2022 Volume 70 Issue 8 Pages 544-549
Fatty acid biosynthesis is essential for bacterial survival. Of these promising targets, β-ketoacyl-acyl carrier protein (ACP) synthase III (FabH) is the most attractive target. FabH would trigger the initiation of fatty acid biosynthesis and it is highly conserved among Gram-positive and -negative bacteria. A series of novel amide derivatives bearing dioxygenated rings were synthesized and developed as potent inhibitors of FabH. These compounds were determined by 1H-NMR, 13C-NMR, MS and further confirmed by crystallographic diffraction study for compound 19. Furthermore, these compounds were evaluated strong broad-spectrum antibacterial activity. Some compounds with potent antibacterial activities were tested for their Escherichia coli (E. coli) FabH inhibitory activity. Especially, compound 19 showed the most potent antibacterial activity with minimum inhibitory concentration (MIC) values of 1.56–3.13 mg/mL against the tested bacterial strains and exhibited the most potent E. coli FabH inhibitory activity with IC50 of 2.4 µM. Docking simulation was performed to position compound 19 into the E. coli FabH active site to determine the probable binding conformation.
Bacterial resistance poses a serious global threat of growing concern to human, animal, and environment health. Thus, it is imperative to seek new therapeutic targets and medications.1,2) Among them, targeting the fatty acid biosynthetic pathway of bacterial is particularly attractive. Since the fatty acid synthesis (FAS) is a vital facet of bacteria, moreover, the FAS between mammals and bacteria possesses prominent selectivity.3–5) The FAS found in mammals (FAS I) are consisted of a single gene that produces a polypeptide, which contains all of the reaction centers required to produce a fatty acid. The FAS in bacteria (FAS II) is a dissociated system wherein each component is encoded by a separate gene that produces a discrete protein, which catalyzes a specific step in the pathway.6)
β-Ketoacyl-acyl carrier protein synthase III (KAS III), also named as FabH, is a vital enzyme for the bacterial FAS II. FabH initiates the FAS cycle by catalyzing the first condensation step between acetyl-CoA and malonyl acyl carrier protein (ACP) to form Acetoacyl-ACP and inhibits the product of the FAS through the feedback mechanism7,8) (Fig. 1). Futhermore, FabH proteins from both Gram-positive and -negative bacteria are highly conserved at the sequence and structural level, while there are no significantly homologous proteins in humans.9,10) These benefits make FabH the key target enzyme in the fatty acid synthesis of bacteria, and prospective target for developing broad-spectrum antibiotics.11–15)
Researchers including us have aroused great interest in new antibiotics agents targeting FabH, and some potential inhibitors with high antibacterial activity have been invented13,16–20) (Fig. 2).
From our previous research, four novel Schiff bases derived from YKAs300317) that retain the structure of the cyclohexylamine moiety were designed. The Schiff base with a dioxygenated seven-membered ring was the most appropriate candidate for the development of new FabH inhibitors via molecular docking. Further studies revealed compounds combined these skeleton and substituents of 16 aliphatic or tert-butyl piperidine carboxylate were the most promising Escherichia coli (E. coli) FabH inhibitors.21,22) Considering the instability of Schiff base, the N-acylhydrazone (–CO–NH–N–CH–) was chosen for the design of novel FabH inhibitors as successive research. Compound with three functional groups: dioxygenated seven-membered ring, N-acylhydrazone and benzen ring bearing three methoxyl groups exhibited better antibacterial activity than Kanamycin B.23) However, all these three compounds displayed poor metabolic stability of primary in vitro screen, which imposed restrictions on in vivo study.
The amide functional group plays a key role in the composition of biomolecules, including many clinically approved drugs.24,25) The amide group enjoys significant attention not only due to its unique ability to form relevant hydrogen bonding interactions, but also it is more stable relative to the ester group and other easily hydrolyzable groups at the same position.26–29) To improve the metabolic stability of the lead compounds, our new design strategy is to replace the Schiff base/N-acylhydrazone moiety with amide in the core structures. The design strategy is presented in Fig. 3. In this paper, we synthesize a series of compounds with amide linked dioxygenated seven-membered ring, and evaluate their biological activities. Thus, molecular docking studies of the E. coli FabH with the most potent inhibitor are performed to explore the binding model of the compound at the active site.
In this study, 31 amide derivatives (Chart 1) were synthesized. The synthetic route of compounds 3–33 was depicted in Chart 1. Compound 1 was obtained from the reaction of 3,4-dihydroxybenzaldehyde and 1,3-dibromopropane in N,N-dimethylformamide (DMF) solvent in the presence of Cs2CO3 as a base. Oxidized compound 1 with KMnO4 to form the required 3,4-dihydro-2H-benzo[b][1,4]dioxepine-7-carboxylic acid (2) in 73% yield. Then, compounds 3–33 was accomplished by amidation of compound 2 with various amines using EDC·HCl and N,N-dimethyl-4-aminopyridine (DMAP) as the coupling reagents with 68–93% yield. Details of the whole synthesis process are shown in the SI. All synthesized compounds were fully characterized by 1H- and 13C-NMR, and mass (electrospray ionization (ESI)-MS) spectral data.
Reagents and conditions: i) 1,3-Dibromopropane, Cs2CO3, DMF, 70 °C, overnight, ii) KMnO4, water, 70–80 °C, 5 h; iii) R-NH2, EDC·HCl, DMAP, room temperature (r.t.), overnight.
The crystal structures of compound 19 were determined by X-ray diffraction analysis. The crystal data presented in Table 1 and Fig. 4 gave the perspective views of 19 with the atomic labeling system. The crystallographic data have been deposited at the Cambridge Crystallographic Data Centre (CCDC) and the deposition number of 19 is 2121248.
| Compound | 19 | ||
|---|---|---|---|
| Empirical formula | C17H17NO4 | Z | 12 |
| Temperature/K | 173.00 | Dcalcd (Mg m−3) | 1.381 |
| Crystal system | Monoclinic | μ (mm−1) | 0.099 |
| Space group | P21/c | F(000) | 1896.0 |
| a (Å) | 28.6548(11) | θ limits (°) | 4.318 to 54.988 |
| b (Å) | 5.1708(2) | Reflections collected | 18929 |
| c (Å) | 29.5064(9) | Independent reflections | 9878 [Rint = 0.0415] |
| α (°) | 90 | Data/restraints/parameters | 9878/34/627 |
| β (°) | 99.0210(10) | GOF | 1.009 |
| γ (°) | 90 | R1/wR2 [I > 2&Gs (I)] | 0.0593/0.1228 |
| Volume (Å3) | 4317.8(3) | R1/wR2 (all data) | 0.1120/0.1510 |
All the synthesized amide derivatives 3–33 were evaluated for their antimicrobial activities against two Gram-negative bacterial strains: E. coli and Pseudomonas aeruginosa and two Gram-positive bacterial strains: Staphylococcus aureus and Bacillus subtilis by the standard two-fold serial broth dilution method.30,31) The minimum inhibitory concentrations (MICs) of those compounds against these bacteria were presented in Table 2. Standard antibacterial agent kanamycin B were also screened under identical conditions for comparison and the results revealed that most of the synthetic compounds exhibited certain antibacterial activities. Dimetyl sulfoxide (DMSO) was added as the negative control within the biological testing.
| Compda) | A log Pb) | Minimum inhibitory concentrations [µg/mL] | |||
|---|---|---|---|---|---|
| Gram-negative | Gram-positive | ||||
| E. coli | P. aeruginosa | B. subtilis | S. aureus | ||
| 3 | 2.215 | >100 | >100 | >100 | 50 |
| 4 | 3.127 | >100 | 50 | 50 | >100 |
| 5 | 4.039 | 12.5 | 6.25 | 50 | 12.5 |
| 6 | 5.864 | 25 | 12.5 | 12.5 | 6.25 |
| 7 | 7.689 | 3.13 | 6.25 | 12.5 | 6.25 |
| 8 | 2.078 | 12.5 | 12.5 | 50 | >100 |
| 9 | 1.817 | >100 | >100 | >100 | 50 |
| 10 | 2.744 | 50 | 50 | >100 | >100 |
| 11 | 2.462 | 12.5 | 6.25 | 50 | 25 |
| 12 | 2.949 | 6.25 | 6.25 | 12.5 | 12.5 |
| 13 | 3.435 | 6.25 | 12.5 | 6.25 | 25 |
| 14 | 3.921 | 25 | 25 | 50 | >100 |
| 15 | 4.851 | 25 | 50 | >100 | 50 |
| 16 | 2.668 | 6.25 | 3.13 | 6.25 | 3.13 |
| 17 | 2.668 | 6.25 | 6.25 | 50 | 12.5 |
| 18 | 3.405 | 1.56 | 3.13 | 6.25 | 3.13 |
| 19 | 2.446 | 1.56 | 1.56 | 3.13 | 3.13 |
| 20 | 2.652 | 3.13 | 1.56 | 3.13 | 6.25 |
| 21 | 3.194 | 6.25 | 25 | 6.25 | 25 |
| 22 | 2.413 | 6.25 | 6.25 | 3.13 | 3.13 |
| 23 | 3.157 | 50 | 12.5 | 12.5 | 6.25 |
| 24 | 2.515 | 6.25 | 6.25 | 3.13 | 6.25 |
| 25 | 3.131 | 6.25 | 6.25 | 6.25 | 12.5 |
| 26 | 3.371 | 50 | >100 | >100 | >100 |
| 27 | 3.371 | >100 | 25 | 50 | 25 |
| 28 | 2.469 | 50 | 25 | 50 | 25 |
| 29 | 2.453 | 6.25 | 12.5 | 6.25 | 3.13 |
| 30 | 2.847 | 50 | 25 | 50 | 25 |
| 31 | 3.412 | 6.25 | 12.5 | 6.25 | 6.25 |
| 32 | 1.651 | 25 | 50 | >100 | 50 |
| 33 | 4.03 | 12.5 | 50 | 50 | 25 |
| Kanamycin B | −7.144 | 3.13 | 3.13 | 1.56 | 1.56 |
a) The compounds tested for antibacterial activity are consistent with the description in the Experimental Section. b) Calculated with discovery studio 4.5.32)
In order to describe the structure–activity relationships (SARs) of these compounds, this series was classified into two groups: compounds with aliphatic amine side chains (3–10) and those with aromatic amine substituents (11–33). Rising the length of the aliphatic chain enhances the antibacterial potency of compounds as seen in the MIC values from compound 3 to 7 (Table 1). While increase the side chains of short aliphatic chains or replace them with cycloalkanes, that does not improve the activity. For the compounds with aliphatic amine chains, the high predicted lipophilicity (Alog P) values accompanied by a relatively high hydrophobicity hydrophobic properties, which contribute more to the antibiotic activity.
The substituents on the benzene ring are important factors affect their antibacterial activity. Based on our previous research, specific substituents such as methyl, methoxy groups and fluorine groups are introduced to improve the antibacterial potency. The compounds with fluoro (16, 17), trifluoromethyl (18), and methoxyl (19–22) substituents of the benzene ring performed better than other moieties. And these substituents of benzylamine (29, 31) also showed higher potencies than other benzylamine or furfurylamine compounds (28, 30, 32, 33). Especially compound 19 exhibited the most potent antibacterial activity with MIC of 1.56, 1.56, 3.13, and 3.13 µg/mL against E. coli, P. fluorescence, B. subtilis, and S. aureus, respectively, which was similar to the broad-spectrum antibiotic Kanamycin B with corresponding MIC of 3.13, 3.13, 1.56, and 1.56 µg/mL. However, compound 21 represents an exception, probably because the bromine substituent could cause the loss of activity. Compounds 24 and 25 with N-heterocyclic rings replacing the benzene ring showed better antibacterial activity than other compounds, which will provide another strategy to explore better antibacterial agents in the future work. Compounds contained α- or β-naphthalene rings (26, 27) displayed low inhibitory activity. A possible explanation for these results is the introducing large steric substituents to this series of compounds cause the adverse effects in antibacterial activity.
Next, the synthetic compounds with potent antibacterial activities (7, 12, 13, 16–25, 29, 30) were further evaluated against E. coli FabH enzymes, and the IC50 values were summarized in Table 3. Most of the test compounds displayed potent inhibitory activity. Among them, compounds with fluoro and methoxyl substituents of the benzene ring once again showed higher inhibitory activities. Particularly those compounds with methoxyl substituents (19, 20, 22, 29) presented considerable inhibitory activities (IC50 < 9 µM). This probably methoxyl group share the same binding site. Also, there are conformational complementarity and strong non-bonding interactions between the methoxyl group and the amino acids in the binding pocket, which significantly stabilizes the binding conformation. All these demonstrated that compounds bearing methoxyl substituents have high affinity for FabH protein. These biological assays indicated that compound 19 is a potent inhibitor of E. coli FabH as antibiotic. The molecular polar surface areas (PSA) of these compounds were in the range of 47–88 Å2, and therefore, this molecular property cannot significantly account for the results obtained.
| Compounda) | E. coli FabH | PSAb) | Hemolysis LCc) | Cytotoxicity |
|---|---|---|---|---|
| IC50 [µM] | [Å2] | [mg/mL] | IC50 [µM] | |
| 7 | 9.4 ± 0.19 | 47.971 | >10 | 135.8 ± 12.3 |
| 12 | 20.3 ± 0.11 | 47.971 | >10 | 127.4 ± 8.8 |
| 13 | 32.6 ± 0.37 | 47.971 | >10 | 113.9 ± 13.5 |
| 16 | 8.5 ± 0.18 | 47.971 | >10 | 158.1 ± 14.7 |
| 17 | 7.7 ± 0.23 | 47.971 | >10 | 178.3 ± 16.2 |
| 18 | 6.8 ± 0.21 | 47.971 | >10 | 193.8 ± 11.9 |
| 19 | 2.4 ± 0.16 | 56.901 | >10 | 182.7 ± 5.3 |
| 20 | 3.8 ± 0.23 | 56.901 | >10 | 198.2 ± 12.8 |
| 21 | 27.1 ± 0.32 | 56.901 | >10 | 106.1 ± 9.7 |
| 22 | 8.6 ± 0.12 | 74.761 | >10 | 145.7 ± 14.4 |
| 23 | 39.4 ± 0.27 | 87.012 | >10 | 162.1 ± 7.2 |
| 24 | 12.9 ± 0.15 | 59.232 | >10 | 130.4 ± 11.3 |
| 25 | 16.6 ± 0.17 | 74.287 | >10 | 122.8 ± 10.6 |
| 29 | 7.2 ± 0.24 | 56.901 | >10 | 182.7 ± 10.6 |
| 31 | 10.9 ± 0.18 | 47.971 | >10 | 210.3 ± 19.7 |
a) The compounds tested for antibacterial activity are consistent with the description in the Experimental Section. b) Molecular polar surface area (PSA); calculated with Discovery studio 4.5.31) c) Lytic concentration 30%.
Not only antibacterial activity, but also biological safety are the important indexes for potent antibiotics. So the compounds selected above were also tested for their hemolytic activity and cytotoxic activity on a mouse embryonic fibroblast cell line (NIH-3T3) using the 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) assay. In summary, as shown in Table 3, these compounds displayed low hemolytic activities. Moreover, the cytotoxicity data indicates that the compounds possessing inhibitory activity were low toxicity.
In order to determine the interaction binding mode between the target protein and small molecules, the protein crystal structure of E. coli FabH (PDB entry code: 5BNM) was employed for in silico molecular docking as the receptor model. The three dimensional (3D) binding pose and the 2D interaction diagram between the key residues of FabH active pocket and the native inhibitor (biphenyl sulfonamide) were presented in Figs. 5a and b, respectively. Judged on the co-crystal structures of E. coli FabH with the biphenyl sulfonamide, the biphenyl portion makes mostly lipophilic interactions, with the sulfonamide too far from the many arginines in the right side of the pocket to form favorable interactions. The hydroxymethyl group engages with the His244−Asn274−Cys112 catalytic triad in a complex of hydrogen bonding.33) Also, several hydrophobic amino acids (labeled as green, such as Phe304, Ile250, Gly209, Gly305, and Phe157) can form hydrophobic pockets, and enhance the FabH binding potency of native inhibitor. The hydroxymethyl group, binding to the catalytic triad, represented an opportunity to introduce polarity into the fatty acid portion of the binding pocket. This was considered advantageous for the development of small-molecule inhibitors of FabH with physical properties suited to support Gram-negative cell permeability. Besides, the difference among target compounds is just substituent, therefore, their binding mode is substantially identical. Since molecular skeleton are consistent, the difference lies in the substituents. Some substituents can form the interaction force, so that the combination is enhanced. The others affect the activity by peripheral electronic arrangement. Their binding patterns for the target protein are similar on the whole.
a) and b): the 3D and the 2D diagram of the native inhibitor interacting with the FabH. c) and d): the 3D optimal conformation and the 2D diagram of compound 19 interacting with the FabH active site. e): interaction maps of native inhibitor and compound 19.
The docking result showed compound 19 can be inserted into in the FabH binding site with the docking score of −7.827 kcal/mol. Similarly, the 3D binding pose and the 2D interaction diagram of compound 19 were displayed in Figs. 5c and 5d. Even without the hydroxymethyl group binding to the catalytic triad, one oxygen of the dioxygenated rings contained in compound 19 are able to form a hydrogen bond with Asn274, which plays a vital role for the stabilization of its binding mode. Significantly, as showed in Fig. 5e, the two compounds (19 and the biphenyl sulfonamide) can be overlapped well in the protein pocket, which also verified the rationality of molecular docking (compound 19 marked in green, the native inhibitor marked in cyan). In addition, the hydrophobic moiety (left) of the two molecules seemed to be perfectly overlapped, and the conformational difference between these two molecules mainly existed on the right side. This would contribute to the better understanding of the most active compound 19. As a result, the more modification of the amide substituent of compound 19 would be effective for the development of excellent antibacterial agents in the future work.
In summary, a series of amide derivatives (3–33) were prepared and tested for their inhibitory activity against E. coli, P. fluorescence, B. subtilis, and S. aureus. Many synthesized compounds showed potent antibacterial and E. coli FabH inhibitory activities. Particularly, compound 19 showed the most potent E. coli FabH inhibitory activity with an IC50 value of 2.4 µM, which was compared with the positive control, kanamycin B. Structure–activity relationships prediction and molecular modeling study provided further insight into interactions between the enzyme and its ligand. Based on the data obtained in this study, we conclude that compound 19 is the E. coli FabH inhibitor most deserving of further research as a potential antibiotic. Additionally, the development of similar novel compounds with amide bearing with a dioxygenated seven-membered ring will be investigated in future studies.
This work was supported by the Natural Science Foundation of Zhejiang Province (LQ19E030006), the S&T Innovation 2025 Major Special Program of Ningbo (2020Z091), the Natural Science Foundation of China (51803227).
The authors declare no conflict of interest.
This article contains supplementary materials.
