ≤ 5930 nM) [23] cover a sufficiently wide concentration range for building of a reliable QSAR model. The ratio between the sizes of training and validation sets remains a critical point of correct classification but is limited by the count of the set of homologous compounds available from the literature [24]. Top five approved drugs identified by the PH4 screening exhibited predicted potencies IC50 pre ranging from 1.7 to 60 nM.

Training and Validation Sets
The training set of 19 BHMBs and validation set of another 6 analogs (Table 1) were selected from a homogeneous series of InhA inhibitors with known experimentally determined inhibitory activities originating from a single laboratory [23]. The whole series was obtained by variations at two positions R1 and R2 of the phenyl ring and amide group as shown in Table 1. The experimental half-maximal inhibitory concentrations (20 ≤ IC50 exp ≤ 5930 nM) [23] cover a sufficiently wide concentration range for building of a reliable QSAR model. The ratio between the sizes of training and validation sets remains a critical point of correct classification but is limited by the count of the set of homologous compounds available from the literature [24]. Top five approved drugs identified by the PH4 screening exhibited predicted potencies IC50 pre ranging from 1.7 to 60 nM.

Training and Validation Sets
The training set of 19 BHMBs and validation set of another 6 analogs (Table 1) were selected from a homogeneous series of InhA inhibitors with known experimentally determined inhibitory activities originating from a single laboratory [23]. The whole series was obtained by variations at two positions R1 and R2 of the phenyl ring and amide group as shown in Table 1. The experimental half-maximal inhibitory concentrations (20 ≤ IC50 exp ≤ 5930 nM) [23] cover a sufficiently wide concentration range for building of a reliable QSAR model. The ratio between the sizes of training and validation sets remains a critical point of correct classification but is limited by the count of the set of homologous compounds available from the literature [24]. Top five approved drugs identified by the PH4 screening exhibited predicted potencies IC50 pre ranging from 1.7 to 60 nM.

Training and Validation Sets
The training set of 19 BHMBs and validation set of another 6 analogs (Table 1) were selected from a homogeneous series of InhA inhibitors with known experimentally determined inhibitory activities originating from a single laboratory [23]. The whole series was obtained by variations at two positions R1 and R2 of the phenyl ring and amide group as shown in Table 1. The experimental half-maximal inhibitory concentrations (20 ≤ IC50 exp ≤ 5930 nM) [23] cover a sufficiently wide concentration range for building of a reliable QSAR model. The ratio between the sizes of training and validation sets remains a critical point of correct classification but is limited by the count of the set of homologous compounds available from the literature [24]. Top five approved drugs identified by the PH4 screening exhibited predicted potencies IC50 pre ranging from 1.7 to 60 nM.

Training and Validation Sets
The training set of 19 BHMBs and validation set of another 6 analogs (Table 1) were selected from a homogeneous series of InhA inhibitors with known experimentally determined inhibitory activities originating from a single laboratory [23]. The whole series was obtained by variations at two positions R1 and R2 of the phenyl ring and amide group as shown in Table 1. The experimental half-maximal inhibitory concentrations (20 ≤ IC50 exp ≤ 5930 nM) [23] cover a sufficiently wide concentration range for building of a reliable QSAR model. The ratio between the sizes of training and validation sets remains a critical point of correct classification but is limited by the count of the set of homologous compounds available from the literature [24]. Top five approved drugs identified by the PH4 screening exhibited predicted potencies IC50 pre ranging from 1.7 to 60 nM.

Training and Validation Sets
The training set of 19 BHMBs and validation set of another 6 analogs (Table 1) were selected from a homogeneous series of InhA inhibitors with known experimentally determined inhibitory activities originating from a single laboratory [23]. The whole series was obtained by variations at two positions R1 and R2 of the phenyl ring and amide group as shown in Table 1. The experimental half-maximal inhibitory concentrations (20 ≤ IC50 exp ≤ 5930 nM) [23] cover a sufficiently wide concentration range for building of a reliable QSAR model. The ratio between the sizes of training and validation sets remains a critical point of correct classification but is limited by the count of the set of homologous compounds available from the literature [24]. Top five approved drugs identified by the PH4 screening exhibited predicted potencies IC50 pre ranging from 1.7 to 60 nM.

Training and Validation Sets
The training set of 19 BHMBs and validation set of another 6 analogs (Table 1) were selected from a homogeneous series of InhA inhibitors with known experimentally determined inhibitory activities originating from a single laboratory [23]. The whole series was obtained by variations at two positions R1 and R2 of the phenyl ring and amide group as shown in Table 1. The experimental half-maximal inhibitory concentrations (20 ≤ IC50 exp ≤ 5930 nM) [23] cover a sufficiently wide concentration range for building of a reliable QSAR model. The ratio between the sizes of training and validation sets remains a critical point of correct classification but is limited by the count of the set of homologous compounds available from the literature [24]. Top five approved drugs identified by the PH4 screening exhibited predicted potencies IC50 pre ranging from 1.7 to 60 nM.

Training and Validation Sets
The training set of 19 BHMBs and validation set of another 6 analogs (Table 1) were selected from a homogeneous series of InhA inhibitors with known experimentally determined inhibitory activities originating from a single laboratory [23]. The whole series was obtained by variations at two positions R1 and R2 of the phenyl ring and amide group as shown in Table 1. The experimental half-maximal inhibitory concentrations (20 ≤ IC50 exp ≤ 5930 nM) [23] cover a sufficiently wide concentration range for building of a reliable QSAR model. The ratio between the sizes of training and validation sets remains a critical point of correct classification but is limited by the count of the set of homologous compounds available from the literature [24]. Top five approved drugs identified by the PH4 screening exhibited predicted potencies IC50 pre ranging from 1.7 to 60 nM.

Training and Validation Sets
The training set of 19 BHMBs and validation set of another 6 analogs (Table 1) were selected from a homogeneous series of InhA inhibitors with known experimentally determined inhibitory activities originating from a single laboratory [23]. The whole series was obtained by variations at two positions R1 and R2 of the phenyl ring and amide group as shown in Table 1. The experimental half-maximal inhibitory concentrations (20 ≤ IC50 exp ≤ 5930 nM) [23] cover a sufficiently wide concentration range for building of a reliable QSAR model. The ratio between the sizes of training and validation sets remains a critical point of correct classification but is limited by the count of the set of homologous compounds available from the literature [24]. Top five approved drugs identified by the PH4 screening exhibited predicted potencies IC50 pre ranging from 1.7 to 60 nM.

Training and Validation Sets
The training set of 19 BHMBs and validation set of another 6 analogs (Table 1) were selected from a homogeneous series of InhA inhibitors with known experimentally determined inhibitory activities originating from a single laboratory [23]. The whole series was obtained by variations at two positions R1 and R2 of the phenyl ring and amide group as shown in Table 1. The experimental half-maximal inhibitory concentrations (20 ≤ IC50 exp ≤ 5930 nM) [23] cover a sufficiently wide concentration range for building of a reliable QSAR model. The ratio between the sizes of training and validation sets remains a critical point of correct classification but is limited by the count of the set of homologous compounds available from the literature [24]. Top five approved drugs identified by the PH4 screening exhibited predicted potencies IC50 pre ranging from 1.7 to 60 nM.

Training and Validation Sets
The training set of 19 BHMBs and validation set of another 6 analogs (Table 1) were selected from a homogeneous series of InhA inhibitors with known experimentally determined inhibitory activities originating from a single laboratory [23]. The whole series was obtained by variations at two positions R1 and R2 of the phenyl ring and amide group as shown in Table 1. The experimental half-maximal inhibitory concentrations (20 ≤ IC50 exp ≤ 5930 nM) [23] cover a sufficiently wide concentration range for building of a reliable QSAR model. The ratio between the sizes of training and validation sets remains a critical point of correct classification but is limited by the count of the set of homologous compounds available from the literature [24]. Top five approved drugs identified by the PH4 screening exhibited predicted potencies IC50 pre ranging from 1.7 to 60 nM.

Training and Validation Sets
The training set of 19 BHMBs and validation set of another 6 analogs (Table 1) were selected from a homogeneous series of InhA inhibitors with known experimentally determined inhibitory activities originating from a single laboratory [23]. The whole series was obtained by variations at two positions R1 and R2 of the phenyl ring and amide group as shown in Table 1. The experimental half-maximal inhibitory concentrations (20 ≤ IC50 exp ≤ 5930 nM) [23] cover a sufficiently wide concentration range for building of a reliable QSAR model. The ratio between the sizes of training and validation sets remains a critical point of correct classification but is limited by the count of the set of homologous compounds available from the literature [24]. Top five approved drugs identified by the PH4 screening exhibited predicted potencies IC50 pre ranging from 1.7 to 60 nM.

Training and Validation Sets
The training set of 19 BHMBs and validation set of another 6 analogs (Table 1) were selected from a homogeneous series of InhA inhibitors with known experimentally determined inhibitory activities originating from a single laboratory [23]. The whole series was obtained by variations at two positions R1 and R2 of the phenyl ring and amide group as shown in Table 1. The experimental half-maximal inhibitory concentrations (20 ≤ IC50 exp ≤ 5930 nM) [23] cover a sufficiently wide concentration range for building of a reliable QSAR model. The ratio between the sizes of training and validation sets remains a critical point of correct classification but is limited by the count of the set of homologous compounds available from the literature [24]. Top five approved drugs identified by the PH4 screening exhibited predicted potencies IC50 pre ranging from 1.7 to 60 nM.

Training and Validation Sets
The training set of 19 BHMBs and validation set of another 6 analogs (Table 1) were selected from a homogeneous series of InhA inhibitors with known experimentally determined inhibitory activities originating from a single laboratory [23]. The whole series was obtained by variations at two positions R1 and R2 of the phenyl ring and amide group as shown in Table 1. The experimental half-maximal inhibitory concentrations (20 ≤ IC50 exp ≤ 5930 nM) [23] cover a sufficiently wide concentration range for building of a reliable QSAR model. The ratio between the sizes of training and validation sets remains a critical point of correct classification but is limited by the count of the set of homologous compounds available from the literature [24]. Top five approved drugs identified by the PH4 screening exhibited predicted potencies IC50 pre ranging from 1.7 to 60 nM.

Training and Validation Sets
The training set of 19 BHMBs and validation set of another 6 analogs (Table 1) were selected from a homogeneous series of InhA inhibitors with known experimentally determined inhibitory activities originating from a single laboratory [23]. The whole series was obtained by variations at two positions R1 and R2 of the phenyl ring and amide group as shown in Table 1. The experimental half-maximal inhibitory concentrations (20 ≤ IC50 exp ≤ 5930 nM) [23] cover a sufficiently wide concentration range for building of a reliable QSAR model. The ratio between the sizes of training and validation sets remains a critical point of correct classification but is limited by the count of the set of homologous compounds available from the literature [24]. Top five approved drugs identified by the PH4 screening exhibited predicted potencies IC50 pre ranging from 1.7 to 60 nM.

Training and Validation Sets
The training set of 19 BHMBs and validation set of another 6 analogs (Table 1) were selected from a homogeneous series of InhA inhibitors with known experimentally determined inhibitory activities originating from a single laboratory [23]. The whole series was obtained by variations at two positions R1 and R2 of the phenyl ring and amide group as shown in Table 1. The experimental half-maximal inhibitory concentrations (20 ≤ IC50 exp ≤ 5930 nM) [23] cover a sufficiently wide concentration range for building of a reliable QSAR model. The ratio between the sizes of training and validation sets remains a critical point of correct classification but is limited by the count of the set of homologous compounds available from the literature [24]. Each of the 19 training sets (TS) and 6 validation sets (VS) InhA-BHMBx complexes (Table 1), was prepared by in situ modification of the refined template crystal structure (PDB entry code 4QXM [23]) of the complex InhA-BHMB2 as described in the Methods section. Further, the relative Gibbs free energy of the InhA-BHMBx complex formation (∆∆G com ) was computed for each of the 25 optimized enzyme-inhibitor complexes. Table 2 lists computed values of ∆∆G com and its components as defined in Equation (7), for the TS and VS of benzamides [23]. The QSAR model explained variation in the BHMBs experimental inhibitory potencies (pIC 50 exp = −log 10 (IC 50 exp ) [23]) by correlating it with computed GFE ∆∆G com through linear regression (Equation (8), Table 2). In addition, significant correlation obtained in this QSAR relationship permitted to determine the active bound conformation of the BHMBs at the InhA binding site and enabled definition of the PH4 pharmacophore. In search for a better insight into the binding affinity of BHBMs towards MtInhA, we have analyzed the enthalpy of complexation in gas phase ∆∆H MM by correlating it with the pIC 50 exp . The validity of this linear correlation (for statistical data of the regression see Table 3, Equation A) allowed assessment of the significance of inhibitor-enzyme interactions (∆∆H MM ) when solvent effect and loss of entropy of the inhibitor upon binding to the enzyme were neglected. This correlation explained about 94% of the pIC 50 exp data variation and underlined the role of the enthalpic contribution to the binding affinity of the ligand. Similarly, the more advanced descriptor, namely the GFE of the InhA-BHMBx complex formation including all components: ∆∆H MM , ∆∆TS vib and ∆∆G sol , has been assessed (for statistical data see Table 3, Equation B). Relatively high values of the regression coefficient R 2 , leave-one-out cross-validated regression coefficient R 2 xv and Fischer F-test of the correlation suggest strong relationship between the 3D model of inhibitor binding and the observed inhibitory potencies of the BHBMs [23]. Therefore, structural information derived from the 3D models of InhA-BHMBx complexes can be expected to lead to reliable prediction of InhA inhibitory potencies for new BHMBs analogs based on the QSAR model B, Table 3.  Table 1; b M w is the molar mass of inhibitors; c ∆∆H MM is the relative enthalpic contribution to the GFE change related to E-I complex formation derived by MM; 50 exp is the experimental half-maximal inhibition concentration of InhA obtained from ref. [23]; h ratio of predicted and experimental half-maximal inhibition concentrations pIC 50 pre /pIC 50 exp (pIC 50 pre = −log 10 IC 50 pre ) was predicted from computed ∆∆G com using the regression equation for InhA shown in Table 3, B. The statistical data confirmed validity of the correlation Equations (A) and (B) plotted on Figure 2. The ratio pIC 50 pre /pIC 50 exp 1 (the pIC 50 pre values were estimated using correlation Equation B, Table 3) calculated for the validation set BHMV1-6 documents the substantial predictive power of the complexation QSAR model from Table 2. Thus, the regression Equation B (Table 3) and computed ∆∆G com GFEs can be used for prediction of inhibitory potencies IC 50 pre against MtInhA for novel BHMB analogs, provided they share the same binding mode as the training set benzamides BHMB1-19.  Table 1; b Mw is the molar mass of inhibitors; c ΔΔH MM is the relative enthalpic contribution to the GFE change related to E-I complex formation derived by MM; 50 exp is the experimental half-maximal inhibition concentration of InhA obtained from ref. [23]; h ratio of predicted and experimental half-maximal inhibition concentrations pIC 50 pre /pIC 50 exp (pIC 50 pre = −log 10 IC 50 pre ) was predicted from computed ΔΔG com using the regression equation for InhA shown in Table 3, B. The statistical data confirmed validity of the correlation Equations (A) and (B) plotted on Figure   2. The ratio pIC 50 pre /pIC 50 exp ≅ 1 (the pIC 50 pre values were estimated using correlation Equation B, Table   3) calculated for the validation set BHMV1-6 documents the substantial predictive power of the complexation QSAR model from Table 2. Thus, the regression Equation B (Table 3) and computed ∆∆G com GFEs can be used for prediction of inhibitory potencies IC 50 pre against MtInhA for novel BHMB analogs, provided they share the same binding mode as the training set benzamides BHMB1-19.

Binding Mode of BHMBs
Structural information enzyme-inhibitor interactions retrieved from the crystal structure of InhA-BHMB1 complex [23] showed that BHMBs are InhA class 2 specific inhibitors. As indicated in Figure 3, in the catalytic site I residue Tyr158 adopted the 'out' conformation [5]; the pyridine moiety of the ligand (referred to as the "warhead" [6]) is in π-alkyl contact with Tyr158 and nicotinamide ring and π-donor with the hydroxyl group adjacent to the ribose of NADH. The central benzene ring of the inhibitor forms a recently observed π-π stacking interaction with Phe97 [6]. In the hydrophobic site II, the dihalide-substituted benzene ring sits in the hydrophobic substrate cavity, surrounded by side chains of predominantly nonpolar residues: Met103, Met98, Ile202, Leu207, Gly104, Phe149, Ala157, Tyr158, and Ile215 [6,23]. In the hydrophilic site III, the inhibitor makes hydrogen bonds with the sidechain of catalytic Met98, a key interaction reported recently [6]. These specific contacts were observed for the novel class of InhA inhibitors except the interaction with the "Phe41-Arg43" pocket previously unreported before Soutter et al. [6] and which will be discussed later.

Binding Mode of BHMBs
Structural information enzyme-inhibitor interactions retrieved from the crystal structure of InhA-BHMB1 complex [23] showed that BHMBs are InhA class 2 specific inhibitors. As indicated in Figure 3, in the catalytic site I residue Tyr158 adopted the 'out' conformation [5]; the pyridine moiety of the ligand (referred to as the "warhead" [6]) is in π-alkyl contact with Tyr158 and nicotinamide ring and π-donor with the hydroxyl group adjacent to the ribose of NADH. The central benzene ring of the inhibitor forms a recently observed π-π stacking interaction with Phe97 [6]. In the hydrophobic site II, the dihalide-substituted benzene ring sits in the hydrophobic substrate cavity, surrounded by side chains of predominantly nonpolar residues: Met103, Met98, Ile202, Leu207, Gly104, Phe149, Ala157, Tyr158, and Ile215 [6,23]. In the hydrophilic site III, the inhibitor makes hydrogen bonds with the sidechain of catalytic Met98, a key interaction reported recently [6]. These specific contacts were observed for the novel class of InhA inhibitors except the interaction with the "Phe41-Arg43" pocket previously unreported before Soutter et al. [6] and which will be discussed later.

Interaction Energy
Other key structural information was provided by the interaction energy (IE, ΔEint) diagram obtained for each training set inhibitor. IE breakdown to contributions from InhA active site residue is helpful for the choice of relevant R1-groups (site I) and R2-groups (site II) which could improve the binding affinity of BHMB analogs to the MtInhA and subsequently enhance the inhibitory potency. A comparative analysis of computed IE for training set BHMBs (Figure 4) divided into three classes (highest, moderate, and lowest activity) has been carried out to identify the residues for which the contribution to binding affinity could be increased. However, the comparative analysis showed about

Interaction Energy
Other key structural information was provided by the interaction energy (IE, ∆E int ) diagram obtained for each training set inhibitor. IE breakdown to contributions from InhA active site residue is helpful for the choice of relevant R 1 -groups (site I) and R 2 -groups (site II) which could improve the binding affinity of BHMB analogs to the MtInhA and subsequently enhance the inhibitory potency. A comparative analysis of computed IE for training set BHMBs (Figure 4) divided into three classes (highest, moderate, and lowest activity) has been carried out to identify the residues for which the contribution to binding affinity could be increased. However, the comparative analysis showed about the same level of IE contributions from active site residues for all three classes of inhibitors. Therefore, no suggestions of suitable substitutions able to improve the binding affinity as we previously reported for thymine-like inhibitors of Mt thymidine monophosphate kinase design could be proposed [25]. Since specific substitutions could not be proposed, we have adopted a combinatorial approach to novel BHMB analogs design and in silico screened a virtual library of 114921 BHMB analogs with help of the PH4 pharmacophore of InhA inhibition derived from the complexation QSAR model. As we can see from the IE analysis (Figure 4), the TS and VS benzamides [23] do not show significant interaction energies with the "Phe41-Arg43" pocket. the same level of IE contributions from active site residues for all three classes of inhibitors. Therefore, no suggestions of suitable substitutions able to improve the binding affinity as we previously reported for thymine-like inhibitors of Mt thymidine monophosphate kinase design could be proposed [25]. Since specific substitutions could not be proposed, we have adopted a combinatorial approach to novel BHMB analogs design and in silico screened a virtual library of 114921 BHMB analogs with help of the PH4 pharmacophore of InhA inhibition derived from the complexation QSAR model. As we can see from the IE analysis (Figure 4), the TS and VS benzamides [23] do not show significant interaction energies with the "Phe41-Arg43" pocket.  The interaction generation protocol in Discovery Studio molecular modeling program [26] provides the pharmacophore features of the active site of a protein. InhA predominantly displays hydrophobic features at the active site ( Figure 1C) as confirmed by previously reported works [16,17], the larger being the site II hydrophobic pocket that accommodates the substrate long alkyl chains. InhA substrate-competitive inhibitors design often exploits the pocket flexibility because of the high mobility of the Tyr158, Phe149 side chains and the substrate-binding loop (Thr196-Gly208) [27].  The interaction generation protocol in Discovery Studio molecular modeling program [26] provides the pharmacophore features of the active site of a protein. InhA predominantly displays hydrophobic features at the active site ( Figure 1C) as confirmed by previously reported works [16,17], the larger being the site II hydrophobic pocket that accommodates the substrate long alkyl chains.
InhA substrate-competitive inhibitors design often exploits the pocket flexibility because of the high mobility of the Tyr158, Phe149 side chains and the substrate-binding loop (Thr196-Gly208) [27].

Generation and Validation of 3D-QSAR Pharmacophore
InhA inhibition 3D-QSAR pharmacophore was generated from the active conformation of 19 TS BHMB1-19 and evaluated by 6 VS BHMV1-6 covering a large range of experimental activity (20-5930 nM) spanning more than two orders of magnitude. The generation process is divided into three main steps: (i) the constructive step, (ii) the subtractive step, and (iii) the optimization step [26] as described earlier [17]. During the constructive phase, BHMB1 alone was retained as the lead (since only the activity of BHMB1 fulfilled the threshold criterion, IC 50 exp ≤ 2 × 20 nM), and used to generate the starting PH4 features. In the subtractive phase, compounds for which IC 50 exp > 20 × 10 3.5 nM = 63246 nM were considered inactive. Accordingly, none of the training set BHMBx was inactive and no starting PH4 features were removed. Finally, during the optimization phase, the score of the pharmacophoric hypotheses was improved. Hypotheses were scored according to errors in activity estimates from regression and complexity via a simulated annealing approach. At the end of the optimization, the top scoring 10 unique pharmacophore hypotheses were kept, all displaying five-point features. The cost values, correlation coefficients, root-mean square deviation (RMSD) values, the pharmacophore features, and the max-fit value of the top 10 ranked hypotheses (Hypo1−Hypo10) are listed in Table 4. They were selected based on significant statistical parameters, such as high correlation coefficient, low total cost, and low RMSD. The generated pharmacophore models were then assessed for their reliability based on the calculated cost parameters ranging from 70.1 (Hypo1) to 184.7 (Hypo10). The relatively small gap between the highest and lowest cost parameter corresponds well with the homogeneity of the generated hypotheses and consistency of the TS of BHMBx. For this PH4 model, the fixed cost (45.4) is lower than the null cost (528.2) by a difference ∆ = 482.8. This difference is a major quality indicator of the PH4 predictability (∆ > 70 corresponds to an excellent chance or a probability higher than 90% that the model represents a true correlation [26]). To be statistically significant, a hypothesis has to be as close as possible to the fixed cost and as far as possible from the null cost. For the set of 10 hypotheses, the difference ∆ ≥ 343.5, which attests to the high quality of the pharmacophore model. The standard indicators such as the RMSD between the hypotheses ranged from 1.610 to 3.809, and the squared correlation coefficient (R 2 ) falls to an interval from 0.97 to 0.84. The first PH4 hypothesis with the closest cost (70.1) to the fixed one (45.4) and best RMSD and R 2 was retained for further analysis. The statistical data for the set of hypotheses (costs, RMSD, R 2 ) are listed in Table 4. The configuration cost (10.63 for all hypotheses) far below 17 confirms this pharmacophore as a reasonable one.
The link between the 98% significance and the number 49 scrambled runs of each hypothesis is based on the formula S = [1 − (1 + X)/Y] × 100, with X the total number of hypotheses having a total cost lower than the original hypothesis (Hypo 1) and Y the total number of HypoGen runs (initial + random runs): The evaluation of Hypo 1 was performed first through Fischer's randomization cross-validation test. The CatScramble program was used to randomize the experimental activities of the training set. At 98% confidence level, each of the 49 scramble runs created ten valid hypotheses, using the same features and parameters as in the generation of the original 10 pharmacophore hypotheses. Among them, the cost value of Hypo1 is the lowest compared with those of the 49 randomly generated hypotheses, as we can see in Table 4 where the lowest cost of the 49 random runs is listed for each original hypothesis, and none of them was as predictive as the original hypotheses generated shown in Table 4. Thus, there is a 98% probability that the best selected hypothesis Hypo1 represents a pharmacophore model for inhibitory activity of BHMBs with a similar level of predictive power as the complexation QSAR model, which relies on the BHMBx active conformation from 3D structures of the InhA-BHMBx complexes and computed GFE of enzyme-inhibitor binding ∆∆G com . Another evaluation of Hypo 1 is the mapping of the best active training set BHMB1 ( configuration cost (10.63 for all hypotheses) far below 17 confirms this pharmacophore as a reasonable one. The link between the 98% significance and the number 49 scrambled runs of each hypothesis is based on the formula S = [1 − (1 + X)/Y] × 100, with X the total number of hypotheses having a total cost lower than the original hypothesis (Hypo 1) and Y the total number of HypoGen runs (initial + random runs): The evaluation of Hypo 1 was performed first through Fischer's randomization cross-validation test. The CatScramble program was used to randomize the experimental activities of the training set. At 98% confidence level, each of the 49 scramble runs created ten valid hypotheses, using the same features and parameters as in the generation of the original 10 pharmacophore hypotheses. Among them, the cost value of Hypo1 is the lowest compared with those of the 49 randomly generated hypotheses, as we can see in Table 4 where the lowest cost of the 49 random runs is listed for each original hypothesis, and none of them was as predictive as the original hypotheses generated shown in Table 4. Thus, there is a 98% probability that the best selected hypothesis Hypo1 represents a pharmacophore model for inhibitory activity of BHMBs with a similar level of predictive power as the complexation QSAR model, which relies on the BHMBx active conformation from 3D structures of the InhA-BHMBx complexes and computed GFE of enzyme-inhibitor binding ΔΔG com. Another evaluation of Hypo 1 is the mapping of the best active training set BHMB1 (    We can carry out computational design and selection of new BHMB analogs with elevated inhibitory potencies against MtInhA, based on a strategy using the noticeable presence of the hydrophobic features included in the best pharmacophore model at the position of R2 coupled with mapping of R1 to the aromatic ring feature and the appropriate ring substitution to the hydrophobic aliphatic feature in Hypo1 ( Figure 5).

Virtual Screnning
In silico screening of a virtual (combinatorial) library can lead to hit identification as it was shown in our previous works on inhibitors design [17,28,29].

Virtual Library
An initial virtual library (VL) was generated by substitutions at positions for R 1 and R2 (see Table 5) on the benzamide scaffold. During the virtual library enumeration, all 339 R-groups listed in Table 5 were attached to in positions R1 and R2 of the BHMB scaffold to form a combinatorial library of the size: R1 × R2 = 339 × 339 = 114,921 analogs. This initial diversity library was generated from building blocks (chemicals) listed in the databases of available chemicals [30]. To design a more focused library of a reduced size and increased content of drug-like molecules, we have introduced a set of filters and penalties such as the Lipinski rule-of-five [31], which helped to select a smaller number of suitable BHMBs that could be submitted to in silico screening. This focusing has reduced the size of the initial library to 73,565 analogs, 64% of its original number size. We can carry out computational design and selection of new BHMB analogs with elevated inhibitory potencies against MtInhA, based on a strategy using the noticeable presence of the hydrophobic features included in the best pharmacophore model at the position of R 2 coupled with mapping of R 1 to the aromatic ring feature and the appropriate ring substitution to the hydrophobic aliphatic feature in Hypo1 ( Figure 5).

Virtual Screnning
In silico screening of a virtual (combinatorial) library can lead to hit identification as it was shown in our previous works on inhibitors design [17,28,29].

Virtual Library
An initial virtual library (VL) was generated by substitutions at positions for R 1 and R 2 (see Table 5) on the benzamide scaffold. During the virtual library enumeration, all 339 R-groups listed in Table 5 were attached to in positions R 1 and R 2 of the BHMB scaffold to form a combinatorial library of the size: R 1 × R 2 = 339 × 339 = 114,921 analogs. This initial diversity library was generated from building blocks (chemicals) listed in the databases of available chemicals [30]. To design a more focused library of a reduced size and increased content of drug-like molecules, we have introduced a set of filters and penalties such as the Lipinski rule-of-five [31], which helped to select a smaller number of suitable BHMBs that could be submitted to in silico screening. This focusing has reduced the size of the initial library to 73,565 analogs, 64% of its original number size.

In Silico Screening of Library of BHMBs
The focused library of 73,565 analogs was further screened for molecular structures matching the 3D-QSAR PH4 pharmacophore model Hypo1 of InhA inhibition. 238 BHMBs mapped to at least 2 pharmacophoric features, 90 of which mapped to at least 4 features of the pharmacophore. These best fitting analogs (PH4 hits) then underwent complexation QSAR model screening. The computed GFE of InhA-BHMBx complex formation, their components, and predicted half-maximal inhibitory concentrations IC 50 pre calculated from the correlation Equation B (Table 3) are listed in Table 6.

In Silico Screening of Library of BHMBs
The focused library of 73,565 analogs was further screened for molecular structures matching the 3D-QSAR PH4 pharmacophore model Hypo1 of InhA inhibition. 238 BHMBs mapped to at least 2 pharmacophoric features, 90 of which mapped to at least 4 features of the pharmacophore. These best fitting analogs (PH4 hits) then underwent complexation QSAR model screening. The computed GFE of InhA-BHMBx complex formation, their components, and predicted half-maximal inhibitory concentrations IC 50 pre calculated from the correlation Equation B (Table 3) are listed in Table 6.

Novel BHMB Analogs
The design of virtual library of novel analogs was guided by structural information retrieved from the BHMBx active conformation and was used for the selection of appropriate substituents (R 1 -and R 2 -groups). In order to identify which substituents lead to new inhibitor candidates with the highest predicted potencies towards the InhA of Mt, we have prepared histograms of the frequency of occurrence of R 1 -and R 2 -groups among the 90 best fit PH4 hits ( Figure 6). The histograms show that the R 1 groups 79, 310, and 318 were represented with the highest frequency of occurrence (5) among the 90 BHMB hits. The R 2 -groups most frequently represented in this subset are 282 (10) and 232, 233 (6) (1). Due to amino acid composition of the larger hydrophobic pocket, the R 2 -groups display preferences for larger aliphatic building blocks from butyl to octyl [5].
e ΔΔG com is the relative Gibbs free energy change related to the enzyme-inhibitor InhA-BHMB complex formation ΔΔG com ≅ ΔΔH MM + ΔΔG sol − ΔΔTS vib ; f IC 50 pre is the predicted inhibition potency towards MtInhA calculated from ΔΔG com using correlation Equation B, Table 3; g IC 50 exp [19] is given for the reference inhibitor BHMB1 instead of the IC 50 pre .

Novel BHMB Analogs
The design of virtual library of novel analogs was guided by structural information retrieved from the BHMBx active conformation and was used for the selection of appropriate substituents (R 1and R 2 -groups). In order to identify which substituents lead to new inhibitor candidates with the highest predicted potencies towards the InhA of Mt, we have prepared histograms of the frequency of occurrence of R 1 -and R 2 -groups among the 90 best fit PH4 hits ( Figure 6). The histograms show that the R 1 groups 79, 310, and 318 were represented with the highest frequency of occurrence (5) among the 90 BHMB hits. The R 2 -groups most frequently represented in this subset are 282 (10) (1). Due to amino acid composition of the larger hydrophobic pocket, the R 2 -groups display preferences for larger aliphatic building blocks from butyl to octyl [5].

Pharmacokinetic Profile of Novel BHMB Analogs
The pharmacokinetic profile of InhA inhibitors remains an important issue [7]. The best designed triclosan derivative with very low oral bioavailability due to its poor water solubility and rapid phase II metabolism has to be optimized for a possible use as antituberculotic and eventually antimalarial in case of high affinity toward PfEACP [32]. Among the ADME-related properties displayed in Table 7, such as octanol-water partitioning coefficient; aqueous solubility; blood-brain partition coefficient; Caco-2 cell permeability; serum protein binding; number of likely metabolic reactions; and another eighteen descriptors related to absorption, distribution, metabolism, and excretion (ADME) of the new analogs, were computed by the QikProp program [33] based on the method of Jorgensen [34,35]. Experimental data from more than 710 compounds including about 500 drugs and related heterocycles were used to produce regression equations correlating experimental and computed descriptors, resulting in an accurate prediction of pharmacokinetic properties of molecules. In line with the observed unfavorable pharmacokinetic exposure for the best active triclosan derivative, the predicted oral bioavailability of novel BHMB analogs ranges from 86% to 100%. Since a value greater than 80% is considered good, ten analogs among the best predicted 14 display a human oral absorption in gastrointestinal tract (HOA) of 100%. Drug likeness (#stars)-the number of property descriptors that fall outside the range of optimal values determined for 95% of known drugs out of 24 selected descriptors computed by the QikProp-was used as an additional ADME-related compound selection criterion. The values for the best active designed BHMBs are compared with those computed for drugs used for treatment of tuberculosis or currently undergoing clinical trials (Table 7). Our best designed analogs all display #stars equal to zero, meaning that the optimal value range of none of the drug-likeness descriptors was violated. Thus the designed BHMBs display a favorable pharmacokinetic profile.

Pharmacokinetic Profile of Novel BHMB Analogs
The pharmacokinetic profile of InhA inhibitors remains an important issue [7]. The best designed triclosan derivative with very low oral bioavailability due to its poor water solubility and rapid phase II metabolism has to be optimized for a possible use as antituberculotic and eventually antimalarial in case of high affinity toward Pf EACP [32]. Among the ADME-related properties displayed in Table 7, such as octanol-water partitioning coefficient; aqueous solubility; blood-brain partition coefficient; Caco-2 cell permeability; serum protein binding; number of likely metabolic reactions; and another eighteen descriptors related to absorption, distribution, metabolism, and excretion (ADME) of the new analogs, were computed by the QikProp program [33] based on the method of Jorgensen [34,35]. Experimental data from more than 710 compounds including about 500 drugs and related heterocycles were used to produce regression equations correlating experimental and computed descriptors, resulting in an accurate prediction of pharmacokinetic properties of molecules. In line with the observed unfavorable pharmacokinetic exposure for the best active triclosan derivative, the predicted oral bioavailability of novel BHMB analogs ranges from 86% to 100%. Since a value greater than 80% is considered good, ten analogs among the best predicted 14 display a human oral absorption in gastrointestinal tract (HOA) of 100%. Drug likeness (#stars)-the number of property descriptors that fall outside the range of optimal values determined for 95% of known drugs out of 24 selected descriptors computed by the QikProp-was used as an additional ADME-related compound selection criterion. The values for the best active designed BHMBs are compared with those computed for drugs used for treatment of tuberculosis or currently undergoing clinical trials (Table 7). Our best designed analogs all display #stars equal to zero, meaning that the optimal value range of none of the drug-likeness descriptors was violated. Thus the designed BHMBs display a favorable pharmacokinetic profile. Table 7. ADME-related properties of the best designed BHMB analogs and known antituberculotic agents either in clinical use or currently undergoing clinical testing computed by QikProp [33].  .0-20.0); j logarithm of partitioning coefficient between n-octanol and water phases (range for 95% of drugs: −2 to 6.5); k logarithm of predicted aqueous solubility, logS. S in [mol·dm-3] is the concentration of the solute in a saturated solution that is in equilibrium with the crystalline solid (range for 95% of drugs: −6.0 to 0.5); l logarithm of predicted binding constant to human serum albumin (range for 95% of drugs: −1.5 to 1.5); m logarithm of predicted brain/blood partition coefficient (range for 95% of drugs: −3.0 to 1.2); n predicted apparent Caco-2 cell membrane permeability in Boehringer-Ingelheim scale in [nm s-1] (range for 95% of drugs: < 25 poor, > 500 nm s −1 great); o number of likely metabolic reactions (range for 95% of drugs: 1-8); p predicted inhibition constants IC50pre. IC50pre was predicted from computed ∆∆Gcom using the regression Equation B shown in Table 3; q human oral absorption (1 = low, 2 = medium, 3 = high); r percentage of human oral absorption in gastrointestinal tract (<25% = poor, >80% = high); * star in any column indicates that the property descriptor value of the compound falls outside the range of values for 95% of known drugs.

Predicted InhA Inhibition Potency for Current Drugs Bearing Benzamide Scaffold
Since the benzamide scaffold has been analyzed in this study, drugs currently used in the clinical practice sharing this scaffold are worth evaluating with the help of our 3D-QSAR generated PH4 pharmacophore. The list of 24 compounds given in Table 8 is mostly indicated for treatment of neural disorders and cancers [36]. As we can see on Figure 9, the mapping of the five most potent predicted MtInhA inhibitors to PH4 pharmacophore sheds light on their affinity towards the enzyme and suggests their experimental evaluation as antituberculotics as they predominantly occupy the large hydrophobic pocket. The best predicted, Sultopride (IC 50 pre = 1.7 nM) is an approved drug, used in Japan, Hong Kong, and Europe for treatment of schizophrenia [36].

Predicted InhA Inhibition Potency for Current Drugs Bearing Benzamide Scaffold
Since the benzamide scaffold has been analyzed in this study, drugs currently used in the clinical practice sharing this scaffold are worth evaluating with the help of our 3D-QSAR generated PH4 pharmacophore. The list of 24 compounds given in Table 8 is mostly indicated for treatment of neural disorders and cancers [36]. As we can see on Figure 9, the mapping of the five most potent predicted MtInhA inhibitors to PH4 pharmacophore sheds light on their affinity towards the enzyme and suggests their experimental evaluation as antituberculotics as they predominantly occupy the large hydrophobic pocket. The best predicted, Sultopride (IC 50 pre = 1.7 nM) is an approved drug, used in Japan, Hong Kong, and Europe for treatment of schizophrenia [36].  Table 8. GFE, its components, and predicted InhA inhibitory potencies of 24 approved drugs which contain benzamide scaffold in their molecular structure (for 2D representation see Table 9).   Table 8. GFE, its components, and predicted InhA inhibitory potencies of 24 approved drugs which contain benzamide scaffold in their molecular structure (for 2D representation see Table 9).

DrugBank Accession Number
Name a M w [g·mol −

Binding Mode of BHMBs
To date, the most comprehensive study on InhA inhibition has been reported by Soutter et al. [6]. An important output of their DNA-encoded screening is the novel binding mode of inhibitors at the MtInhA active site, where the ligand, in addition to the typical stacking contact with Phe97 reaches a pocket surrounded by Phe41 and Arg43 side chains. This results in either a π-cation or a conventional hydrogen-bond contact. According to our analysis of the InhA-BHMBx complexes of most potent inhibitors, this interaction plays a key role in the significant improvement of predicted inhibitory potencies of novel benzamides. The enzyme-inhibitor interaction energy (Eint) breakdown to InhA active site residues contribution of best designed BHMB analogs ( Figure 10) shows that in contrast to the TS and VS inhibitors [23] (Figure 4), while the contribution of Phe97 to Eint is almost the same, the impact of interaction with polar Arg43 increased by at least 4 kcal·mol −1 . Surprisingly, the novel benzamide inhibitors display noticeable interaction with Phe97 and Arg43, suggesting that their length, which is comparable to that of TS BHMBs, is sufficient for concomitant interactions. Recently, site exclusion was observed for the ligands in complex with InhA (a pyrrolidine core and three substituents: benzoyl and 1-t-butoxy ethyl groups directly connected to the pyrrolidine core and a pyrazole benzaldehyde connected via the t-butoxy ethyl) benzoyl π-π stacks Phe97, methyl amide HB with Arg43), (PDB: 5G0W) [6]. As displayed in Figure 10, distance between the benzamide carbonyl oxygen and the Arg43 sidechain warhead carbon atom is relatively high for the training set compounds (7.5-9.5 Å). Interestingly, the plot of this distance vs. inhibitory activity pIC50 exp displays a correlation. This distance seems to be one of the key determinants for binding affinity for the class 2 InhA direct inhibitors instead of the existence of a HB. This structural specificity that emerged during the PH4 screening of the virtual library of benzamide analogs led to identification of new hits, the best of which are capable of forming a HB with the Arg43 residue.  Table 2. Table 9. Representation for 24 approved drugs containing benzamide scaffold listed in Table 8.

Binding Mode of BHMBs
To date, the most comprehensive study on InhA inhibition has been reported by Soutter et al. [6]. An important output of their DNA-encoded screening is the novel binding mode of inhibitors at the MtInhA active site, where the ligand, in addition to the typical stacking contact with Phe97 reaches a pocket surrounded by Phe41 and Arg43 side chains. This results in either a π-cation or a conventional hydrogen-bond contact. According to our analysis of the InhA-BHMBx complexes of most potent inhibitors, this interaction plays a key role in the significant improvement of predicted inhibitory potencies of novel benzamides. The enzyme-inhibitor interaction energy (Eint) breakdown to InhA active site residues contribution of best designed BHMB analogs ( Figure 10) shows that in contrast to the TS and VS inhibitors [23] (Figure 4), while the contribution of Phe97 to Eint is almost the same, the impact of interaction with polar Arg43 increased by at least 4 kcal·mol −1 . Surprisingly, the novel benzamide inhibitors display noticeable interaction with Phe97 and Arg43, suggesting that their length, which is comparable to that of TS BHMBs, is sufficient for concomitant interactions. Recently, site exclusion was observed for the ligands in complex with InhA (a pyrrolidine core and three substituents: benzoyl and 1-t-butoxy ethyl groups directly connected to the pyrrolidine core and a pyrazole benzaldehyde connected via the t-butoxy ethyl) benzoyl π-π stacks Phe97, methyl amide HB with Arg43), (PDB: 5G0W) [6]. As displayed in Figure 10, distance between the benzamide carbonyl oxygen and the Arg43 sidechain warhead carbon atom is relatively high for the training set compounds (7.5-9.5 Å). Interestingly, the plot of this distance vs. inhibitory activity pIC50 exp displays a correlation. This distance seems to be one of the key determinants for binding affinity for the class 2 InhA direct inhibitors instead of the existence of a HB. This structural specificity that emerged during the PH4 screening of the virtual library of benzamide analogs led to identification of new hits, the best of which are capable of forming a HB with the Arg43 residue.  Table 2. Table 9. Representation for 24 approved drugs containing benzamide scaffold listed in Table 8.

Binding Mode of BHMBs
To date, the most comprehensive study on InhA inhibition has been reported by Soutter et al. [6]. An important output of their DNA-encoded screening is the novel binding mode of inhibitors at the MtInhA active site, where the ligand, in addition to the typical stacking contact with Phe97 reaches a pocket surrounded by Phe41 and Arg43 side chains. This results in either a π-cation or a conventional hydrogen-bond contact. According to our analysis of the InhA-BHMBx complexes of most potent inhibitors, this interaction plays a key role in the significant improvement of predicted inhibitory potencies of novel benzamides. The enzyme-inhibitor interaction energy (Eint) breakdown to InhA active site residues contribution of best designed BHMB analogs ( Figure 10) shows that in contrast to the TS and VS inhibitors [23] (Figure 4), while the contribution of Phe97 to Eint is almost the same, the impact of interaction with polar Arg43 increased by at least 4 kcal·mol −1 . Surprisingly, the novel benzamide inhibitors display noticeable interaction with Phe97 and Arg43, suggesting that their length, which is comparable to that of TS BHMBs, is sufficient for concomitant interactions. Recently, site exclusion was observed for the ligands in complex with InhA (a pyrrolidine core and three substituents: benzoyl and 1-t-butoxy ethyl groups directly connected to the pyrrolidine core and a pyrazole benzaldehyde connected via the t-butoxy ethyl) benzoyl π-π stacks Phe97, methyl amide HB with Arg43), (PDB: 5G0W) [6]. As displayed in Figure 10, distance between the benzamide carbonyl oxygen and the Arg43 sidechain warhead carbon atom is relatively high for the training set compounds (7.5-9.5 Å). Interestingly, the plot of this distance vs. inhibitory activity pIC50 exp displays a correlation. This distance seems to be one of the key determinants for binding affinity for the class 2 InhA direct inhibitors instead of the existence of a HB. This structural specificity that emerged during the PH4 screening of the virtual library of benzamide analogs led to identification of new hits, the best of which are capable of forming a HB with the Arg43 residue.  Table 2. Table 9. Representation for 24 approved drugs containing benzamide scaffold listed in Table 8.   DB00345  DB00391  DB00409  DB00604  DB00619  DB01035  DB01168  DB01171   DB01233  DB01393  DB06288  DB06421  DB06422  DB06626  DB07069  DB08950   DB09018  DB11282  DB11480  DB12518  DB13025  DB13273  DB13523  DB15445 3. Discussions

Binding Mode of BHMBs
To date, the most comprehensive study on InhA inhibition has been reported by Soutter et al. [6]. An important output of their DNA-encoded screening is the novel binding mode of inhibitors at the MtInhA active site, where the ligand, in addition to the typical stacking contact with Phe97 reaches a pocket surrounded by Phe41 and Arg43 side chains. This results in either a π-cation or a conventional hydrogen-bond contact. According to our analysis of the InhA-BHMBx complexes of most potent inhibitors, this interaction plays a key role in the significant improvement of predicted inhibitory potencies of novel benzamides. The enzyme-inhibitor interaction energy (Eint) breakdown to InhA active site residues contribution of best designed BHMB analogs ( Figure 10) shows that in contrast to the TS and VS inhibitors [23] (Figure 4), while the contribution of Phe97 to Eint is almost the same, the impact of interaction with polar Arg43 increased by at least 4 kcal·mol −1 . Surprisingly, the novel benzamide inhibitors display noticeable interaction with Phe97 and Arg43, suggesting that their length, which is comparable to that of TS BHMBs, is sufficient for concomitant interactions. Recently, site exclusion was observed for the ligands in complex with InhA (a pyrrolidine core and three substituents: benzoyl and 1-t-butoxy ethyl groups directly connected to the pyrrolidine core and a pyrazole benzaldehyde connected via the t-butoxy ethyl) benzoyl π-π stacks Phe97, methyl amide HB with Arg43), (PDB: 5G0W) [6]. As displayed in Figure 10, distance between the benzamide carbonyl oxygen and the Arg43 sidechain warhead carbon atom is relatively high for the training set compounds (7.5-9.5 Å). Interestingly, the plot of this distance vs. inhibitory activity pIC50 exp displays a correlation. This distance seems to be one of the key determinants for binding affinity for the class 2 InhA direct inhibitors instead of the existence of a HB. This structural specificity that emerged during the PH4 screening of the virtual library of benzamide analogs led to identification of new hits, the best of which are capable of forming a HB with the Arg43 residue.  Table 2. Table 9. Representation for 24 approved drugs containing benzamide scaffold listed in Table 8.   DB00345  DB00391  DB00409  DB00604  DB00619  DB01035  DB01168  DB01171   DB01233  DB01393  DB06288  DB06421  DB06422  DB06626  DB07069  DB08950   DB09018  DB11282  DB11480  DB12518  DB13025  DB13273  DB13523  DB15445 3. Discussions

Binding Mode of BHMBs
To date, the most comprehensive study on InhA inhibition has been reported by Soutter et al. [6]. An important output of their DNA-encoded screening is the novel binding mode of inhibitors at the MtInhA active site, where the ligand, in addition to the typical stacking contact with Phe97 reaches a pocket surrounded by Phe41 and Arg43 side chains. This results in either a π-cation or a conventional hydrogen-bond contact. According to our analysis of the InhA-BHMBx complexes of most potent inhibitors, this interaction plays a key role in the significant improvement of predicted inhibitory potencies of novel benzamides. The enzyme-inhibitor interaction energy (Eint) breakdown to InhA active site residues contribution of best designed BHMB analogs ( Figure 10) shows that in contrast to the TS and VS inhibitors [23] (Figure 4), while the contribution of Phe97 to Eint is almost the same, the impact of interaction with polar Arg43 increased by at least 4 kcal·mol −1 . Surprisingly, the novel benzamide inhibitors display noticeable interaction with Phe97 and Arg43, suggesting that their length, which is comparable to that of TS BHMBs, is sufficient for concomitant interactions. Recently, site exclusion was observed for the ligands in complex with InhA (a pyrrolidine core and three substituents: benzoyl and 1-t-butoxy ethyl groups directly connected to the pyrrolidine core and a pyrazole benzaldehyde connected via the t-butoxy ethyl) benzoyl π-π stacks Phe97, methyl amide HB with Arg43), (PDB: 5G0W) [6]. As displayed in Figure 10, distance between the benzamide carbonyl oxygen and the Arg43 sidechain warhead carbon atom is relatively high for the training set compounds (7.5-9.5 Å). Interestingly, the plot of this distance vs. inhibitory activity pIC50 exp displays a correlation. This distance seems to be one of the key determinants for binding affinity for the class 2 InhA direct inhibitors instead of the existence of a HB. This structural specificity that emerged during the PH4 screening of the virtual library of benzamide analogs led to identification of new hits, the best of which are capable of forming a HB with the Arg43 residue.  Table 2. Table 9. Representation for 24 approved drugs containing benzamide scaffold listed in Table 8.   DB00345  DB00391  DB00409  DB00604  DB00619  DB01035  DB01168  DB01171   DB01233  DB01393  DB06288  DB06421  DB06422  DB06626  DB07069  DB08950   DB09018  DB11282  DB11480  DB12518  DB13025  DB13273  DB13523  DB15445 3. Discussions

Binding Mode of BHMBs
To date, the most comprehensive study on InhA inhibition has been reported by Soutter et al. [6]. An important output of their DNA-encoded screening is the novel binding mode of inhibitors at the MtInhA active site, where the ligand, in addition to the typical stacking contact with Phe97 reaches a pocket surrounded by Phe41 and Arg43 side chains. This results in either a π-cation or a conventional hydrogen-bond contact. According to our analysis of the InhA-BHMBx complexes of most potent inhibitors, this interaction plays a key role in the significant improvement of predicted inhibitory potencies of novel benzamides. The enzyme-inhibitor interaction energy (Eint) breakdown to InhA active site residues contribution of best designed BHMB analogs ( Figure 10) shows that in contrast to the TS and VS inhibitors [23] (Figure 4), while the contribution of Phe97 to Eint is almost the same, the impact of interaction with polar Arg43 increased by at least 4 kcal·mol −1 . Surprisingly, the novel benzamide inhibitors display noticeable interaction with Phe97 and Arg43, suggesting that their length, which is comparable to that of TS BHMBs, is sufficient for concomitant interactions. Recently, site exclusion was observed for the ligands in complex with InhA (a pyrrolidine core and three substituents: benzoyl and 1-t-butoxy ethyl groups directly connected to the pyrrolidine core and a pyrazole benzaldehyde connected via the t-butoxy ethyl) benzoyl π-π stacks Phe97, methyl amide HB with Arg43), (PDB: 5G0W) [6]. As displayed in Figure 10, distance between the benzamide carbonyl oxygen and the Arg43 sidechain warhead carbon atom is relatively high for the training set compounds (7.5-9.5 Å). Interestingly, the plot of this distance vs. inhibitory activity pIC50 exp displays a correlation. This distance seems to be one of the key determinants for binding affinity for the class 2 InhA direct inhibitors instead of the existence of a HB. This structural specificity that emerged during the PH4 screening of the virtual library of benzamide analogs led to identification of new hits, the best of which are capable of forming a HB with the Arg43 residue.  Table 2. Table 9. Representation for 24 approved drugs containing benzamide scaffold listed in Table 8.   DB00345  DB00391  DB00409  DB00604  DB00619  DB01035  DB01168  DB01171   DB01233  DB01393  DB06288  DB06421  DB06422  DB06626  DB07069  DB08950   DB09018  DB11282  DB11480  DB12518  DB13025  DB13273  DB13523  DB15445 3. Discussions

Binding Mode of BHMBs
To date, the most comprehensive study on InhA inhibition has been reported by Soutter et al. [6]. An important output of their DNA-encoded screening is the novel binding mode of inhibitors at the MtInhA active site, where the ligand, in addition to the typical stacking contact with Phe97 reaches a pocket surrounded by Phe41 and Arg43 side chains. This results in either a π-cation or a conventional hydrogen-bond contact. According to our analysis of the InhA-BHMBx complexes of most potent inhibitors, this interaction plays a key role in the significant improvement of predicted inhibitory potencies of novel benzamides. The enzyme-inhibitor interaction energy (Eint) breakdown to InhA active site residues contribution of best designed BHMB analogs ( Figure 10) shows that in contrast to the TS and VS inhibitors [23] (Figure 4), while the contribution of Phe97 to Eint is almost the same, the impact of interaction with polar Arg43 increased by at least 4 kcal·mol −1 . Surprisingly, the novel benzamide inhibitors display noticeable interaction with Phe97 and Arg43, suggesting that their length, which is comparable to that of TS BHMBs, is sufficient for concomitant interactions. Recently, site exclusion was observed for the ligands in complex with InhA (a pyrrolidine core and three substituents: benzoyl and 1-t-butoxy ethyl groups directly connected to the pyrrolidine core and a pyrazole benzaldehyde connected via the t-butoxy ethyl) benzoyl π-π stacks Phe97, methyl amide HB with Arg43), (PDB: 5G0W) [6]. As displayed in Figure 10, distance between the benzamide carbonyl oxygen and the Arg43 sidechain warhead carbon atom is relatively high for the training set compounds (7.5-9.5 Å). Interestingly, the plot of this distance vs. inhibitory activity pIC50 exp displays a correlation. This distance seems to be one of the key determinants for binding affinity for the class 2 InhA direct inhibitors instead of the existence of a HB. This structural specificity that emerged during the PH4 screening of the virtual library of benzamide analogs led to identification of new hits, the best of which are capable of forming a HB with the Arg43 residue.  Table 2. Table 9. Representation for 24 approved drugs containing benzamide scaffold listed in Table 8.   DB00345  DB00391  DB00409  DB00604  DB00619  DB01035  DB01168  DB01171   DB01233  DB01393  DB06288  DB06421  DB06422  DB06626  DB07069  DB08950   DB09018  DB11282  DB11480  DB12518  DB13025  DB13273  DB13523  DB15445 3. Discussions

Binding Mode of BHMBs
To date, the most comprehensive study on InhA inhibition has been reported by Soutter et al. [6]. An important output of their DNA-encoded screening is the novel binding mode of inhibitors at the MtInhA active site, where the ligand, in addition to the typical stacking contact with Phe97 reaches a pocket surrounded by Phe41 and Arg43 side chains. This results in either a π-cation or a conventional hydrogen-bond contact. According to our analysis of the InhA-BHMBx complexes of most potent inhibitors, this interaction plays a key role in the significant improvement of predicted inhibitory potencies of novel benzamides. The enzyme-inhibitor interaction energy (Eint) breakdown to InhA active site residues contribution of best designed BHMB analogs ( Figure 10) shows that in contrast to the TS and VS inhibitors [23] (Figure 4), while the contribution of Phe97 to Eint is almost the same, the impact of interaction with polar Arg43 increased by at least 4 kcal·mol −1 . Surprisingly, the novel benzamide inhibitors display noticeable interaction with Phe97 and Arg43, suggesting that their length, which is comparable to that of TS BHMBs, is sufficient for concomitant interactions. Recently, site exclusion was observed for the ligands in complex with InhA (a pyrrolidine core and three substituents: benzoyl and 1-t-butoxy ethyl groups directly connected to the pyrrolidine core and a pyrazole benzaldehyde connected via the t-butoxy ethyl) benzoyl π-π stacks Phe97, methyl amide HB with Arg43), (PDB: 5G0W) [6]. As displayed in Figure 10, distance between the benzamide carbonyl oxygen and the Arg43 sidechain warhead carbon atom is relatively high for the training set compounds (7.5-9.5 Å). Interestingly, the plot of this distance vs. inhibitory activity pIC50 exp displays a correlation. This distance seems to be one of the key determinants for binding affinity for the class 2 InhA direct inhibitors instead of the existence of a HB. This structural specificity that emerged during the PH4 screening of the virtual library of benzamide analogs led to identification of new hits, the best of which are capable of forming a HB with the Arg43 residue.  Table 2. Table 9. Representation for 24 approved drugs containing benzamide scaffold listed in Table 8.   DB00345  DB00391  DB00409  DB00604  DB00619  DB01035  DB01168  DB01171   DB01233  DB01393  DB06288  DB06421  DB06422  DB06626  DB07069  DB08950   DB09018  DB11282  DB11480  DB12518  DB13025  DB13273  DB13523  DB15445 3. Discussions

Binding Mode of BHMBs
To date, the most comprehensive study on InhA inhibition has been reported by Soutter et al. [6]. An important output of their DNA-encoded screening is the novel binding mode of inhibitors at the MtInhA active site, where the ligand, in addition to the typical stacking contact with Phe97 reaches a pocket surrounded by Phe41 and Arg43 side chains. This results in either a π-cation or a conventional hydrogen-bond contact. According to our analysis of the InhA-BHMBx complexes of most potent inhibitors, this interaction plays a key role in the significant improvement of predicted inhibitory potencies of novel benzamides. The enzyme-inhibitor interaction energy (Eint) breakdown to InhA active site residues contribution of best designed BHMB analogs ( Figure 10) shows that in contrast to the TS and VS inhibitors [23] (Figure 4), while the contribution of Phe97 to Eint is almost the same, the impact of interaction with polar Arg43 increased by at least 4 kcal·mol −1 . Surprisingly, the novel benzamide inhibitors display noticeable interaction with Phe97 and Arg43, suggesting that their length, which is comparable to that of TS BHMBs, is sufficient for concomitant interactions. Recently, site exclusion was observed for the ligands in complex with InhA (a pyrrolidine core and three substituents: benzoyl and 1-t-butoxy ethyl groups directly connected to the pyrrolidine core and a pyrazole benzaldehyde connected via the t-butoxy ethyl) benzoyl π-π stacks Phe97, methyl amide HB with Arg43), (PDB: 5G0W) [6]. As displayed in Figure 10, distance between the benzamide carbonyl oxygen and the Arg43 sidechain warhead carbon atom is relatively high for the training set compounds (7.5-9.5 Å). Interestingly, the plot of this distance vs. inhibitory activity pIC50 exp displays a correlation. This distance seems to be one of the key determinants for binding affinity for the class 2 InhA direct inhibitors instead of the existence of a HB. This structural specificity that emerged during the PH4 screening of the virtual library of benzamide analogs led to identification of new hits, the best of which are capable of forming a HB with the Arg43 residue.  Table 2. Table 9. Representation for 24 approved drugs containing benzamide scaffold listed in Table 8.   DB00345  DB00391  DB00409  DB00604  DB00619  DB01035  DB01168  DB01171   DB01233  DB01393  DB06288  DB06421  DB06422  DB06626  DB07069  DB08950   DB09018  DB11282  DB11480  DB12518  DB13025  DB13273  DB13523  DB15445 3. Discussions

Binding Mode of BHMBs
To date, the most comprehensive study on InhA inhibition has been reported by Soutter et al. [6]. An important output of their DNA-encoded screening is the novel binding mode of inhibitors at the MtInhA active site, where the ligand, in addition to the typical stacking contact with Phe97 reaches a pocket surrounded by Phe41 and Arg43 side chains. This results in either a π-cation or a conventional hydrogen-bond contact. According to our analysis of the InhA-BHMBx complexes of most potent inhibitors, this interaction plays a key role in the significant improvement of predicted inhibitory potencies of novel benzamides. The enzyme-inhibitor interaction energy (Eint) breakdown to InhA active site residues contribution of best designed BHMB analogs ( Figure 10) shows that in contrast to the TS and VS inhibitors [23] (Figure 4), while the contribution of Phe97 to Eint is almost the same, the impact of interaction with polar Arg43 increased by at least 4 kcal·mol −1 . Surprisingly, the novel benzamide inhibitors display noticeable interaction with Phe97 and Arg43, suggesting that their length, which is comparable to that of TS BHMBs, is sufficient for concomitant interactions. Recently, site exclusion was observed for the ligands in complex with InhA (a pyrrolidine core and three substituents: benzoyl and 1-t-butoxy ethyl groups directly connected to the pyrrolidine core and a pyrazole benzaldehyde connected via the t-butoxy ethyl) benzoyl π-π stacks Phe97, methyl amide HB with Arg43), (PDB: 5G0W) [6]. As displayed in Figure 10, distance between the benzamide carbonyl oxygen and the Arg43 sidechain warhead carbon atom is relatively high for the training set compounds (7.5-9.5 Å). Interestingly, the plot of this distance vs. inhibitory activity pIC50 exp displays a correlation. This distance seems to be one of the key determinants for binding affinity for the class 2 InhA direct inhibitors instead of the existence of a HB. This structural specificity that emerged during the PH4 screening of the virtual library of benzamide analogs led to identification of new hits, the best of which are capable of forming a HB with the Arg43 residue.  Table 2. Table 9. Representation for 24 approved drugs containing benzamide scaffold listed in Table 8.

Binding Mode of BHMBs
To date, the most comprehensive study on InhA inhibition has been reported by Soutter et al. [6]. An important output of their DNA-encoded screening is the novel binding mode of inhibitors at the MtInhA active site, where the ligand, in addition to the typical stacking contact with Phe97 reaches a pocket surrounded by Phe41 and Arg43 side chains. This results in either a π-cation or a conventional hydrogen-bond contact. According to our analysis of the InhA-BHMBx complexes of most potent inhibitors, this interaction plays a key role in the significant improvement of predicted inhibitory potencies of novel benzamides. The enzyme-inhibitor interaction energy (Eint) breakdown to InhA active site residues contribution of best designed BHMB analogs ( Figure 10) shows that in contrast to the TS and VS inhibitors [23] (Figure 4), while the contribution of Phe97 to Eint is almost the same, the impact of interaction with polar Arg43 increased by at least 4 kcal·mol −1 . Surprisingly, the novel benzamide inhibitors display noticeable interaction with Phe97 and Arg43, suggesting that their length, which is comparable to that of TS BHMBs, is sufficient for concomitant interactions. Recently, site exclusion was observed for the ligands in complex with InhA (a pyrrolidine core and three substituents: benzoyl and 1-t-butoxy ethyl groups directly connected to the pyrrolidine core and a pyrazole benzaldehyde connected via the t-butoxy ethyl) benzoyl π-π stacks Phe97, methyl amide HB with Arg43), (PDB: 5G0W) [6]. As displayed in Figure 10, distance between the benzamide carbonyl oxygen and the Arg43 sidechain warhead carbon atom is relatively high for the training set compounds (7.5-9.5 Å). Interestingly, the plot of this distance vs. inhibitory activity pIC50 exp displays a correlation. This distance seems to be one of the key determinants for binding affinity for the class 2 InhA direct inhibitors instead of the existence of a HB. This structural specificity that emerged during the PH4 screening of the virtual library of benzamide analogs led to identification of new hits, the best of which are capable of forming a HB with the Arg43 residue.  Table 2. Table 9. Representation for 24 approved drugs containing benzamide scaffold listed in Table 8.

Binding Mode of BHMBs
To date, the most comprehensive study on InhA inhibition has been reported by Soutter et al. [6]. An important output of their DNA-encoded screening is the novel binding mode of inhibitors at the MtInhA active site, where the ligand, in addition to the typical stacking contact with Phe97 reaches a pocket surrounded by Phe41 and Arg43 side chains. This results in either a π-cation or a conventional hydrogen-bond contact. According to our analysis of the InhA-BHMBx complexes of most potent inhibitors, this interaction plays a key role in the significant improvement of predicted inhibitory potencies of novel benzamides. The enzyme-inhibitor interaction energy (Eint) breakdown to InhA active site residues contribution of best designed BHMB analogs ( Figure 10) shows that in contrast to the TS and VS inhibitors [23] (Figure 4), while the contribution of Phe97 to Eint is almost the same, the impact of interaction with polar Arg43 increased by at least 4 kcal·mol −1 . Surprisingly, the novel benzamide inhibitors display noticeable interaction with Phe97 and Arg43, suggesting that their length, which is comparable to that of TS BHMBs, is sufficient for concomitant interactions. Recently, site exclusion was observed for the ligands in complex with InhA (a pyrrolidine core and three substituents: benzoyl and 1-t-butoxy ethyl groups directly connected to the pyrrolidine core and a pyrazole benzaldehyde connected via the t-butoxy ethyl) benzoyl π-π stacks Phe97, methyl amide HB with Arg43), (PDB: 5G0W) [6]. As displayed in Figure 10, distance between the benzamide carbonyl oxygen and the Arg43 sidechain warhead carbon atom is relatively high for the training set compounds (7.5-9.5 Å). Interestingly, the plot of this distance vs. inhibitory activity pIC50 exp displays a correlation. This distance seems to be one of the key determinants for binding affinity for the class 2 InhA direct inhibitors instead of the existence of a HB. This structural specificity that emerged during the PH4 screening of the virtual library of benzamide analogs led to identification of new hits, the best of which are capable of forming a HB with the Arg43 residue.  Table 2. Table 9. Representation for 24 approved drugs containing benzamide scaffold listed in Table 8.

Binding Mode of BHMBs
To date, the most comprehensive study on InhA inhibition has been reported by Soutter et al. [6]. An important output of their DNA-encoded screening is the novel binding mode of inhibitors at the MtInhA active site, where the ligand, in addition to the typical stacking contact with Phe97 reaches a pocket surrounded by Phe41 and Arg43 side chains. This results in either a π-cation or a conventional hydrogen-bond contact. According to our analysis of the InhA-BHMBx complexes of most potent inhibitors, this interaction plays a key role in the significant improvement of predicted inhibitory potencies of novel benzamides. The enzyme-inhibitor interaction energy (Eint) breakdown to InhA active site residues contribution of best designed BHMB analogs ( Figure 10) shows that in contrast to the TS and VS inhibitors [23] (Figure 4), while the contribution of Phe97 to Eint is almost the same, the impact of interaction with polar Arg43 increased by at least 4 kcal·mol −1 . Surprisingly, the novel benzamide inhibitors display noticeable interaction with Phe97 and Arg43, suggesting that their length, which is comparable to that of TS BHMBs, is sufficient for concomitant interactions. Recently, site exclusion was observed for the ligands in complex with InhA (a pyrrolidine core and three substituents: benzoyl and 1-t-butoxy ethyl groups directly connected to the pyrrolidine core and a pyrazole benzaldehyde connected via the t-butoxy ethyl) benzoyl π-π stacks Phe97, methyl amide HB with Arg43), (PDB: 5G0W) [6]. As displayed in Figure 10, distance between the benzamide carbonyl oxygen and the Arg43 sidechain warhead carbon atom is relatively high for the training set compounds (7.5-9.5 Å). Interestingly, the plot of this distance vs. inhibitory activity pIC50 exp displays a correlation. This distance seems to be one of the key determinants for binding affinity for the class 2 InhA direct inhibitors instead of the existence of a HB. This structural specificity that emerged during the PH4 screening of the virtual library of benzamide analogs led to identification of new hits, the best of which are capable of forming a HB with the Arg43 residue.  Table 2. Table 9. Representation for 24 approved drugs containing benzamide scaffold listed in Table 8.

Binding Mode of BHMBs
To date, the most comprehensive study on InhA inhibition has been reported by Soutter et al. [6]. An important output of their DNA-encoded screening is the novel binding mode of inhibitors at the MtInhA active site, where the ligand, in addition to the typical stacking contact with Phe97 reaches a pocket surrounded by Phe41 and Arg43 side chains. This results in either a π-cation or a conventional hydrogen-bond contact. According to our analysis of the InhA-BHMBx complexes of most potent inhibitors, this interaction plays a key role in the significant improvement of predicted inhibitory potencies of novel benzamides. The enzyme-inhibitor interaction energy (Eint) breakdown to InhA active site residues contribution of best designed BHMB analogs ( Figure 10) shows that in contrast to the TS and VS inhibitors [23] (Figure 4), while the contribution of Phe97 to Eint is almost the same, the impact of interaction with polar Arg43 increased by at least 4 kcal·mol −1 . Surprisingly, the novel benzamide inhibitors display noticeable interaction with Phe97 and Arg43, suggesting that their length, which is comparable to that of TS BHMBs, is sufficient for concomitant interactions. Recently, site exclusion was observed for the ligands in complex with InhA (a pyrrolidine core and three substituents: benzoyl and 1-t-butoxy ethyl groups directly connected to the pyrrolidine core and a pyrazole benzaldehyde connected via the t-butoxy ethyl) benzoyl π-π stacks Phe97, methyl amide HB with Arg43), (PDB: 5G0W) [6]. As displayed in Figure 10, distance between the benzamide carbonyl oxygen and the Arg43 sidechain warhead carbon atom is relatively high for the training set compounds (7.5-9.5 Å). Interestingly, the plot of this distance vs. inhibitory activity pIC50 exp displays a correlation. This distance seems to be one of the key determinants for binding affinity for the class 2 InhA direct inhibitors instead of the existence of a HB. This structural specificity that emerged during the PH4 screening of the virtual library of benzamide analogs led to identification of new hits, the best of which are capable of forming a HB with the Arg43 residue.  Table 2. Table 9. Representation for 24 approved drugs containing benzamide scaffold listed in Table 8.

Binding Mode of BHMBs
To date, the most comprehensive study on InhA inhibition has been reported by Soutter et al. [6]. An important output of their DNA-encoded screening is the novel binding mode of inhibitors at the MtInhA active site, where the ligand, in addition to the typical stacking contact with Phe97 reaches a pocket surrounded by Phe41 and Arg43 side chains. This results in either a π-cation or a conventional hydrogen-bond contact. According to our analysis of the InhA-BHMBx complexes of most potent inhibitors, this interaction plays a key role in the significant improvement of predicted inhibitory potencies of novel benzamides. The enzyme-inhibitor interaction energy (Eint) breakdown to InhA active site residues contribution of best designed BHMB analogs ( Figure 10) shows that in contrast to the TS and VS inhibitors [23] (Figure 4), while the contribution of Phe97 to Eint is almost the same, the impact of interaction with polar Arg43 increased by at least 4 kcal·mol −1 . Surprisingly, the novel benzamide inhibitors display noticeable interaction with Phe97 and Arg43, suggesting that their length, which is comparable to that of TS BHMBs, is sufficient for concomitant interactions. Recently, site exclusion was observed for the ligands in complex with InhA (a pyrrolidine core and three substituents: benzoyl and 1-t-butoxy ethyl groups directly connected to the pyrrolidine core and a pyrazole benzaldehyde connected via the t-butoxy ethyl) benzoyl π-π stacks Phe97, methyl amide HB with Arg43), (PDB: 5G0W) [6]. As displayed in Figure 10, distance between the benzamide carbonyl oxygen and the Arg43 sidechain warhead carbon atom is relatively high for the training set compounds (7.5-9.5 Å). Interestingly, the plot of this distance vs. inhibitory activity pIC50 exp displays a correlation. This distance seems to be one of the key determinants for binding affinity for the class 2 InhA direct inhibitors instead of the existence of a HB. This structural specificity that emerged during the PH4 screening of the virtual library of benzamide analogs led to identification of new hits, the best of which are capable of forming a HB with the Arg43 residue.  Table 2. Table 9. Representation for 24 approved drugs containing benzamide scaffold listed in Table 8.

Binding Mode of BHMBs
To date, the most comprehensive study on InhA inhibition has been reported by Soutter et al. [6]. An important output of their DNA-encoded screening is the novel binding mode of inhibitors at the MtInhA active site, where the ligand, in addition to the typical stacking contact with Phe97 reaches a pocket surrounded by Phe41 and Arg43 side chains. This results in either a π-cation or a conventional hydrogen-bond contact. According to our analysis of the InhA-BHMBx complexes of most potent inhibitors, this interaction plays a key role in the significant improvement of predicted inhibitory potencies of novel benzamides. The enzyme-inhibitor interaction energy (Eint) breakdown to InhA active site residues contribution of best designed BHMB analogs ( Figure 10) shows that in contrast to the TS and VS inhibitors [23] (Figure 4), while the contribution of Phe97 to Eint is almost the same, the impact of interaction with polar Arg43 increased by at least 4 kcal·mol −1 . Surprisingly, the novel benzamide inhibitors display noticeable interaction with Phe97 and Arg43, suggesting that their length, which is comparable to that of TS BHMBs, is sufficient for concomitant interactions. Recently, site exclusion was observed for the ligands in complex with InhA (a pyrrolidine core and three substituents: benzoyl and 1-t-butoxy ethyl groups directly connected to the pyrrolidine core and a pyrazole benzaldehyde connected via the t-butoxy ethyl) benzoyl π-π stacks Phe97, methyl amide HB with Arg43), (PDB: 5G0W) [6]. As displayed in Figure 10, distance between the benzamide carbonyl oxygen and the Arg43 sidechain warhead carbon atom is relatively high for the training set compounds (7.5-9.5 Å). Interestingly, the plot of this distance vs. inhibitory activity pIC50 exp displays a correlation. This distance seems to be one of the key determinants for binding affinity for the class 2 InhA direct inhibitors instead of the existence of a HB. This structural specificity that emerged during the PH4 screening of the virtual library of benzamide analogs led to identification of new hits, the best of which are capable of forming a HB with the Arg43 residue.

Binding Mode of BHMBs
To date, the most comprehensive study on InhA inhibition has been reported by Soutter et al. [6]. An important output of their DNA-encoded screening is the novel binding mode of inhibitors at the MtInhA active site, where the ligand, in addition to the typical stacking contact with Phe97 reaches a pocket surrounded by Phe41 and Arg43 side chains. This results in either a π-cation or a conventional hydrogen-bond contact. According to our analysis of the InhA-BHMBx complexes of most potent inhibitors, this interaction plays a key role in the significant improvement of predicted inhibitory potencies of novel benzamides. The enzyme-inhibitor interaction energy (Eint) breakdown to InhA active site residues contribution of best designed BHMB analogs ( Figure 10) shows that in contrast to the TS and VS inhibitors [23] (Figure 4), while the contribution of Phe97 to Eint is almost the same, the impact of interaction with polar Arg43 increased by at least 4 kcal·mol −1 . Surprisingly, the novel benzamide inhibitors display noticeable interaction with Phe97 and Arg43, suggesting that their length, which is comparable to that of TS BHMBs, is sufficient for concomitant interactions. Recently, site exclusion was observed for the ligands in complex with InhA (a pyrrolidine core and three substituents: benzoyl and 1-t-butoxy ethyl groups directly connected to the pyrrolidine core and a pyrazole benzaldehyde connected via the t-butoxy ethyl) benzoyl π-π stacks Phe97, methyl amide HB with Arg43), (PDB: 5G0W) [6]. As displayed in Figure 10, distance between the benzamide carbonyl oxygen and the Arg43 sidechain warhead carbon atom is relatively high for the training set compounds (7.5-9.5 Å). Interestingly, the plot of this distance vs. inhibitory activity pIC50 exp displays a correlation. This distance seems to be one of the key determinants for binding affinity for the class 2 InhA direct inhibitors instead of the existence of a HB. This structural specificity that emerged during the PH4 screening of the virtual library of benzamide analogs led to identification of new hits, the best of which are capable of forming a HB with the Arg43 residue.  Table 2. Table 9. Representation for 24 approved drugs containing benzamide scaffold listed in Table 8.   DB00345  DB00391  DB00409  DB00604  DB00619  DB01035  DB01168  DB01171   DB01233  DB01393  DB06288  DB06421  DB06422  DB06626  DB07069  DB08950   DB09018  DB11282  DB11480  DB12518  DB13025  DB13273  DB13523  DB15445 3. Discussions

Binding Mode of BHMBs
To date, the most comprehensive study on InhA inhibition has been reported by Soutter et al. [6]. An important output of their DNA-encoded screening is the novel binding mode of inhibitors at the MtInhA active site, where the ligand, in addition to the typical stacking contact with Phe97 reaches a pocket surrounded by Phe41 and Arg43 side chains. This results in either a π-cation or a conventional hydrogen-bond contact. According to our analysis of the InhA-BHMBx complexes of most potent inhibitors, this interaction plays a key role in the significant improvement of predicted inhibitory potencies of novel benzamides. The enzyme-inhibitor interaction energy (Eint) breakdown to InhA active site residues contribution of best designed BHMB analogs ( Figure 10) shows that in contrast to the TS and VS inhibitors [23] (Figure 4), while the contribution of Phe97 to Eint is almost the same, the impact of interaction with polar Arg43 increased by at least 4 kcal·mol −1 . Surprisingly, the novel benzamide inhibitors display noticeable interaction with Phe97 and Arg43, suggesting that their length, which is comparable to that of TS BHMBs, is sufficient for concomitant interactions. Recently, site exclusion was observed for the ligands in complex with InhA (a pyrrolidine core and three substituents: benzoyl and 1-t-butoxy ethyl groups directly connected to the pyrrolidine core and a pyrazole benzaldehyde connected via the t-butoxy ethyl) benzoyl π-π stacks Phe97, methyl amide HB with Arg43), (PDB: 5G0W) [6]. As displayed in Figure 10, distance between the benzamide carbonyl oxygen and the Arg43 sidechain warhead carbon atom is relatively high for the training set compounds (7.5-9.5 Å). Interestingly, the plot of this distance vs. inhibitory activity pIC50 exp displays a correlation. This distance seems to be one of the key determinants for binding affinity for the class 2 InhA direct inhibitors instead of the existence of a HB. This structural specificity that emerged during the PH4 screening of the virtual library of benzamide analogs led to identification of new hits, the best of which are capable of forming a HB with the Arg43 residue.  Table 2. Table 9. Representation for 24 approved drugs containing benzamide scaffold listed in Table 8.   DB00345  DB00391  DB00409  DB00604  DB00619  DB01035  DB01168  DB01171   DB01233  DB01393  DB06288  DB06421  DB06422  DB06626  DB07069  DB08950   DB09018  DB11282  DB11480  DB12518  DB13025  DB13273  DB13523  DB15445 3. Discussions

Binding Mode of BHMBs
To date, the most comprehensive study on InhA inhibition has been reported by Soutter et al. [6]. An important output of their DNA-encoded screening is the novel binding mode of inhibitors at the MtInhA active site, where the ligand, in addition to the typical stacking contact with Phe97 reaches a pocket surrounded by Phe41 and Arg43 side chains. This results in either a π-cation or a conventional hydrogen-bond contact. According to our analysis of the InhA-BHMBx complexes of most potent inhibitors, this interaction plays a key role in the significant improvement of predicted inhibitory potencies of novel benzamides. The enzyme-inhibitor interaction energy (Eint) breakdown to InhA active site residues contribution of best designed BHMB analogs ( Figure 10) shows that in contrast to the TS and VS inhibitors [23] (Figure 4), while the contribution of Phe97 to Eint is almost the same, the impact of interaction with polar Arg43 increased by at least 4 kcal·mol −1 . Surprisingly, the novel benzamide inhibitors display noticeable interaction with Phe97 and Arg43, suggesting that their length, which is comparable to that of TS BHMBs, is sufficient for concomitant interactions. Recently, site exclusion was observed for the ligands in complex with InhA (a pyrrolidine core and three substituents: benzoyl and 1-t-butoxy ethyl groups directly connected to the pyrrolidine core and a pyrazole benzaldehyde connected via the t-butoxy ethyl) benzoyl π-π stacks Phe97, methyl amide HB with Arg43), (PDB: 5G0W) [6]. As displayed in Figure 10, distance between the benzamide carbonyl oxygen and the Arg43 sidechain warhead carbon atom is relatively high for the training set compounds (7.5-9.5 Å). Interestingly, the plot of this distance vs. inhibitory activity pIC50 exp displays a correlation. This distance seems to be one of the key determinants for binding affinity for the class 2 InhA direct inhibitors instead of the existence of a HB. This structural specificity that emerged during the PH4 screening of the virtual library of benzamide analogs led to identification of new hits, the best of which are capable of forming a HB with the Arg43 residue.  Table 2. Table 9. Representation for 24 approved drugs containing benzamide scaffold listed in Table 8.   DB00345  DB00391  DB00409  DB00604  DB00619  DB01035  DB01168  DB01171   DB01233  DB01393  DB06288  DB06421  DB06422  DB06626  DB07069  DB08950   DB09018  DB11282  DB11480  DB12518  DB13025  DB13273  DB13523  DB15445 3. Discussions

Binding Mode of BHMBs
To date, the most comprehensive study on InhA inhibition has been reported by Soutter et al. [6]. An important output of their DNA-encoded screening is the novel binding mode of inhibitors at the MtInhA active site, where the ligand, in addition to the typical stacking contact with Phe97 reaches a pocket surrounded by Phe41 and Arg43 side chains. This results in either a π-cation or a conventional hydrogen-bond contact. According to our analysis of the InhA-BHMBx complexes of most potent inhibitors, this interaction plays a key role in the significant improvement of predicted inhibitory potencies of novel benzamides. The enzyme-inhibitor interaction energy (Eint) breakdown to InhA active site residues contribution of best designed BHMB analogs ( Figure 10) shows that in contrast to the TS and VS inhibitors [23] (Figure 4), while the contribution of Phe97 to Eint is almost the same, the impact of interaction with polar Arg43 increased by at least 4 kcal·mol −1 . Surprisingly, the novel benzamide inhibitors display noticeable interaction with Phe97 and Arg43, suggesting that their length, which is comparable to that of TS BHMBs, is sufficient for concomitant interactions. Recently, site exclusion was observed for the ligands in complex with InhA (a pyrrolidine core and three substituents: benzoyl and 1-t-butoxy ethyl groups directly connected to the pyrrolidine core and a pyrazole benzaldehyde connected via the t-butoxy ethyl) benzoyl π-π stacks Phe97, methyl amide HB with Arg43), (PDB: 5G0W) [6]. As displayed in Figure 10, distance between the benzamide carbonyl oxygen and the Arg43 sidechain warhead carbon atom is relatively high for the training set compounds (7.5-9.5 Å). Interestingly, the plot of this distance vs. inhibitory activity pIC50 exp displays a correlation. This distance seems to be one of the key determinants for binding affinity for the class 2 InhA direct inhibitors instead of the existence of a HB. This structural specificity that emerged during the PH4 screening of the virtual library of benzamide analogs led to identification of new hits, the best of which are capable of forming a HB with the Arg43 residue.  Table 2. Table 9. Representation for 24 approved drugs containing benzamide scaffold listed in Table 8.   DB00345  DB00391  DB00409  DB00604  DB00619  DB01035  DB01168  DB01171   DB01233  DB01393  DB06288  DB06421  DB06422  DB06626  DB07069  DB08950   DB09018  DB11282  DB11480  DB12518  DB13025  DB13273  DB13523  DB15445 3. Discussions

Binding Mode of BHMBs
To date, the most comprehensive study on InhA inhibition has been reported by Soutter et al. [6]. An important output of their DNA-encoded screening is the novel binding mode of inhibitors at the MtInhA active site, where the ligand, in addition to the typical stacking contact with Phe97 reaches a pocket surrounded by Phe41 and Arg43 side chains. This results in either a π-cation or a conventional hydrogen-bond contact. According to our analysis of the InhA-BHMBx complexes of most potent inhibitors, this interaction plays a key role in the significant improvement of predicted inhibitory potencies of novel benzamides. The enzyme-inhibitor interaction energy (Eint) breakdown to InhA active site residues contribution of best designed BHMB analogs ( Figure 10) shows that in contrast to the TS and VS inhibitors [23] (Figure 4), while the contribution of Phe97 to Eint is almost the same, the impact of interaction with polar Arg43 increased by at least 4 kcal·mol −1 . Surprisingly, the novel benzamide inhibitors display noticeable interaction with Phe97 and Arg43, suggesting that their length, which is comparable to that of TS BHMBs, is sufficient for concomitant interactions. Recently, site exclusion was observed for the ligands in complex with InhA (a pyrrolidine core and three substituents: benzoyl and 1-t-butoxy ethyl groups directly connected to the pyrrolidine core and a pyrazole benzaldehyde connected via the t-butoxy ethyl) benzoyl π-π stacks Phe97, methyl amide HB with Arg43), (PDB: 5G0W) [6]. As displayed in Figure 10, distance between the benzamide carbonyl oxygen and the Arg43 sidechain warhead carbon atom is relatively high for the training set compounds (7.5-9.5 Å). Interestingly, the plot of this distance vs. inhibitory activity pIC50 exp displays a correlation. This distance seems to be one of the key determinants for binding affinity for the class 2 InhA direct inhibitors instead of the existence of a HB. This structural specificity that emerged during the PH4 screening of the virtual library of benzamide analogs led to identification of new hits, the best of which are capable of forming a HB with the Arg43 residue.  Table 2. Table 9. Representation for 24 approved drugs containing benzamide scaffold listed in Table 8.   DB00345  DB00391  DB00409  DB00604  DB00619  DB01035  DB01168  DB01171   DB01233  DB01393  DB06288  DB06421  DB06422  DB06626  DB07069  DB08950   DB09018  DB11282  DB11480  DB12518  DB13025  DB13273  DB13523  DB15445 3. Discussions

Binding Mode of BHMBs
To date, the most comprehensive study on InhA inhibition has been reported by Soutter et al. [6]. An important output of their DNA-encoded screening is the novel binding mode of inhibitors at the MtInhA active site, where the ligand, in addition to the typical stacking contact with Phe97 reaches a pocket surrounded by Phe41 and Arg43 side chains. This results in either a π-cation or a conventional hydrogen-bond contact. According to our analysis of the InhA-BHMBx complexes of most potent inhibitors, this interaction plays a key role in the significant improvement of predicted inhibitory potencies of novel benzamides. The enzyme-inhibitor interaction energy (Eint) breakdown to InhA active site residues contribution of best designed BHMB analogs ( Figure 10) shows that in contrast to the TS and VS inhibitors [23] (Figure 4), while the contribution of Phe97 to Eint is almost the same, the impact of interaction with polar Arg43 increased by at least 4 kcal·mol −1 . Surprisingly, the novel benzamide inhibitors display noticeable interaction with Phe97 and Arg43, suggesting that their length, which is comparable to that of TS BHMBs, is sufficient for concomitant interactions. Recently, site exclusion was observed for the ligands in complex with InhA (a pyrrolidine core and three substituents: benzoyl and 1-t-butoxy ethyl groups directly connected to the pyrrolidine core and a pyrazole benzaldehyde connected via the t-butoxy ethyl) benzoyl π-π stacks Phe97, methyl amide HB with Arg43), (PDB: 5G0W) [6]. As displayed in Figure 10, distance between the benzamide carbonyl oxygen and the Arg43 sidechain warhead carbon atom is relatively high for the training set compounds (7.5-9.5 Å). Interestingly, the plot of this distance vs. inhibitory activity pIC50 exp displays a correlation. This distance seems to be one of the key determinants for binding affinity for the class 2 InhA direct inhibitors instead of the existence of a HB. This structural specificity that emerged during the PH4 screening of the virtual library of benzamide analogs led to identification of new hits, the best of which are capable of forming a HB with the Arg43 residue.  Table 2. Table 9. Representation for 24 approved drugs containing benzamide scaffold listed in Table 8.   DB00345  DB00391  DB00409  DB00604  DB00619  DB01035  DB01168  DB01171   DB01233  DB01393  DB06288  DB06421  DB06422  DB06626  DB07069  DB08950   DB09018  DB11282  DB11480  DB12518  DB13025  DB13273  DB13523  DB15445 3. Discussions

Binding Mode of BHMBs
To date, the most comprehensive study on InhA inhibition has been reported by Soutter et al. [6]. An important output of their DNA-encoded screening is the novel binding mode of inhibitors at the MtInhA active site, where the ligand, in addition to the typical stacking contact with Phe97 reaches a pocket surrounded by Phe41 and Arg43 side chains. This results in either a π-cation or a conventional hydrogen-bond contact. According to our analysis of the InhA-BHMBx complexes of most potent inhibitors, this interaction plays a key role in the significant improvement of predicted inhibitory potencies of novel benzamides. The enzyme-inhibitor interaction energy (Eint) breakdown to InhA active site residues contribution of best designed BHMB analogs ( Figure 10) shows that in contrast to the TS and VS inhibitors [23] (Figure 4), while the contribution of Phe97 to Eint is almost the same, the impact of interaction with polar Arg43 increased by at least 4 kcal·mol −1 . Surprisingly, the novel benzamide inhibitors display noticeable interaction with Phe97 and Arg43, suggesting that their length, which is comparable to that of TS BHMBs, is sufficient for concomitant interactions. Recently, site exclusion was observed for the ligands in complex with InhA (a pyrrolidine core and three substituents: benzoyl and 1-t-butoxy ethyl groups directly connected to the pyrrolidine core and a pyrazole benzaldehyde connected via the t-butoxy ethyl) benzoyl π-π stacks Phe97, methyl amide HB with Arg43), (PDB: 5G0W) [6]. As displayed in Figure 10, distance between the benzamide carbonyl oxygen and the Arg43 sidechain warhead carbon atom is relatively high for the training set compounds (7.5-9.5 Å). Interestingly, the plot of this distance vs. inhibitory activity pIC50 exp displays a correlation. This distance seems to be one of the key determinants for binding affinity for the class 2 InhA direct inhibitors instead of the existence of a HB. This structural specificity that emerged during the PH4 screening of the virtual library of benzamide analogs led to identification of new hits, the best of which are capable of forming a HB with the Arg43 residue.  Table 2. Table 9. Representation for 24 approved drugs containing benzamide scaffold listed in Table 8.   DB00345  DB00391  DB00409  DB00604  DB00619  DB01035  DB01168  DB01171   DB01233  DB01393  DB06288  DB06421  DB06422  DB06626  DB07069  DB08950   DB09018  DB11282  DB11480  DB12518  DB13025  DB13273  DB13523  DB15445 3. Discussions

Binding Mode of BHMBs
To date, the most comprehensive study on InhA inhibition has been reported by Soutter et al. [6]. An important output of their DNA-encoded screening is the novel binding mode of inhibitors at the MtInhA active site, where the ligand, in addition to the typical stacking contact with Phe97 reaches a pocket surrounded by Phe41 and Arg43 side chains. This results in either a π-cation or a conventional hydrogen-bond contact. According to our analysis of the InhA-BHMBx complexes of most potent inhibitors, this interaction plays a key role in the significant improvement of predicted inhibitory potencies of novel benzamides. The enzyme-inhibitor interaction energy (Eint) breakdown to InhA active site residues contribution of best designed BHMB analogs ( Figure 10) shows that in contrast to the TS and VS inhibitors [23] (Figure 4), while the contribution of Phe97 to Eint is almost the same, the impact of interaction with polar Arg43 increased by at least 4 kcal·mol −1 . Surprisingly, the novel benzamide inhibitors display noticeable interaction with Phe97 and Arg43, suggesting that their length, which is comparable to that of TS BHMBs, is sufficient for concomitant interactions. Recently, site exclusion was observed for the ligands in complex with InhA (a pyrrolidine core and three substituents: benzoyl and 1-t-butoxy ethyl groups directly connected to the pyrrolidine core and a pyrazole benzaldehyde connected via the t-butoxy ethyl) benzoyl π-π stacks Phe97, methyl amide HB with Arg43), (PDB: 5G0W) [6]. As displayed in Figure 10, distance between the benzamide carbonyl oxygen and the Arg43 sidechain warhead carbon atom is relatively high for the training set compounds (7.5-9.5 Å). Interestingly, the plot of this distance vs. inhibitory activity pIC50 exp displays a correlation. This distance seems to be one of the key determinants for binding affinity for the class 2 InhA direct inhibitors instead of the existence of a HB. This structural specificity that emerged during the PH4 screening of the virtual library of benzamide analogs led to identification of new hits, the best of which are capable of forming a HB with the Arg43 residue.

Binding Mode of BHMBs
To date, the most comprehensive study on InhA inhibition has been reported by Soutter et al. [6]. An important output of their DNA-encoded screening is the novel binding mode of inhibitors at the MtInhA active site, where the ligand, in addition to the typical stacking contact with Phe97 reaches a pocket surrounded by Phe41 and Arg43 side chains. This results in either a π-cation or a conventional hydrogen-bond contact. According to our analysis of the InhA-BHMBx complexes of most potent inhibitors, this interaction plays a key role in the significant improvement of predicted inhibitory potencies of novel benzamides. The enzyme-inhibitor interaction energy (E int ) breakdown to InhA active site residues contribution of best designed BHMB analogs ( Figure 10) shows that in contrast to the TS and VS inhibitors [23] (Figure 4), while the contribution of Phe97 to E int is almost the same, the impact of interaction with polar Arg43 increased by at least 4 kcal·mol −1 . Surprisingly, the novel benzamide inhibitors display noticeable interaction with Phe97 and Arg43, suggesting that their length, which is comparable to that of TS BHMBs, is sufficient for concomitant interactions. Recently, site exclusion was observed for the ligands in complex with InhA (a pyrrolidine core and three substituents: benzoyl and 1-t-butoxy ethyl groups directly connected to the pyrrolidine core and a pyrazole benzaldehyde connected via the t-butoxy ethyl) benzoyl π-π stacks Phe97, methyl amide HB with Arg43), (PDB: 5G0W) [6]. As displayed in Figure 10, distance between the benzamide carbonyl oxygen and the Arg43 sidechain warhead carbon atom is relatively high for the training set compounds (7.5-9.5 Å). Interestingly, the plot of this distance vs. inhibitory activity pIC 50 exp displays a correlation. This distance seems to be one of the key determinants for binding affinity for the class 2 InhA direct inhibitors instead of the existence of a HB. This structural specificity that emerged during the PH4 screening of the virtual library of benzamide analogs led to identification of new hits, the best of which are capable of forming a HB with the Arg43 residue.

Binding Mode of BHMBs
To date, the most comprehensive study on InhA inhibition has been reported by Soutter et al. [6]. An important output of their DNA-encoded screening is the novel binding mode of inhibitors at the MtInhA active site, where the ligand, in addition to the typical stacking contact with Phe97 reaches a pocket surrounded by Phe41 and Arg43 side chains. This results in either a π-cation or a conventional hydrogen-bond contact. According to our analysis of the InhA-BHMBx complexes of most potent inhibitors, this interaction plays a key role in the significant improvement of predicted inhibitory potencies of novel benzamides. The enzyme-inhibitor interaction energy (Eint) breakdown to InhA active site residues contribution of best designed BHMB analogs ( Figure 10) shows that in contrast to the TS and VS inhibitors [23] (Figure 4), while the contribution of Phe97 to Eint is almost the same, the impact of interaction with polar Arg43 increased by at least 4 kcal·mol −1 . Surprisingly, the novel benzamide inhibitors display noticeable interaction with Phe97 and Arg43, suggesting that their length, which is comparable to that of TS BHMBs, is sufficient for concomitant interactions. Recently, site exclusion was observed for the ligands in complex with InhA (a pyrrolidine core and three substituents: benzoyl and 1-t-butoxy ethyl groups directly connected to the pyrrolidine core and a pyrazole benzaldehyde connected via the t-butoxy ethyl) benzoyl π-π stacks Phe97, methyl amide HB with Arg43), (PDB: 5G0W) [6]. As displayed in Figure 10, distance between the benzamide carbonyl oxygen and the Arg43 sidechain warhead carbon atom is relatively high for the training set compounds (7.5-9.5 Å). Interestingly, the plot of this distance vs. inhibitory activity pIC50 exp displays a correlation. This distance seems to be one of the key determinants for binding affinity for the class 2 InhA direct inhibitors instead of the existence of a HB. This structural specificity that emerged during the PH4 screening of the virtual library of benzamide analogs led to identification of new hits, the best of which are capable of forming a HB with the Arg43 residue.

PH4-Based Screening of Approved Drugs Containing Benzamide Scaffold
A library of 24 benzamide scaffold drugs from DrugBank [36] was screened with the class 2 InhA inhibition PH4. These drugs are mostly dedicated to a variety of neurological and psychiatric disorders or nausea and vomiting treatment. The top five best PH4 hits have been evaluated also with the complexation QSAR model Equation B, Table 3, to predict their antituberculotic IC50 pre towards the MtInhA: Sultopride (IC50 pre = 1.7 nM), Diethyltoluamide (9 nM), Tricalopride (10 nM), Veralipride (23 nM) and Remoxipride (60 nM). Their mapping to the pharmacophore features is displayed in Figure 9. Therefore, experimental testing of these five approved drugs for MtInhA inhibition and perhaps also for antibacterial effect against the Mt would be worthwhile.

Training and Validation Sets
Chemical structures and biological activities (IC 50 exp ) of training and validation sets of N-benzyl-4-((heteroaryl)methyl) benzamides inhibitors of InhA used in this study were taken from literature [23]. The potencies of these compounds cover a sufficiently broad range of half-maximal inhibitory concentrations (20 ≤ IC50 exp ≤ 5930 nM) to allow construction of a QSAR model. The training set (TS) containing 19 BHMB inhibitors and the validation set (VS) including 6 BHMBs were taken from the ref. [23].
The structures of InhA and the E-I complexes were considered to be at pH of 7 with neutral Nand C-terminal residues and all protonizable and ionizable residues charged. No crystallographic water molecules were included in the model. The inhibitors were built into the reference structure 4QXM [23] by in situ replacing of derivatized groups in the molecular scaffold of the template inhibitor BHMB2. An exhaustive conformational search over all rotatable bonds of the replacing function groups coupled with a careful gradual energy-minimization of the modified inhibitor and active site residues of the InhA located in the vicinity of the inhibitor (within 5 Å distance) was

PH4-Based Screening of Approved Drugs Containing Benzamide Scaffold
A library of 24 benzamide scaffold drugs from DrugBank [36] was screened with the class 2 InhA inhibition PH4. These drugs are mostly dedicated to a variety of neurological and psychiatric disorders or nausea and vomiting treatment. The top five best PH4 hits have been evaluated also with the complexation QSAR model Equation B, Table 3, to predict their antituberculotic IC 50 pre towards the MtInhA: Sultopride (IC 50 pre = 1.7 nM), Diethyltoluamide (9 nM), Tricalopride (10 nM), Veralipride (23 nM) and Remoxipride (60 nM). Their mapping to the pharmacophore features is displayed in Figure 9. Therefore, experimental testing of these five approved drugs for MtInhA inhibition and perhaps also for antibacterial effect against the Mt would be worthwhile.

Training and Validation Sets
Chemical structures and biological activities (IC 50 exp ) of training and validation sets of N-benzyl-4-((heteroaryl)methyl) benzamides inhibitors of InhA used in this study were taken from literature [23]. The potencies of these compounds cover a sufficiently broad range of half-maximal inhibitory concentrations (20 ≤ IC 50 exp ≤ 5930 nM) to allow construction of a QSAR model. The training set (TS) containing 19 BHMB inhibitors and the validation set (VS) including 6 BHMBs were taken from the ref. [23].
The structures of InhA and the E-I complexes were considered to be at pH of 7 with neutral N-and C-terminal residues and all protonizable and ionizable residues charged. No crystallographic water molecules were included in the model. The inhibitors were built into the reference structure 4QXM [23] by in situ replacing of derivatized groups in the molecular scaffold of the template inhibitor BHMB2. An exhaustive conformational search over all rotatable bonds of the replacing function groups coupled with a careful gradual energy-minimization of the modified inhibitor and active site residues of the InhA located in the vicinity of the inhibitor (within 5 Å distance) was employed to identify low-energy bound conformations of the modified inhibitor. The resulting low-energy structures of the E-I complexes were then carefully refined by minimization of the whole complex. This procedure has been successfully used for model building of viral, bacterial, and protozoal enzyme-inhibitor complexes and design of peptidomimetic, hydroxynaphthoic, thymidine, triclosan, pyrrolidine carboxamide, nitriles, and chalcone-based inhibitors [16,17,29,[39][40][41][42][43][44][45][46][47].

Molecular Mechanics
Modeling of inhibitors, InhA, and E-I complexes was carried out by molecular mechanics using CFF91 force field [48] as described earlier [17].

Conformational Search
Free inhibitor conformations were derived from their bound conformations in the E-I complexes by gradual relaxation to the nearest local energy minimum as described earlier [17].

Solvation Gibbs Free Energies
The electrostatic component of solvation Gibbs free energy (GFE) that includes also the effects of ionic strength via solving nonlinear Poisson-Boltzmann equation [49,50] was computed by the DelPhi module in Discovery Studio [26] as described earlier [17].

Calculation of Binding Affinity and QSAR Model
The calculation of binding affinity expressed as complexation GFE has been described fully earlier [17].

Interaction Energy
The calculation of MM interaction energy (E int ) between enzyme residues and the inhibitor CFF91 force field [48] was performed as described earlier [17].

Pharmacophore Generation
Bound conformations of inhibitors taken from the models of E-I complexes were used for constructing of 3D-QSAR pharmacophore (PH4) by means of Catalyst HypoGen algorithm [51] implemented in Discovery Studio [26] as described earlier [17].

ADME Properties
The pharmacokinetics profile of BHMBs were computed by the QikProp program [33] as described earlier [17].

Virtual Library Generation
The virtual library generation was performed as described earlier [17].

ADME-Based Library Searching
The drug-likeness selection criterion served to focus the initial virtual library as described earlier [17].

Pharmacophore-Based Library Searching
The pharmacophore model (PH4) described in Section 4.8 and derived from the bound conformations of BHMBs at the active site of InhA served as library searching tool as described earlier [17].

Inhibitory Potency Prediction
The conformer with the best mapping on the PH4 pharmacophore in each cluster of the focused library subset was used for ∆∆G com calculation and IC 50 pre estimation (virtual screening) by the complexation QSAR model as described earlier [17].

Conclusions
In this work novel class 2 InhA direct inhibitors with benzamide scaffold have been designed to reach the picomolar inhibitory concentration range of the predicted IC 50 pre ( Table 6, Figure 8).
Even though these predicted inhibitory potencies may be somewhat too optimistic, they suggest that benzamide-type MtInhA inhibitors more potent than the known TS and VS analogs [23] may exist. Our QSAR model provided bound InhA inhibitor conformation, from which the enzyme-inhibitor interaction energy breakdown to active site residue contribution clearly revealed structural determinants needed for binding improvement involving the favorable contacts in the three active site subsites (I, II, and III) including combination of π-π stacking with Phe97 and the HB contacts with Arg43. The derived 3D QSAR pharmacophores identified during chemical space exploration around R 1