- Email: [email protected]
Discovery of novel inhibitors of multidrug-resistant Acinetobacter baumannii
Bioorganic & Medicinal Chemistry 25 (2017) 5477–5482
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry journal homepage: www.elsevier.com/locate/bmc
Discovery of novel inhibitors of multidrug-resistant Acinetobacter baumannii Iswarduth Soojhawon a,d, Nagarajan Pattabiraman b,d, Arthur Tsang a, Amanda L. Roth c, Ellen Kang a, Schroeder M. Noble a,⇑ a b c
Wound Infections Department, Bacterial Diseases Branch, Walter Reed Army Institute of Research, Silver Spring, MD 20910, USA MolBox LLC, Silver Spring, MD 20910, USA Multidrug-resistant Organism Repository and Surveillance Network, Bacterial Diseases Branch, Walter Reed Army Institute of Research, Silver Spring, MD 20910, USA
a r t i c l e
i n f o
Article history: Received 10 May 2017 Revised 4 August 2017 Accepted 8 August 2017 Available online 9 August 2017 Keywords: Acinetobacter baumannii In silico Docking Antibacterial Outer membrane protein
a b s t r a c t The recent emergence of multidrug-resistant Acinetobacter baumannii strains and the non-efficacy of currently available antibiotics against such infections have led to an urgent need for the development of novel antibacterials. In an effort to address this problem, we have identified three novel inhibitors, namely, D5, D12 and D6 using in silico screening with a homology model of the outer membrane protein W2 (OmpW2) from A. baumannii, as the proposed new drug target. OmpW is an eight-stranded b-barrel protein involved in the transport of hydrophobic molecules across the outer membrane and maintenance of homeostasis under cellular stress. The antimicrobial activities of compounds D5, D12 and D6 were evaluated against a panel of clinical isolates of A. baumannii strains. These compounds inhibited the growth of the strains with minimum inhibitory concentration (MIC) ranges of 1–32 lg/mL. Time-kill kinetic studies with the highly virulent and multidrug-resistant strain, A. baumannii 5075, indicated that D6 exhibited the highest bactericidal activity as a 3 log10 CFU/mL (99.9%) reduction in colony count from the initial inoculum was observed after 30 min incubation. D5 and D12 reduced at least 1 log10 CFU/mL (90%) of the initial inoculum after 24 h. In conclusion, these three lead inhibitors have provided two distinct chemical scaffolds for further analog design and optimizations, using chemical synthesis, to develop more potent inhibitors of the pathogen. Published by Elsevier Ltd.
1. Introduction The rapid emergence of multidrug-resistant (MDR) bacteria threatens the effectiveness of currently available antibiotics used to treat the civilian population and deployed forces infected with these pathogens. To combat resistance, new antibacterials with novel modes of action or binding sites are urgently required, especially against drug resistant Acinetobacter baumannii infections, which are difficult to treat. A. baumannii is a gram-negative, aerobic, non-fermentative, catalase-positive, oxidase-negative, coccobacillus opportunistic pathogen.1–4 This pathogen can survive hospital environments due to its ability to form biofilms on abiotic surfaces, such as catheters, ventilators, and medical devices etc.5–7 A. baumannii is a major cause of hospital-associated infections such Abbreviations: AB_OmpW2, A. baumannii OmpW; MIC, minimum inhibitory concentration; MDR, multidrug-resistant; CFU, colony-forming units. ⇑ Corresponding author. E-mail address: [email protected] (S.M. Noble). d Equally Contributed. http://dx.doi.org/10.1016/j.bmc.2017.08.014 0968-0896/Published by Elsevier Ltd.
as ventilator-associated pneumonia (VAP),8 skin,1 wound,1 urinary tract,4 bacteremia,9 meningitis10 and soft tissue11 infections in patients within civilian and military heath care centers. Interest in A. baumannii (MDR) arose after it was detected in infected combat troops returning from war zones.12,13 The consequences of such infections are prolonged wound healing, which sometimes lead to amputations and accounts for the increased morbidity rate.13 A. baumannii is reported to be responsible for more than 10% of all hospital-acquired infections in the United States and has a greater than 50% mortality rate in patients with sepsis and pneumonia.14–16 A. baumannii was initially sensitive to most available antibiotics; however, it is now resistant to most first-line antibiotics.17 One of the distinguishing features of the bacterium, which made it a ‘‘red-alert’’ human pathogen, is its impressive array of acquired antibiotic resistance mechanisms.3 Indeed, most A. baumannii isolates collected in military and civilian health care centers were often sensitive to carbapenems in the 1990s, but now demonstrate decreased sensitivity to carbapenem, if not outright resistance to these last resort antibiotics.18
5478
I. Soojhawon et al. / Bioorganic & Medicinal Chemistry 25 (2017) 5477–5482
Currently available antibiotics are designed to target cellular functions, such as cell wall, DNA, RNA, or protein biosynthesis in bacterial pathogens. However, the decreased susceptibility of A. baumannii strains to cephalosporins, aminoglycosides, quinolones and, lately, carbapenems has necessitated the development of novel antibacterials, which act on new drug targets crucial for bacterial growth. Promising drug targets in gram-negative bacteria include the outer membrane proteins (OMPs), specifically the outer membrane protein W (OmpW). OmpW is an eight-stranded b-barrel shaped outer membrane protein, which is widespread in many gram-negative bacteria, such as Vibrio cholerae, Escherichia coli, and A. baumannii.19,20 OmpW protein is believed to be involved in the transport of hydrophobic or aromatic molecules across the outer membrane.20–22 It is also implicated in the maintenance of homeostasis (osmoregulation) under cellular stress. Studies have demonstrated that E. coli colicin S4 exploits OmpW for entry into the bacterial cell and eventually kills the bacterium through pore formation. Deletion of OmpW from E. coli was associated with decreased MICs of several antibiotics and the protein was found to be down regulated when E. coli was exposed to chlortetracycline.23 Moreover, mutant strains of V. cholerae, which expressed no OmpW, were shown to have reduced colonization in the intestine of mice relative to those that expressed OmpW.19 All these aforementioned studies indicate that OmpW could be a potential drug target in gram-negative bacteria. The objective of this study was to use a structure-based approach to develop novel inhibitors of A. baumannii. Computational modeling was utilized to generate a homology model of one of the isoforms of A. baumannii OmpW (AB_OmpW2) based on the X-ray crystal structure of E. coli OmpW (EC_OmpW). Molecular docking studies were performed using the homology model to determine the binding pockets in the protein. Using a 3D stereoelectronic pharmacophore, in silico screening of the NCI database chemical library was performed for potential A. baumannii inhibitors. The antimicrobial activities of the identified compounds were then evaluated on the growth of a panel of clinical isolates of A. baumannii.
PDB database (http://www.rcsb.org). The protein sequence of the X-ray crystal structure of E. coli OmpW (EC_OmpW)24 (PDB code: 2F1 T; http://www.rcsb.org) was found to have a homology score of 59% (including identity and similarity of amino acids) with that of AB_OmpW2. Full sequence alignment of AB_OmpW2 and EC_OmpW (UniProt Accession No: P0A915) was done using CLUSTAL OMEGA (http://www.ebi.ac.uk). A 3 D homology model of AB_OmpW2 was then generated by the MODELLER program using the EC_OmpW structure (2F1T) as the template.25 The model was further energy minimized using AMBER force field until the RMS deviation was less than 0.09 kcal Å 1.26 2.3. In silico screening of inhibitors of A. baumannii Structural differences between AB_OmpW2 and EC_OmpW exist in the loop regions between the b-strands pointing to the extracellular surface of the outer membrane. One strategy to develop inhibitors against AB_OmpW2 is to plug the b-barrel channel. Thus, the homology model of AB_OmpW2, particularly the bbarrel region, was used for the in silico inhibitor screening. We analyzed the binding pockets in the b-barrel region of the energy minimized homology model of AB_OmpW2 and a promising inhibitor binding site was identified in the periplasmic side of the b-barrel structure, as shown in Fig. 1. The potential binding pocket is formed by the following fifteen amino acid residues (S108, T113, E115, N142, N144, E152, H231, Q233, R243, Y245, T315, D317, F323, V325 and R371). Seven of the residues (S108, Q233, R243, Y245, F323, V325 and R371) are located towards the inner side of the b-barrel structure. As there are two aromatic amino acid residues, Y245 and F323, we developed a three-point pharmacophore with a six-member aromatic ring as the first point and a short hydrophobic group of atoms or aromatic ring as the second point. The third point is a hydrogen bond acceptor or donor near
2. Materials and methods 2.1. Bacterial strains and reagents Clinical isolates of A. baumannii strains (AB5075, AB5711, AB3560, AB3638, AB0087, AB4052, AB5674, AB0967, AB4795, AB4957, AB585, and AB499) were obtained from the Multi-drugresistant organism Repository and Surveillance Network (MRSN) at Walter Reed Army Institute of Research. E. coli (EC25922) and A. baumannii (AB19606 and AB17978) strains were purchased from the American Type Culture Collection (ATCC, Manassas, VA, USA). The bacterial isolates were subcultured on blood agar plates 24 h prior to antimicrobial susceptibility tests. Small molecule inhibitors identified by virtual screening were obtained from the National Cancer Institute (NCI) repository. Bacterial culture media was purchased from Fisher Scientific (New Jersey, USA) and all other chemicals were obtained from other commercially available sources. Inhibitors stock solution were prepared in DMSO and stored at -80 °C. The final percentage of DMSO in the assays was less than 1% (v/v). 2.2. Homology model of AB_OmpW2 Currently, no X-ray crystal structure of AB_OmpW2 is available. The protein sequence of AB_OmpW2 (NCBI Accession No: 213057207) was used to identify homologous structures in the
Fig. 1. A homology model of A. baumannii outer membrane protein W2 generated using the program Modeller. The inhibitor-binding pocket is depicted in a spacefilled representation near the periplasmic side of the b-barrel structure.
I. Soojhawon et al. / Bioorganic & Medicinal Chemistry 25 (2017) 5477–5482
the side chain of Q233. Using this pharmacophore, we searched the NCI database to identify inhibitors that would bind to the promising pocket of AB_OmpW2. The search resulted in 3000 top ranking compounds out of 8073 compounds, matching within 0.5 angstrom of the three-point pharmacophore. The top 10 energetically favorable conformers of these compounds were docked into the pocket of AB_OmpW2 using AutoDock software27 and further rankordered based on the energy and surface-based scoring of the complex. 2.4. Antimicrobial susceptibility assays The minimum inhibitory concentration (MIC) of each novel compound was evaluated using the broth microdilution method according to the guidelines described in the Clinical and Laboratory Standards Institute (CLSI).28 MIC determination of each compound was carried out in 96-well round-bottom, polystyrene microtiter plates against a panel of A. baumannii strains (Table 1). In brief, the compounds were twofold diluted in cation-adjusted MuellerHinton Broth (CAMHB) medium to final concentrations ranging from 0.250 to 128 lg/mL. 100 ll of the drug containing broth was dispensed into the wells of the 96-well plate in triplicate. Wells containing only broth serve as the growth and sterility controls. Fresh overnight colonies of A. baumannii strains from 5% horse blood agar plates were suspended into phosphate buffered saline (PBS) and adjusted spectrophotometrically to an optical density of 0.1, at wavelength of 600 nm, to yield approximately 1 108 colony-forming units (CFU)/mL. The bacterial suspension was further diluted in CAMHB and 5 ll of the suspension was inoculated into respective wells to yield a starting inoculum concentration of 5 x 105 CFU/mL. The 96-well plates were incubated at 37 °C for 16 to 20 h. E. coli strain ATCC 25922 was used as the quality control strain with ampicillin as the antibiotic to verify accuracy of the susceptibility results. The MIC was recorded as the lowest concentration in lg/mL of the antimicrobial agent that inhibits the visible growth of the bacterial isolate after overnight incubation. 2.5. Time-kill kinetics assay The bactericidal (99.9% kill or a 3 log10 CFU/mL reduction in colony count from the initial inoculum) or bacteriostatic (<3 log10 CFU/mL reduction) nature of the lead compounds (D5, D12, and D6) was determined by time-kill kinetic assays. Briefly, 1 mL of cation-adjusted Mueller-Hinton Broth (CAMHB), containing the test compound with concentrations corresponding to 1, 2, 4 and 8 MIC, was seeded with an initial inoculum of 5 105 to Table 1 MIC of novel compounds against clinical of A. baumannii strains. A. baumannii isolate
MIC (lg/mL)
Strain
D5
D12
D6
AB19606 (ATCC) AB17978 (ATCC) AB5075 AB5711 AB3560 AB3638 AB0087 AB4052 AB5674 AB0967 AB4795 AB4957 AB585 AB499
16 16 16 32 32 16 4 32 16 16 8 32 8 1
16 16 16 16 32 16 2 32 16 16 8 32 8 1
16 16 32 16 16 16 8 16 16 16 16 16 8 8
5479
7 105 CFU/mL of A. baumannii strain AB5075, and incubated at 37 °C for 24 h, with constant shaking. At various time intervals (0, 6, 12 and 24 h), aliquots of the bacterial culture were removed, serially diluted, plated on Luria Broth (LB) agar, and incubated at 37 °C for 24 h. Colony counts were determined and kill curves were constructed by plotting Log10 CFU/mL vs time. The control sample contains only the test organism, AB5075 (no drug), and the experiment was performed in duplicate.29,30 Time-kill assays data were analyzed by determining the log10 decrease in the number of CFU/ mL achieved after a specific incubation time with respect to the original inoculum. A 1 log10 CFU/mL, 2 log10 CFU/mL and 3 log10 CFU/mL decrease represents 90%, 99% and 99.9% killing of the initial inoculum respectively.31 A compound is defined as bactericidal when it kills 99.9% (3 log10 CFU/mL) of the initial inoculum at a concentration no greater than four times the MIC or bacteriostatic when it kills <99% (<3 log10 CFU/mL reduction) of the bacteria.32,33 3. Results and discussion Emerging resistance to currently available antimicrobials has necessitated the urgent need for novel antibiotics effective against gram-negative bacteria. One of the strategies to combat antimicrobial resistance is to develop new classes of antibacterials, which act on novel drug-targets.34–36 However, achieving antibacterial activity against gram-negative pathogens is challenging, as the drug needs to penetrate the cell envelope without effluence by efflux pumps. Thus, in an effort to develop novel antibacterials against A. baumannii, we used an in silico screening approach with a homology model of the outer membrane protein W2 from the pathogen A. baumannii (AB_OmpW2), as the proposed drug target. 3.1. Homology modeling of A. baumannii OmpW2 The homology model of A. baumannii OmpW2 (AB_OmpW2) was generated from the X-ray crystal structure of E. coli OmpW (Fig. 1). The latter was used as a template, since both proteins share sequence homology, particularly in the eight-stranded b-barrel region. Structurally, the b-barrel-shaped outer membrane protein W forms a long and narrow channel as depicted in OmpW of E. coli.24 The amino acid sequence length for AB_OmpW2 is 372 residues whereas for EC_OmpW is 212. From CLUSTALW OMEGA sequence alignment, the first N-terminal 107 amino acid residues of AB_OmpW2 do not align with that of EC_OmpW and is located in the periplasmic side of the bacterial outer membrane. In addition, a gap of 47 amino acid residues (165–211) in AB_OmpW2 is found in the sequence alignment and is exposed to the extracellular surface of the outer membrane. The RMS deviation between Ca atoms of the residues forming the b-barrel together with the loops pointing to the periplasmic side of EC_OmpW and the Ca of the residues forming the b-barrel with the corresponding loops closer to the periplasmic side of AB_OmpW2 is 0.72 Å. Unlike in EC_OmpW, the loops connecting the b-barrel-barrel structure facing the extracellular side of the outer membrane of AB_OmpW2 are flexible and vary in size. Based on these structural features, we focused our in silico screening against the pocket in the b-barrel facing the periplasmic side. Fig. 1 shows the ribbon representation of the energy-minimized model of AB_OmpW2. The coordinates of the homology model are available upon request. 3.2. In silico screening of compounds The lower interior-binding pocket of the channel is hydrophobic as opposed to other outer membrane proteins, which are usually hydrophilic.24 Drugs would most likely bind to the lower interior-binding pocket to inhibit the growth of the pathogen. Thus,
5480
I. Soojhawon et al. / Bioorganic & Medicinal Chemistry 25 (2017) 5477–5482
using a 3D pharmacophore model and the AB_OmpW2 homology model, we screened around 3000 compounds from the NCI database. Based on scoring functions, the top ranking 100 compounds were selected and visually checked for their position and interactions with the amino acid residues in the binding pocket. Fifty compounds were selected for antimicrobial susceptibility testing. These compounds fulfilled the Lipinski’s rule of five: molar mass less than 500, hydrogen bond donors (–OH, NH) less than five, hydrogen bond acceptors (N, O) less than ten and calculated LogP less than five. The rule of five is beneficial to assess in vivo absorption abilities of any designed compounds.37 As mentioned earlier, the binding preference of the drugs would be near the periplasmic side of the outer membrane. Figure 2 shows one of the lead compounds, D5, depicted as ball-and-stick representation docked into the binding pocket of the homology model of AB_OmpW2. The atoms are color-coded: carbon (beige), oxygen (red), and nitrogen (blue), with the exception of the carbon atoms of the lead compound shown in magenta. The two hydrogen bonds are shown by dashed lines between the lead compound and amino acids residues, S108 and Q233, in the binding pocket. Compound D5 contains three chiral centers resulting in six isomers. However, when the compound was obtained from NCI, no information about the nature of isomers was provided. Our conformational analysis and docking studies revealed that two of the six isomers of D5 have ‘‘boat” conformation. Two of the remaining four isomers dock in similar orientation and have favorable docking scores. One of the isomers is shown in Figs. 2 and 3. Unlike D12, D5 is a chiral molecule, however, both compounds dock in the same orientation and location. The docking of D12 shows that it forms two hydrogen bonds; the first one with the side chain hydroxyl group (–OH) of residue T113 and the (–NH) of the amide group of D12. The second hydrogen bond is formed between amino group (–NH2) of Q233 and carbonyl oxygen (–C@O) of D12. The docking orientation of D6 is slightly different from that of D5 and D12, due to the presence of only one methylene group (short linker) between the two rings. However, all three compounds occupy the same pocket. D6 forms three hydrogen bonds; one of the hydroxyl groups of the compound forms hydrogen bonds with the amino (–NH2) side chain of R371 and hydroxyl group (–OH) of T113. The third hydro-
Fig. 3. Chemical structures of lead compounds with PubChem CID numbers and their computed DG (kcal/mol) values (docking scores). These compounds are designated as D5, D12 and D6.
gen bond is formed between the second hydroxyl group of D6 with side chain (–OH) of S108. 3.3. Determination of MIC of novel compounds against A. baumannii strains The antimicrobial susceptibility of a panel of clinical isolates of A. baumannii to the compounds was evaluated using the broth microdilution method. Three compounds, namely, D5, D12 and D6 displayed promising antibacterial activities. The MICs of the lead compounds are summarized in Table 1. All of the A. baumannii strains tested were susceptible to D5 and D12, with MIC ranges of 1–32 lg/mL respectively. Similarly, the bacterial strains were susceptible to D6, with MIC ranges of 8–32 lg/mL. It is to be noted that A. baumannii strain AB5075 is the most virulent and multidrug-resistant (MDR) clinical isolate38 and the MIC of D5 and D12 against this strain was 16 lg/mL, while that of D6 was 32 lg/mL. The chemical structure of the leads is shown in Fig. 3. D5 and D12 belong to the hydrazine class of compounds, whereas D6 is a phenolic derivative. 3.4. In vitro inhibitory activity of D5, D12 and D6 against AB5075 strain
Fig. 2. Compound D5 docked into AB_OmpW2. Dashed lines represents hydrogen bonds.
Time-kill kinetic assays were performed to determine the bactericidal (a 3 log10 CFU/mL reduction in colony count from initial inoculum) or bacteriostatic (<3 log10 CFU/mL reduction) nature of the compounds.32,33 Time-kill studies with D5, D12 and D6 were performed with the A. baumannii strain AB5075. The A. baumannii strains shown in Table 1 are multidrug-resistant and are equally important for inhibitor development against the pathogen. However AB5075 is the most virulent and multidrug-resistant (MDR) strain among them and it was selected for the kill-curve kinetics assay.38,39 D5 reduced the colony count from the initial bacterial inoculum, by 0.46, 0.75, 1.24 and 1.74 log10 CFU/mL at 1, 2, 4 and 8 MIC (16, 32, 64 & 128 lg/mL, respectively) after 24 h
I. Soojhawon et al. / Bioorganic & Medicinal Chemistry 25 (2017) 5477–5482
(Fig 4). Similarly, D12 reduced the initial inoculum by 0.73, 1.02, 1.42 and 2.1 log10 CFU/mL at 1, 2, 4 and 8 the MIC (16, 32, 64 & 128 lg/mL, respectively) after 24 h (Fig 5). However, exposure of AB5075 to D6 resulted in a 3 log10 CFU/mL (99.9%) reduction in colony count from the initial inoculum at 1 the MIC (32 lg/mL) after 30 min (Fig 6). D6 was found to be the most bactericidal compound, as 99.9% reduction of the initial inoculum was observed. D5 reduced at least 90% (1 log10 CFU/mL) of the initial inoculum at 4 MIC (64 lg/mL), whereas D12 caused similar reduction at 2 or 4 MIC (32 & 64 lg/mL). An antimicrobial compound is usually considered as bactericidal if its minimum bactericidal concentration (MBC) is no more than four times its MIC. In time–kill studies, the MBC is the minimal amount of the compound that results in 99.9% decrease of the initial inoculum within 24 h of incubation. Thus, in the case of D6, its MIC is also equal to the MBC (1 MIC, 32 lg/mL) as 99.9% reduction in colony count from initial inoculum is achieved. In general, the bactericidal activity of an antimicrobial compound tends to be related to its mechanism of action against a particular pathogen. Although 99.9% reduction of colony counts was not observed with exposure to D5 and D12, even at 8
5481
MIC (128 lg/mL), the inhibitory nature of these compounds could be further optimized by synthesizing new derivative. 4. Conclusions This is the first report where the homology model of the outer membrane protein W2 of A. baumannii (AB_OmpW2) was used for in silico screening of inhibitors of the pathogen. We successfully identified three novel inhibitors of multidrug-resistant A. baumannii, namely D5, D12 and D6 by in silico screening techniques, using the homology model of AB_OmpW2. The study evaluated the in vitro antibacterial activities of these three compounds against a panel of clinical isolates of A. baumannii strains. All the bacterial strains were found to be susceptible to the inhibitors with MIC ranges of 1 to 32 lg/mL. From time-kill kinetic studies with one of the most virulent strains (AB5075), treatment with D5 and D12 resulted in at least 90% reduction in colony count from the initial inoculum, whereas D6 was the most bactericidal (99.9%). Given the clinical relevancy of the strains tested in this study, these results are very promising for new antibacterial drug development. This result has provided two distinct chemical scaffolds, which would facilitate further design and optimization of the hit compounds by chemical synthesis and structure-activity relationship study supported by molecular docking calculations. Results from the docking studies will also allow further development of nonchiral analogs. Further, biophysical characterization of the interactions of hit compounds with AB_OmpW2 will be undertaken to substantiate its role as potential drug target. 5. Statistical analysis Statistical analysis was performed using GraphPad Prism software version 5.01 (La Jolla, California). Acknowledgements
Fig. 4. Time-kill curve of D5 against AB5075.
m m m m
The findings and opinions expressed herein belong to the authors and do not necessarily reflect the official views of the WRAIR, the U.S. Army, or the Department of Defense. The authors would like to thank Ms. Rae Heitkamp, Ms. Amy Summers and SGT Rebecca Fant for technical assistance. Funding This work was supported by a Military Infectious Diseases Research Program (MIDRP) grant awarded to Dr. S. Noble. References
Fig. 5. Time-kill curve of D12 against AB5075.
m
Fig. 6. Time-kill curve of D6 against AB5075.
1. Alsan M, Klompas M. Acinetobacter an emerging and important pathogen. J Clin Outcomes Manag. 2010;17:363–369. 2. Howard A, O’Donoghue M, Feeney A, Sleator RD. Acinetobacter baumannii: an emerging opportunistic pathogen. Virulence. 2012;3:243–250. 3. Peleg AY, Seifert H, Paterson DL. Acinetobacter baumannii: emergence of a successful pathogen. Clin Microbiol Rev. 2008;21:538–582. 4. Mortensen BL, Skaar EP. Host-microbe interactions that shape the pathogenesis of Acinetobacter baumannii infection. Cell Microbiol. 2012;14:1336–1344. 5. Vidal R, Dominguez M, Urrutia H, et al. Biofilm formation by Acinetobacter baumannii. Microbios. 1996;86:49–58. 6. Tomaras AP, Dorsey CW, Edelmann RE, Actis LA. Attachment to and biofilm formation on abiotic surfaces by Acinetobacter baumannii: involvement of a novel chaperone-usher pili assembly system. Microbiology. 2003;149:3473–3484. 7. Pour NK, Dusane DH, Dhakephalkar PK, Zamin FR, Zinjarde SS, Chopade BA. Biofilm formation by Acinetobacter baumannii strains isolated from urinary tract infection and urinary catheters. FEMS Immunol Med Microbiol. 2011;62:328–338. 8. Garnacho-Montero J, Ortiz-Leyba C, Fernandez-Hinojosa E, et al. Acinetobacter baumannii ventilator-associated pneumonia: epidemiological and clinical findings. Intensive Care Med. 2005;31:649–655.
5482
I. Soojhawon et al. / Bioorganic & Medicinal Chemistry 25 (2017) 5477–5482
9. Wisplinghoff H, Bischoff T, Tallent SM, Seifert H, Wenzel RP, Edmond MB. Nosocomial bloodstream infections in US hospitals: analysis of 24,179 cases from a prospective nationwide surveillance study. Clin Infect Dis. 2004;39:309–317. 10. Doyle JS, Buising KL, Thursky KA, Worth LJ, Richards MJ. Epidemiology of infections acquired in intensive care units. Semin Respir Crit Care Med. 2011;32:115–138. 11. Bergogne-Berezin E, Towner KJ. Acinetobacter spp. as nosocomial pathogens: microbiological, clinical, and epidemiological features. Clin Microbiol Rev. 1996;9:148–165. 12. Acinetobacter baumannii infections among patients at military medical facilities treating injured U.S. service members, 2002-2004. MMWR Morb Mortal Wkly Rep. 2004;53:1063–1066. 13. Hujer KM, Hujer AM, Hulten EA, et al. Analysis of antibiotic resistance genes in multidrug-resistant Acinetobacter sp. isolates from military and civilian patients treated at the Walter Reed Army Medical Center. Antimicrob Agents Chemother. 2006;50:4114–4123. 14. Dijkshoorn L, Nemec A, Seifert H. An increasing threat in hospitals: multidrugresistant Acinetobacter baumannii. Nat Rev Microbiol. 2007;5:939–951. 15. Joly-Guillou ML. Clinical impact and pathogenicity of Acinetobacter. Clin Microbiol Infect. 2005;11:868–873. 16. Davis KA, Moran KA, McAllister CK, Gray PJ. Multidrug-resistant Acinetobacter extremity infections in soldiers. Emerg Infect Dis. 2005;11:1218–1224. 17. Fournier PE, Vallenet D, Barbe V, et al. Comparative genomics of multidrug resistance in Acinetobacter baumannii. PLoS Genet. 2006;2:e7. 18. Keen 3rd EF, Murray CK, Robinson BJ, Hospenthal DR, Co EM, Aldous WK. Changes in the incidences of multidrug-resistant and extensively drugresistant organisms isolated in a military medical center. Infect Control Hosp Epidemiol. 2010;31:728–732. 19. Nandi B, Nandy RK, Sarkar A, Ghose AC. Structural features, properties and regulation of the outer-membrane protein W (OmpW) of Vibrio cholerae. Microbiology. 2005;151:2975–2986. 20. Pilsl H, Smajs D, Braun V. Characterization of colicin S4 and its receptor, OmpW, a minor protein of the Escherichia coli outer membrane. J Bacteriol. 1999;181:3578–3581. 21. Gil F, Ipinza F, Fuentes J, et al. The ompW (porin) gene mediates methyl viologen (paraquat) efflux in Salmonella enterica serovar typhimurium. Res Microbiol. 2007;158:529–536. 22. Neher TM, Lueking DR. Pseudomonas fluorescens ompW: plasmid localization and requirement for naphthalene uptake. Can J Microbiol. 2009;55:553–563. 23. Lin XM, Yang JN, Peng XX, Li H. A novel negative regulation mechanism of bacterial outer membrane proteins in response to antibiotic resistance. J Proteome Res. 2010;9:5952–5959.
24. Hong H, Patel DR, Tamm LK, van den Berg B. The outer membrane protein OmpW forms an eight-stranded beta-barrel with a hydrophobic channel. J Biol Chem. 2006;281:7568–7577. 25. Webb B, Sali A. Comparative protein structure modeling using MODELLER. Curr Protoc Bioinformatics. 2014;47. 5.6.1-32. 26. Case DA, Cheatham 3rd TE, Darden T, et al. The Amber biomolecular simulation programs. J Comput Chem. 2005;26:1668–1688. 27. Morris GM, Huey R, Lindstrom W, et al. AutoDock4 and AutoDockTools4: automated docking with selective receptor flexibility. J Comput Chem. 2009;30:2785–2791. 28. Methods for Dilution Antimicrobial Susceptibility Tests for Bacteria That Grow Aerobically; Approved Standard. Clinical and Laboratory Standards Institute. 2012; M07–A9. 29. Makobongo MO, Einck L, Peek Jr RM, Merrell DS. In vitro characterization of the anti-bacterial activity of SQ109 against Helicobacter pylori. PLoS ONE. 2013;8: e68917. 30. Methods for Determining Bactericidal Activity of Antimicrobial Agents; Approved Guideline. Clinical and Laboratory Standards Institute. 1999; M26–A. 31. Pankuch GA, Jacobs MR, Appelbaum PC. MIC and time-kill study of antipneumococcal activities of RPR 106972 (a new oral streptogramin), RP 59500 (quinupristin-dalfopristin), pyostacine (RP 7293), penicillin G, cefotaxime, erythromycin, and clarithromycin against 10 penicillinsusceptible and -resistant pneumococci. Antimicrob Agents Chemother. 1996;40:2071–2074. 32. Levison ME. Pharmacodynamics of antimicrobial drugs. Infect Dis Clin North Am. 2004;18:451–465. vii. 33. French GL. Bactericidal agents in the treatment of MRSA infections–the potential role of daptomycin. J Antimicrob Chemother. 2006;58:1107–1117. 34. Nathan C. Fresh approaches to anti-infective therapies. Sci Transl Med. 2012;4:140–142. 35. Livermore DM. The need for new antibiotics. Clin Microbiol Infect. 2004;10:1–9. 36. Freire-Moran L, Aronsson B, Manz C, et al. Critical shortage of new antibiotics in development against multidrug-resistant bacteria-Time to react is now. Drug Resist Updat. 2011;14:118–124. 37. Lipinski CA, Lombardo F, Dominy BW, Feeney PJ. Experimental and computational approaches to estimate solubility and permeability in drug discovery and development settings. Adv Drug Deliv Rev. 2001;46:3–26. 38. Jacobs AC, Thompson MG, Black CC, et al. AB5075, a Highly Virulent Isolate of Acinetobacter baumannii, as a Model Strain for the Evaluation of Pathogenesis and Antimicrobial Treatments. MBio. 2014;5:e01076–01014. 39. Huggins WM, Minrovic BM, Corey BW, et al. 1,2,4-Triazolidine-3-thiones as Narrow Spectrum Antibiotics against Multidrug-Resistant Acinetobacter baumannii. ACS Med Chem Lett. 2017;8:27–31.
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry journal homepage: www.elsevier.com/locate/bmc
Discovery of novel inhibitors of multidrug-resistant Acinetobacter baumannii Iswarduth Soojhawon a,d, Nagarajan Pattabiraman b,d, Arthur Tsang a, Amanda L. Roth c, Ellen Kang a, Schroeder M. Noble a,⇑ a b c
Wound Infections Department, Bacterial Diseases Branch, Walter Reed Army Institute of Research, Silver Spring, MD 20910, USA MolBox LLC, Silver Spring, MD 20910, USA Multidrug-resistant Organism Repository and Surveillance Network, Bacterial Diseases Branch, Walter Reed Army Institute of Research, Silver Spring, MD 20910, USA
a r t i c l e
i n f o
Article history: Received 10 May 2017 Revised 4 August 2017 Accepted 8 August 2017 Available online 9 August 2017 Keywords: Acinetobacter baumannii In silico Docking Antibacterial Outer membrane protein
a b s t r a c t The recent emergence of multidrug-resistant Acinetobacter baumannii strains and the non-efficacy of currently available antibiotics against such infections have led to an urgent need for the development of novel antibacterials. In an effort to address this problem, we have identified three novel inhibitors, namely, D5, D12 and D6 using in silico screening with a homology model of the outer membrane protein W2 (OmpW2) from A. baumannii, as the proposed new drug target. OmpW is an eight-stranded b-barrel protein involved in the transport of hydrophobic molecules across the outer membrane and maintenance of homeostasis under cellular stress. The antimicrobial activities of compounds D5, D12 and D6 were evaluated against a panel of clinical isolates of A. baumannii strains. These compounds inhibited the growth of the strains with minimum inhibitory concentration (MIC) ranges of 1–32 lg/mL. Time-kill kinetic studies with the highly virulent and multidrug-resistant strain, A. baumannii 5075, indicated that D6 exhibited the highest bactericidal activity as a 3 log10 CFU/mL (99.9%) reduction in colony count from the initial inoculum was observed after 30 min incubation. D5 and D12 reduced at least 1 log10 CFU/mL (90%) of the initial inoculum after 24 h. In conclusion, these three lead inhibitors have provided two distinct chemical scaffolds for further analog design and optimizations, using chemical synthesis, to develop more potent inhibitors of the pathogen. Published by Elsevier Ltd.
1. Introduction The rapid emergence of multidrug-resistant (MDR) bacteria threatens the effectiveness of currently available antibiotics used to treat the civilian population and deployed forces infected with these pathogens. To combat resistance, new antibacterials with novel modes of action or binding sites are urgently required, especially against drug resistant Acinetobacter baumannii infections, which are difficult to treat. A. baumannii is a gram-negative, aerobic, non-fermentative, catalase-positive, oxidase-negative, coccobacillus opportunistic pathogen.1–4 This pathogen can survive hospital environments due to its ability to form biofilms on abiotic surfaces, such as catheters, ventilators, and medical devices etc.5–7 A. baumannii is a major cause of hospital-associated infections such Abbreviations: AB_OmpW2, A. baumannii OmpW; MIC, minimum inhibitory concentration; MDR, multidrug-resistant; CFU, colony-forming units. ⇑ Corresponding author. E-mail address: [email protected] (S.M. Noble). d Equally Contributed. http://dx.doi.org/10.1016/j.bmc.2017.08.014 0968-0896/Published by Elsevier Ltd.
as ventilator-associated pneumonia (VAP),8 skin,1 wound,1 urinary tract,4 bacteremia,9 meningitis10 and soft tissue11 infections in patients within civilian and military heath care centers. Interest in A. baumannii (MDR) arose after it was detected in infected combat troops returning from war zones.12,13 The consequences of such infections are prolonged wound healing, which sometimes lead to amputations and accounts for the increased morbidity rate.13 A. baumannii is reported to be responsible for more than 10% of all hospital-acquired infections in the United States and has a greater than 50% mortality rate in patients with sepsis and pneumonia.14–16 A. baumannii was initially sensitive to most available antibiotics; however, it is now resistant to most first-line antibiotics.17 One of the distinguishing features of the bacterium, which made it a ‘‘red-alert’’ human pathogen, is its impressive array of acquired antibiotic resistance mechanisms.3 Indeed, most A. baumannii isolates collected in military and civilian health care centers were often sensitive to carbapenems in the 1990s, but now demonstrate decreased sensitivity to carbapenem, if not outright resistance to these last resort antibiotics.18
5478
I. Soojhawon et al. / Bioorganic & Medicinal Chemistry 25 (2017) 5477–5482
Currently available antibiotics are designed to target cellular functions, such as cell wall, DNA, RNA, or protein biosynthesis in bacterial pathogens. However, the decreased susceptibility of A. baumannii strains to cephalosporins, aminoglycosides, quinolones and, lately, carbapenems has necessitated the development of novel antibacterials, which act on new drug targets crucial for bacterial growth. Promising drug targets in gram-negative bacteria include the outer membrane proteins (OMPs), specifically the outer membrane protein W (OmpW). OmpW is an eight-stranded b-barrel shaped outer membrane protein, which is widespread in many gram-negative bacteria, such as Vibrio cholerae, Escherichia coli, and A. baumannii.19,20 OmpW protein is believed to be involved in the transport of hydrophobic or aromatic molecules across the outer membrane.20–22 It is also implicated in the maintenance of homeostasis (osmoregulation) under cellular stress. Studies have demonstrated that E. coli colicin S4 exploits OmpW for entry into the bacterial cell and eventually kills the bacterium through pore formation. Deletion of OmpW from E. coli was associated with decreased MICs of several antibiotics and the protein was found to be down regulated when E. coli was exposed to chlortetracycline.23 Moreover, mutant strains of V. cholerae, which expressed no OmpW, were shown to have reduced colonization in the intestine of mice relative to those that expressed OmpW.19 All these aforementioned studies indicate that OmpW could be a potential drug target in gram-negative bacteria. The objective of this study was to use a structure-based approach to develop novel inhibitors of A. baumannii. Computational modeling was utilized to generate a homology model of one of the isoforms of A. baumannii OmpW (AB_OmpW2) based on the X-ray crystal structure of E. coli OmpW (EC_OmpW). Molecular docking studies were performed using the homology model to determine the binding pockets in the protein. Using a 3D stereoelectronic pharmacophore, in silico screening of the NCI database chemical library was performed for potential A. baumannii inhibitors. The antimicrobial activities of the identified compounds were then evaluated on the growth of a panel of clinical isolates of A. baumannii.
PDB database (http://www.rcsb.org). The protein sequence of the X-ray crystal structure of E. coli OmpW (EC_OmpW)24 (PDB code: 2F1 T; http://www.rcsb.org) was found to have a homology score of 59% (including identity and similarity of amino acids) with that of AB_OmpW2. Full sequence alignment of AB_OmpW2 and EC_OmpW (UniProt Accession No: P0A915) was done using CLUSTAL OMEGA (http://www.ebi.ac.uk). A 3 D homology model of AB_OmpW2 was then generated by the MODELLER program using the EC_OmpW structure (2F1T) as the template.25 The model was further energy minimized using AMBER force field until the RMS deviation was less than 0.09 kcal Å 1.26 2.3. In silico screening of inhibitors of A. baumannii Structural differences between AB_OmpW2 and EC_OmpW exist in the loop regions between the b-strands pointing to the extracellular surface of the outer membrane. One strategy to develop inhibitors against AB_OmpW2 is to plug the b-barrel channel. Thus, the homology model of AB_OmpW2, particularly the bbarrel region, was used for the in silico inhibitor screening. We analyzed the binding pockets in the b-barrel region of the energy minimized homology model of AB_OmpW2 and a promising inhibitor binding site was identified in the periplasmic side of the b-barrel structure, as shown in Fig. 1. The potential binding pocket is formed by the following fifteen amino acid residues (S108, T113, E115, N142, N144, E152, H231, Q233, R243, Y245, T315, D317, F323, V325 and R371). Seven of the residues (S108, Q233, R243, Y245, F323, V325 and R371) are located towards the inner side of the b-barrel structure. As there are two aromatic amino acid residues, Y245 and F323, we developed a three-point pharmacophore with a six-member aromatic ring as the first point and a short hydrophobic group of atoms or aromatic ring as the second point. The third point is a hydrogen bond acceptor or donor near
2. Materials and methods 2.1. Bacterial strains and reagents Clinical isolates of A. baumannii strains (AB5075, AB5711, AB3560, AB3638, AB0087, AB4052, AB5674, AB0967, AB4795, AB4957, AB585, and AB499) were obtained from the Multi-drugresistant organism Repository and Surveillance Network (MRSN) at Walter Reed Army Institute of Research. E. coli (EC25922) and A. baumannii (AB19606 and AB17978) strains were purchased from the American Type Culture Collection (ATCC, Manassas, VA, USA). The bacterial isolates were subcultured on blood agar plates 24 h prior to antimicrobial susceptibility tests. Small molecule inhibitors identified by virtual screening were obtained from the National Cancer Institute (NCI) repository. Bacterial culture media was purchased from Fisher Scientific (New Jersey, USA) and all other chemicals were obtained from other commercially available sources. Inhibitors stock solution were prepared in DMSO and stored at -80 °C. The final percentage of DMSO in the assays was less than 1% (v/v). 2.2. Homology model of AB_OmpW2 Currently, no X-ray crystal structure of AB_OmpW2 is available. The protein sequence of AB_OmpW2 (NCBI Accession No: 213057207) was used to identify homologous structures in the
Fig. 1. A homology model of A. baumannii outer membrane protein W2 generated using the program Modeller. The inhibitor-binding pocket is depicted in a spacefilled representation near the periplasmic side of the b-barrel structure.
I. Soojhawon et al. / Bioorganic & Medicinal Chemistry 25 (2017) 5477–5482
the side chain of Q233. Using this pharmacophore, we searched the NCI database to identify inhibitors that would bind to the promising pocket of AB_OmpW2. The search resulted in 3000 top ranking compounds out of 8073 compounds, matching within 0.5 angstrom of the three-point pharmacophore. The top 10 energetically favorable conformers of these compounds were docked into the pocket of AB_OmpW2 using AutoDock software27 and further rankordered based on the energy and surface-based scoring of the complex. 2.4. Antimicrobial susceptibility assays The minimum inhibitory concentration (MIC) of each novel compound was evaluated using the broth microdilution method according to the guidelines described in the Clinical and Laboratory Standards Institute (CLSI).28 MIC determination of each compound was carried out in 96-well round-bottom, polystyrene microtiter plates against a panel of A. baumannii strains (Table 1). In brief, the compounds were twofold diluted in cation-adjusted MuellerHinton Broth (CAMHB) medium to final concentrations ranging from 0.250 to 128 lg/mL. 100 ll of the drug containing broth was dispensed into the wells of the 96-well plate in triplicate. Wells containing only broth serve as the growth and sterility controls. Fresh overnight colonies of A. baumannii strains from 5% horse blood agar plates were suspended into phosphate buffered saline (PBS) and adjusted spectrophotometrically to an optical density of 0.1, at wavelength of 600 nm, to yield approximately 1 108 colony-forming units (CFU)/mL. The bacterial suspension was further diluted in CAMHB and 5 ll of the suspension was inoculated into respective wells to yield a starting inoculum concentration of 5 x 105 CFU/mL. The 96-well plates were incubated at 37 °C for 16 to 20 h. E. coli strain ATCC 25922 was used as the quality control strain with ampicillin as the antibiotic to verify accuracy of the susceptibility results. The MIC was recorded as the lowest concentration in lg/mL of the antimicrobial agent that inhibits the visible growth of the bacterial isolate after overnight incubation. 2.5. Time-kill kinetics assay The bactericidal (99.9% kill or a 3 log10 CFU/mL reduction in colony count from the initial inoculum) or bacteriostatic (<3 log10 CFU/mL reduction) nature of the lead compounds (D5, D12, and D6) was determined by time-kill kinetic assays. Briefly, 1 mL of cation-adjusted Mueller-Hinton Broth (CAMHB), containing the test compound with concentrations corresponding to 1, 2, 4 and 8 MIC, was seeded with an initial inoculum of 5 105 to Table 1 MIC of novel compounds against clinical of A. baumannii strains. A. baumannii isolate
MIC (lg/mL)
Strain
D5
D12
D6
AB19606 (ATCC) AB17978 (ATCC) AB5075 AB5711 AB3560 AB3638 AB0087 AB4052 AB5674 AB0967 AB4795 AB4957 AB585 AB499
16 16 16 32 32 16 4 32 16 16 8 32 8 1
16 16 16 16 32 16 2 32 16 16 8 32 8 1
16 16 32 16 16 16 8 16 16 16 16 16 8 8
5479
7 105 CFU/mL of A. baumannii strain AB5075, and incubated at 37 °C for 24 h, with constant shaking. At various time intervals (0, 6, 12 and 24 h), aliquots of the bacterial culture were removed, serially diluted, plated on Luria Broth (LB) agar, and incubated at 37 °C for 24 h. Colony counts were determined and kill curves were constructed by plotting Log10 CFU/mL vs time. The control sample contains only the test organism, AB5075 (no drug), and the experiment was performed in duplicate.29,30 Time-kill assays data were analyzed by determining the log10 decrease in the number of CFU/ mL achieved after a specific incubation time with respect to the original inoculum. A 1 log10 CFU/mL, 2 log10 CFU/mL and 3 log10 CFU/mL decrease represents 90%, 99% and 99.9% killing of the initial inoculum respectively.31 A compound is defined as bactericidal when it kills 99.9% (3 log10 CFU/mL) of the initial inoculum at a concentration no greater than four times the MIC or bacteriostatic when it kills <99% (<3 log10 CFU/mL reduction) of the bacteria.32,33 3. Results and discussion Emerging resistance to currently available antimicrobials has necessitated the urgent need for novel antibiotics effective against gram-negative bacteria. One of the strategies to combat antimicrobial resistance is to develop new classes of antibacterials, which act on novel drug-targets.34–36 However, achieving antibacterial activity against gram-negative pathogens is challenging, as the drug needs to penetrate the cell envelope without effluence by efflux pumps. Thus, in an effort to develop novel antibacterials against A. baumannii, we used an in silico screening approach with a homology model of the outer membrane protein W2 from the pathogen A. baumannii (AB_OmpW2), as the proposed drug target. 3.1. Homology modeling of A. baumannii OmpW2 The homology model of A. baumannii OmpW2 (AB_OmpW2) was generated from the X-ray crystal structure of E. coli OmpW (Fig. 1). The latter was used as a template, since both proteins share sequence homology, particularly in the eight-stranded b-barrel region. Structurally, the b-barrel-shaped outer membrane protein W forms a long and narrow channel as depicted in OmpW of E. coli.24 The amino acid sequence length for AB_OmpW2 is 372 residues whereas for EC_OmpW is 212. From CLUSTALW OMEGA sequence alignment, the first N-terminal 107 amino acid residues of AB_OmpW2 do not align with that of EC_OmpW and is located in the periplasmic side of the bacterial outer membrane. In addition, a gap of 47 amino acid residues (165–211) in AB_OmpW2 is found in the sequence alignment and is exposed to the extracellular surface of the outer membrane. The RMS deviation between Ca atoms of the residues forming the b-barrel together with the loops pointing to the periplasmic side of EC_OmpW and the Ca of the residues forming the b-barrel with the corresponding loops closer to the periplasmic side of AB_OmpW2 is 0.72 Å. Unlike in EC_OmpW, the loops connecting the b-barrel-barrel structure facing the extracellular side of the outer membrane of AB_OmpW2 are flexible and vary in size. Based on these structural features, we focused our in silico screening against the pocket in the b-barrel facing the periplasmic side. Fig. 1 shows the ribbon representation of the energy-minimized model of AB_OmpW2. The coordinates of the homology model are available upon request. 3.2. In silico screening of compounds The lower interior-binding pocket of the channel is hydrophobic as opposed to other outer membrane proteins, which are usually hydrophilic.24 Drugs would most likely bind to the lower interior-binding pocket to inhibit the growth of the pathogen. Thus,
5480
I. Soojhawon et al. / Bioorganic & Medicinal Chemistry 25 (2017) 5477–5482
using a 3D pharmacophore model and the AB_OmpW2 homology model, we screened around 3000 compounds from the NCI database. Based on scoring functions, the top ranking 100 compounds were selected and visually checked for their position and interactions with the amino acid residues in the binding pocket. Fifty compounds were selected for antimicrobial susceptibility testing. These compounds fulfilled the Lipinski’s rule of five: molar mass less than 500, hydrogen bond donors (–OH, NH) less than five, hydrogen bond acceptors (N, O) less than ten and calculated LogP less than five. The rule of five is beneficial to assess in vivo absorption abilities of any designed compounds.37 As mentioned earlier, the binding preference of the drugs would be near the periplasmic side of the outer membrane. Figure 2 shows one of the lead compounds, D5, depicted as ball-and-stick representation docked into the binding pocket of the homology model of AB_OmpW2. The atoms are color-coded: carbon (beige), oxygen (red), and nitrogen (blue), with the exception of the carbon atoms of the lead compound shown in magenta. The two hydrogen bonds are shown by dashed lines between the lead compound and amino acids residues, S108 and Q233, in the binding pocket. Compound D5 contains three chiral centers resulting in six isomers. However, when the compound was obtained from NCI, no information about the nature of isomers was provided. Our conformational analysis and docking studies revealed that two of the six isomers of D5 have ‘‘boat” conformation. Two of the remaining four isomers dock in similar orientation and have favorable docking scores. One of the isomers is shown in Figs. 2 and 3. Unlike D12, D5 is a chiral molecule, however, both compounds dock in the same orientation and location. The docking of D12 shows that it forms two hydrogen bonds; the first one with the side chain hydroxyl group (–OH) of residue T113 and the (–NH) of the amide group of D12. The second hydrogen bond is formed between amino group (–NH2) of Q233 and carbonyl oxygen (–C@O) of D12. The docking orientation of D6 is slightly different from that of D5 and D12, due to the presence of only one methylene group (short linker) between the two rings. However, all three compounds occupy the same pocket. D6 forms three hydrogen bonds; one of the hydroxyl groups of the compound forms hydrogen bonds with the amino (–NH2) side chain of R371 and hydroxyl group (–OH) of T113. The third hydro-
Fig. 3. Chemical structures of lead compounds with PubChem CID numbers and their computed DG (kcal/mol) values (docking scores). These compounds are designated as D5, D12 and D6.
gen bond is formed between the second hydroxyl group of D6 with side chain (–OH) of S108. 3.3. Determination of MIC of novel compounds against A. baumannii strains The antimicrobial susceptibility of a panel of clinical isolates of A. baumannii to the compounds was evaluated using the broth microdilution method. Three compounds, namely, D5, D12 and D6 displayed promising antibacterial activities. The MICs of the lead compounds are summarized in Table 1. All of the A. baumannii strains tested were susceptible to D5 and D12, with MIC ranges of 1–32 lg/mL respectively. Similarly, the bacterial strains were susceptible to D6, with MIC ranges of 8–32 lg/mL. It is to be noted that A. baumannii strain AB5075 is the most virulent and multidrug-resistant (MDR) clinical isolate38 and the MIC of D5 and D12 against this strain was 16 lg/mL, while that of D6 was 32 lg/mL. The chemical structure of the leads is shown in Fig. 3. D5 and D12 belong to the hydrazine class of compounds, whereas D6 is a phenolic derivative. 3.4. In vitro inhibitory activity of D5, D12 and D6 against AB5075 strain
Fig. 2. Compound D5 docked into AB_OmpW2. Dashed lines represents hydrogen bonds.
Time-kill kinetic assays were performed to determine the bactericidal (a 3 log10 CFU/mL reduction in colony count from initial inoculum) or bacteriostatic (<3 log10 CFU/mL reduction) nature of the compounds.32,33 Time-kill studies with D5, D12 and D6 were performed with the A. baumannii strain AB5075. The A. baumannii strains shown in Table 1 are multidrug-resistant and are equally important for inhibitor development against the pathogen. However AB5075 is the most virulent and multidrug-resistant (MDR) strain among them and it was selected for the kill-curve kinetics assay.38,39 D5 reduced the colony count from the initial bacterial inoculum, by 0.46, 0.75, 1.24 and 1.74 log10 CFU/mL at 1, 2, 4 and 8 MIC (16, 32, 64 & 128 lg/mL, respectively) after 24 h
I. Soojhawon et al. / Bioorganic & Medicinal Chemistry 25 (2017) 5477–5482
(Fig 4). Similarly, D12 reduced the initial inoculum by 0.73, 1.02, 1.42 and 2.1 log10 CFU/mL at 1, 2, 4 and 8 the MIC (16, 32, 64 & 128 lg/mL, respectively) after 24 h (Fig 5). However, exposure of AB5075 to D6 resulted in a 3 log10 CFU/mL (99.9%) reduction in colony count from the initial inoculum at 1 the MIC (32 lg/mL) after 30 min (Fig 6). D6 was found to be the most bactericidal compound, as 99.9% reduction of the initial inoculum was observed. D5 reduced at least 90% (1 log10 CFU/mL) of the initial inoculum at 4 MIC (64 lg/mL), whereas D12 caused similar reduction at 2 or 4 MIC (32 & 64 lg/mL). An antimicrobial compound is usually considered as bactericidal if its minimum bactericidal concentration (MBC) is no more than four times its MIC. In time–kill studies, the MBC is the minimal amount of the compound that results in 99.9% decrease of the initial inoculum within 24 h of incubation. Thus, in the case of D6, its MIC is also equal to the MBC (1 MIC, 32 lg/mL) as 99.9% reduction in colony count from initial inoculum is achieved. In general, the bactericidal activity of an antimicrobial compound tends to be related to its mechanism of action against a particular pathogen. Although 99.9% reduction of colony counts was not observed with exposure to D5 and D12, even at 8
5481
MIC (128 lg/mL), the inhibitory nature of these compounds could be further optimized by synthesizing new derivative. 4. Conclusions This is the first report where the homology model of the outer membrane protein W2 of A. baumannii (AB_OmpW2) was used for in silico screening of inhibitors of the pathogen. We successfully identified three novel inhibitors of multidrug-resistant A. baumannii, namely D5, D12 and D6 by in silico screening techniques, using the homology model of AB_OmpW2. The study evaluated the in vitro antibacterial activities of these three compounds against a panel of clinical isolates of A. baumannii strains. All the bacterial strains were found to be susceptible to the inhibitors with MIC ranges of 1 to 32 lg/mL. From time-kill kinetic studies with one of the most virulent strains (AB5075), treatment with D5 and D12 resulted in at least 90% reduction in colony count from the initial inoculum, whereas D6 was the most bactericidal (99.9%). Given the clinical relevancy of the strains tested in this study, these results are very promising for new antibacterial drug development. This result has provided two distinct chemical scaffolds, which would facilitate further design and optimization of the hit compounds by chemical synthesis and structure-activity relationship study supported by molecular docking calculations. Results from the docking studies will also allow further development of nonchiral analogs. Further, biophysical characterization of the interactions of hit compounds with AB_OmpW2 will be undertaken to substantiate its role as potential drug target. 5. Statistical analysis Statistical analysis was performed using GraphPad Prism software version 5.01 (La Jolla, California). Acknowledgements
Fig. 4. Time-kill curve of D5 against AB5075.
m m m m
The findings and opinions expressed herein belong to the authors and do not necessarily reflect the official views of the WRAIR, the U.S. Army, or the Department of Defense. The authors would like to thank Ms. Rae Heitkamp, Ms. Amy Summers and SGT Rebecca Fant for technical assistance. Funding This work was supported by a Military Infectious Diseases Research Program (MIDRP) grant awarded to Dr. S. Noble. References
Fig. 5. Time-kill curve of D12 against AB5075.
m
Fig. 6. Time-kill curve of D6 against AB5075.
1. Alsan M, Klompas M. Acinetobacter an emerging and important pathogen. J Clin Outcomes Manag. 2010;17:363–369. 2. Howard A, O’Donoghue M, Feeney A, Sleator RD. Acinetobacter baumannii: an emerging opportunistic pathogen. Virulence. 2012;3:243–250. 3. Peleg AY, Seifert H, Paterson DL. Acinetobacter baumannii: emergence of a successful pathogen. Clin Microbiol Rev. 2008;21:538–582. 4. Mortensen BL, Skaar EP. Host-microbe interactions that shape the pathogenesis of Acinetobacter baumannii infection. Cell Microbiol. 2012;14:1336–1344. 5. Vidal R, Dominguez M, Urrutia H, et al. Biofilm formation by Acinetobacter baumannii. Microbios. 1996;86:49–58. 6. Tomaras AP, Dorsey CW, Edelmann RE, Actis LA. Attachment to and biofilm formation on abiotic surfaces by Acinetobacter baumannii: involvement of a novel chaperone-usher pili assembly system. Microbiology. 2003;149:3473–3484. 7. Pour NK, Dusane DH, Dhakephalkar PK, Zamin FR, Zinjarde SS, Chopade BA. Biofilm formation by Acinetobacter baumannii strains isolated from urinary tract infection and urinary catheters. FEMS Immunol Med Microbiol. 2011;62:328–338. 8. Garnacho-Montero J, Ortiz-Leyba C, Fernandez-Hinojosa E, et al. Acinetobacter baumannii ventilator-associated pneumonia: epidemiological and clinical findings. Intensive Care Med. 2005;31:649–655.
5482
I. Soojhawon et al. / Bioorganic & Medicinal Chemistry 25 (2017) 5477–5482
9. Wisplinghoff H, Bischoff T, Tallent SM, Seifert H, Wenzel RP, Edmond MB. Nosocomial bloodstream infections in US hospitals: analysis of 24,179 cases from a prospective nationwide surveillance study. Clin Infect Dis. 2004;39:309–317. 10. Doyle JS, Buising KL, Thursky KA, Worth LJ, Richards MJ. Epidemiology of infections acquired in intensive care units. Semin Respir Crit Care Med. 2011;32:115–138. 11. Bergogne-Berezin E, Towner KJ. Acinetobacter spp. as nosocomial pathogens: microbiological, clinical, and epidemiological features. Clin Microbiol Rev. 1996;9:148–165. 12. Acinetobacter baumannii infections among patients at military medical facilities treating injured U.S. service members, 2002-2004. MMWR Morb Mortal Wkly Rep. 2004;53:1063–1066. 13. Hujer KM, Hujer AM, Hulten EA, et al. Analysis of antibiotic resistance genes in multidrug-resistant Acinetobacter sp. isolates from military and civilian patients treated at the Walter Reed Army Medical Center. Antimicrob Agents Chemother. 2006;50:4114–4123. 14. Dijkshoorn L, Nemec A, Seifert H. An increasing threat in hospitals: multidrugresistant Acinetobacter baumannii. Nat Rev Microbiol. 2007;5:939–951. 15. Joly-Guillou ML. Clinical impact and pathogenicity of Acinetobacter. Clin Microbiol Infect. 2005;11:868–873. 16. Davis KA, Moran KA, McAllister CK, Gray PJ. Multidrug-resistant Acinetobacter extremity infections in soldiers. Emerg Infect Dis. 2005;11:1218–1224. 17. Fournier PE, Vallenet D, Barbe V, et al. Comparative genomics of multidrug resistance in Acinetobacter baumannii. PLoS Genet. 2006;2:e7. 18. Keen 3rd EF, Murray CK, Robinson BJ, Hospenthal DR, Co EM, Aldous WK. Changes in the incidences of multidrug-resistant and extensively drugresistant organisms isolated in a military medical center. Infect Control Hosp Epidemiol. 2010;31:728–732. 19. Nandi B, Nandy RK, Sarkar A, Ghose AC. Structural features, properties and regulation of the outer-membrane protein W (OmpW) of Vibrio cholerae. Microbiology. 2005;151:2975–2986. 20. Pilsl H, Smajs D, Braun V. Characterization of colicin S4 and its receptor, OmpW, a minor protein of the Escherichia coli outer membrane. J Bacteriol. 1999;181:3578–3581. 21. Gil F, Ipinza F, Fuentes J, et al. The ompW (porin) gene mediates methyl viologen (paraquat) efflux in Salmonella enterica serovar typhimurium. Res Microbiol. 2007;158:529–536. 22. Neher TM, Lueking DR. Pseudomonas fluorescens ompW: plasmid localization and requirement for naphthalene uptake. Can J Microbiol. 2009;55:553–563. 23. Lin XM, Yang JN, Peng XX, Li H. A novel negative regulation mechanism of bacterial outer membrane proteins in response to antibiotic resistance. J Proteome Res. 2010;9:5952–5959.
24. Hong H, Patel DR, Tamm LK, van den Berg B. The outer membrane protein OmpW forms an eight-stranded beta-barrel with a hydrophobic channel. J Biol Chem. 2006;281:7568–7577. 25. Webb B, Sali A. Comparative protein structure modeling using MODELLER. Curr Protoc Bioinformatics. 2014;47. 5.6.1-32. 26. Case DA, Cheatham 3rd TE, Darden T, et al. The Amber biomolecular simulation programs. J Comput Chem. 2005;26:1668–1688. 27. Morris GM, Huey R, Lindstrom W, et al. AutoDock4 and AutoDockTools4: automated docking with selective receptor flexibility. J Comput Chem. 2009;30:2785–2791. 28. Methods for Dilution Antimicrobial Susceptibility Tests for Bacteria That Grow Aerobically; Approved Standard. Clinical and Laboratory Standards Institute. 2012; M07–A9. 29. Makobongo MO, Einck L, Peek Jr RM, Merrell DS. In vitro characterization of the anti-bacterial activity of SQ109 against Helicobacter pylori. PLoS ONE. 2013;8: e68917. 30. Methods for Determining Bactericidal Activity of Antimicrobial Agents; Approved Guideline. Clinical and Laboratory Standards Institute. 1999; M26–A. 31. Pankuch GA, Jacobs MR, Appelbaum PC. MIC and time-kill study of antipneumococcal activities of RPR 106972 (a new oral streptogramin), RP 59500 (quinupristin-dalfopristin), pyostacine (RP 7293), penicillin G, cefotaxime, erythromycin, and clarithromycin against 10 penicillinsusceptible and -resistant pneumococci. Antimicrob Agents Chemother. 1996;40:2071–2074. 32. Levison ME. Pharmacodynamics of antimicrobial drugs. Infect Dis Clin North Am. 2004;18:451–465. vii. 33. French GL. Bactericidal agents in the treatment of MRSA infections–the potential role of daptomycin. J Antimicrob Chemother. 2006;58:1107–1117. 34. Nathan C. Fresh approaches to anti-infective therapies. Sci Transl Med. 2012;4:140–142. 35. Livermore DM. The need for new antibiotics. Clin Microbiol Infect. 2004;10:1–9. 36. Freire-Moran L, Aronsson B, Manz C, et al. Critical shortage of new antibiotics in development against multidrug-resistant bacteria-Time to react is now. Drug Resist Updat. 2011;14:118–124. 37. Lipinski CA, Lombardo F, Dominy BW, Feeney PJ. Experimental and computational approaches to estimate solubility and permeability in drug discovery and development settings. Adv Drug Deliv Rev. 2001;46:3–26. 38. Jacobs AC, Thompson MG, Black CC, et al. AB5075, a Highly Virulent Isolate of Acinetobacter baumannii, as a Model Strain for the Evaluation of Pathogenesis and Antimicrobial Treatments. MBio. 2014;5:e01076–01014. 39. Huggins WM, Minrovic BM, Corey BW, et al. 1,2,4-Triazolidine-3-thiones as Narrow Spectrum Antibiotics against Multidrug-Resistant Acinetobacter baumannii. ACS Med Chem Lett. 2017;8:27–31.