- Email: [email protected]
Structural signature of Ser83Leu and Asp87Asn mutations in DNA gyrase from enterotoxigenic Escherichia coli and impact on quinolone resistance
GENE-40881; No. of pages: 8; 4C: Gene xxx (2015) xxx–xxx
Contents lists available at ScienceDirect
Gene journal homepage: www.elsevier.com/locate/gene
Research paper
Structural signature of Ser83Leu and Asp87Asn mutations in DNA gyrase from enterotoxigenic Escherichia coli and impact on quinolone resistance Kusum Mehla, Jayashree Ramana ⁎ Department of Biotechnology and Bioinformatics, Jaypee University of Information Technology, Solan PIN-173234, Himachal Pradesh, India
a r t i c l e
i n f o
Article history: Received 10 March 2015 Received in revised form 22 September 2015 Accepted 23 September 2015 Available online xxxx Keywords: Diarrhea Drug resistance Amino acid substitutions Homology modeling Binding affinity
a b s t r a c t Enterotoxigenic Escherichia coli (ETEC) is among the most frequent microorganisms causing traveler's diarrhea (TD). Quinolones are potent antimicrobial agents used for the treatment of TD. Resistance to quinolones is typically caused by substitutions in QRDR region of gyrA subunit of DNA gyrase. The aim of this study was to seek insights into the effect of these substitutions at structural level and their association with observed quinolone resistance. Majority of the ETEC strains have gyrA mutations at amino acid position 83 and 87. To understand the quinolone resistance mechanism at molecular level, we have studied the interaction of wild type and mutant forms of ETEC gyrA with nalidixic acid and ciprofloxacin by molecular modeling using Discovery Studio and LeadIt. All the mutants had reduced affinity towards both ciprofloxacin and nalidixic acid relative to the wild type due to the mutations introduced in gyrA. Besides Ser83 and Asp87, for nalidixic acid binding Arg91 and His45 residues were observed to be critical while in ciprofloxacin binding Lys42 and Arg91 residues played a significant role. Amino acid substitutions contribute to the emergence of drug resistance in sensitive strains by causing structural alterations leading to reduced affinity of the drug towards receptor. Analysis of the effect of amino acid substitutions at structural level is of utmost importance to establish possible associations between mutations and the diseases. These studies accelerate the identification of pharmaceutical targets for relevant treatments and could also be helpful in guiding the design of further experimental research. © 2015 Elsevier B.V. All rights reserved.
1. Introduction Progressive increase in antimicrobial resistance among enteric pathogens has become a critical area of public concern. Enterotoxigenic Escherichia coli (ETEC) is the leading bacterial cause of diarrhea in developing world where inadequate clean water and poor sanitation facilities are prevailing and a predominant cause of TD (Sizemore et al., 2004). Enteroaggregative E. coli, Shigella spp. and Salmonella spp. are other common bacterial agents, while Campylobacter, Yersinia, Aeromonas, and Plesiomonas spp. cases are less frequently found (De la Cabada and Dupont, 2011). ETEC's diarrheal incidence rate is estimated to be 280–400 million cases per year in children under 5 years of age (WHO, 2006). Furthermore there are currently no vaccines licensed against ETEC (Sizemore et al., 2004) though several vaccine candidates are in various phases of development (WHO, 2006). LT/ST-based antitoxin vaccine, fimbrial tip adhesin vaccine candidate, EtpA glycoprotein Abbreviations: ETEC, enterotoxigenic Escherichia coli; TD, traveler's diarrhea; LT/ST, heat labile/heat stable toxin; gyrA, gyraseA; gyrB, gyraseB; QRDR, quinolone resistance determining region; PCR, polymerase chain reaction; DS, discovery studio software; SDF, structure data file; DOPE, discrete optimized protein energy; WT, wild type; MT, mutant type; SIS, single interaction scan. ⁎ Corresponding author. E-mail address: [email protected] (J. Ramana).
vaccine candidate are in preclinical stage; inactivated ETEC vaccine candidate, vectored ETEC vaccine candidate (attenuated Salmonella typhi and attenuated Shigella) and transgenic plant vaccine candidates are in phase I; attenuated ETEC vaccine candidate is in its second phase of clinical trials. Dukoral®, the inactivated whole-cell cholera vaccine has shown short-term protection against ETEC but with limited success, and does not have significant impact, particularly in developing countries (PATH and bvgh, 2011). ETEC bacteria colonize the mucosal surface of the small intestine mediated by fimbrial adhesins followed by elaboration of diarrheagenic enterotoxin(s) (Kaper et al., 2004). ETEC makes two toxins, heat-labile (LT) and heat-stable (ST), which cause intestinal epithelial cells to secrete excess fluid. Use of antimicrobials in ETEC diarrhea is problematic because clinical laboratories do not perform toxin detection tests; hence accurate diagnosis cannot be made rapidly between various E. coli pathotypes. Moreover various studies witness a continuous increase in resistance against frequent antimicrobials including trimethoprim–sulfamethoxazole ampicillin, chloramphenicol, and tetracycline in different parts of the world (Vila et al., 1999; Jiang et al., 2000; Isenbarger et al., 2002). Fluoroquinolones have been found to be effective in traveler's diarrhea (Taylor et al., 1991; Ericsson et al., 1987) but the growing body of evidence has reported varying degree of reduced susceptibility or resistance to these antibiotics in different pathogens like
http://dx.doi.org/10.1016/j.gene.2015.09.063 0378-1119/© 2015 Elsevier B.V. All rights reserved.
Please cite this article as: Mehla, K., Ramana, J., Structural signature of Ser83Leu and Asp87Asn mutations in DNA gyrase from enterotoxigenic Escherichia coli and impact on quinolone resistance, Gene (2015), http://dx.doi.org/10.1016/j.gene.2015.09.063
2
K. Mehla, J. Ramana / Gene xxx (2015) xxx–xxx
Fig. 1. 2D chemical structures of marketed quinolones nalidixic acid and ciprofloxacin.
Shigella and ETEC (Vila et al., 2000; Nguyen et al., 2005). Quinolone resistance in E. coli is mainly associated with mutations in DNA topoisomerase II (Vila et al., 1994). DNA topoisomerase II is a tetramer of A and B subunits encoded by gyrA and gyrB genes, which associate to form an A2B2 complex in the active enzyme (Champoux, 2001). GyrA has a unique feature that it introduces negative supercoils into DNA, thereby relieving the torsional stress arising from transcription and replication complexes that leave the DNA in an energetically activated state (Gellert, 1981). Quinolones, a family of entirely synthetic broad spectrum drugs (Andersson and MacGowan, 2003) bind to DNA-gyrase complex forming a quinolonegyrase-DNA complex. Gyrase enzyme cleaves the DNA and quinolone
precludes the broken DNA strands from re-ligation (Hawkey, 2003). Nalidixic acid, the first of its class has a narrow spectrum while ciprofloxacin is a more potent and evolved member of the quinolone class of drug (Fàbrega et al., 2009). Quinolones unlike many other antibiotics are not subjected to transferrable plasmid mediated resistance; rather resistance is due to chromosomal mutations. The mutations generally fall in a small region between amino acids 67 and 106 of subunit A, termed as QRDR (Quinolone Resistance Determining Region) (Reece and Maxwell, 1991; Yoshida et al., 1990). Mutations are the major force behind evolution which marks significant phenotypic changes. Mutational resistance has special clinical relevance when considering resistance to quinolones (Woodford and Ellington, 2007). Mutations
Fig. 2. Sequence alignments for ETEC gyrA and template structure of gyrA from E. coli showing 91.2% identity.
Please cite this article as: Mehla, K., Ramana, J., Structural signature of Ser83Leu and Asp87Asn mutations in DNA gyrase from enterotoxigenic Escherichia coli and impact on quinolone resistance, Gene (2015), http://dx.doi.org/10.1016/j.gene.2015.09.063
K. Mehla, J. Ramana / Gene xxx (2015) xxx–xxx
3
Fig. 3. The secondary structure elements assignment in the corresponding sequence of the modeled protein structure.
at amino acid residues Ser83 and Asp87 implicated in quinolone resistance in E. coli have been widely reported in the literature (Conrad et al., 1996; Sáenz et al., 2003; Chen et al., 2001; Mendez Arancibia et al., 2009; Vila et al., 2000; Pazhani et al., 2011). The present study was inspired from the experimental work of Mendez Arancibia et al. (2009a); Pazhani et al. (2011); Vila et al. (2000) who focused their study on clinical isolates of diarrhea from patients who had traveled to North Africa and India. Minimum inhibitory concentration (MIC) of the clinical isolates was determined for various
antimicrobial agents like ampicillin, chloramphenicol, nalidixic acid, tetracycline, trimethoprim/sulfamethoxazole, ciprofloxacin and amoxicillin/clavulanic acid. All the isolates were analyzed for mutations in the QRDR region of gyrA and/or parC genes using PCR and DNA sequencing techniques. It was found that the strains which developed intermediate to high level of resistance harbored mutations in the gyrA and parC genes. In the work presented here, in silico docking studies have been exploited with a goal to provide a direct link between deleterious mutations in gyrA of ETEC and resistance to quinolones at structural level.
Fig. 4. Superimposition of ETEC gyrA and the template (PDB ID: 1AB4) produced by structural modeling with color coding scheme for helix, sheet and loop as red, yellow, green in template structure and cyan, magenta, salman in the modeled structure. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
Please cite this article as: Mehla, K., Ramana, J., Structural signature of Ser83Leu and Asp87Asn mutations in DNA gyrase from enterotoxigenic Escherichia coli and impact on quinolone resistance, Gene (2015), http://dx.doi.org/10.1016/j.gene.2015.09.063
4
K. Mehla, J. Ramana / Gene xxx (2015) xxx–xxx
Fig. 5. Highest scoring pose of ciprofloxacin at the active site of ETEC WT gyrA. Interacting residues and atoms are labeled. Receptor is shown as cartoons. Ligand is shown in stick mode, gyrA interacting residues are shown in ball and stick mode. Binding interactions are shown as dashed lines indicating salt bridge and hydrogen bonding interactions.
score was selected. DOPE score is a statistical measure for assessment of the quality of homology modeled structure as a whole. The residue profiles of the model were further verified for stereo-chemical properties by subjecting it to Verify3D (Eisenberg et al., 1997) and PROCHECK (Laskowski et al., 1993). The modeled native and mutant structures were prepared and energy minimized in Discovery Studio. Four mutant type structures (MT) were obtained from wild type (WT) by introducing amino acid substitutions at position 83 and 87, generating three mutants with single mutation Ser83Leu, Ser83Ala, Asp87Glu and a double mutant with Ser83Leu/ Asp87Asn. The receptor structures were prepared by optimizing the hydrogen network, inserting missing loops based on SEQRES data in PDB, optimizing short and medium loop regions with a looper algorithm and minimizing remaining loop regions, assigning appropriate ionization states at physiological pH 7.4. All structures were then energy minimized to RMS gradient 0.1 with 200 steps of minimization, so as to remove steric clashes, by a Smart Minimizer Algorithm with CHARMM force field. The secondary structure of the modeled gyrA was predicted using STRIDE (Frishman and Argos, 1995). The modeled structure was superimposed on the template crystal structure without changing the coordinate system of atoms in the template using a jCE algorithm in the Combinatorial Extension method (Shindyalov and Bourne, 1998).
2. Material and methods
2.3. Molecular docking
All the docking calculations were carried out in Accelrys Discovery Studio (DS) version 4.1 and LeadIt version 2.1.6 (Schneider et al., 2012). DS Client 4.1 was used for docking preparation; LeadIt was used for binding energy calculations. LeadIt uses the FlexX algorithm which is a fast, flexible docking method that uses an incremental construction algorithm for ligand placement into an active site.
Amino acids at position 83 and 87 in the receptor were selected to define a sphere center and amino acids within 10 Å radius of this sphere and specify the binding site to allow the ligand to rotate freely even in its fully extended conformation. The minimized structures of ligands and receptors were used for docking in LeadIt using the FlexX incremental algorithm which decomposes ligand into fragments and then selects some base fragments and uses them as anchors for systematic conformational analysis of remainder of the ligand using a variety of placement strategies. For present work, we employed an entropy approach which utilizes single interaction scan (SIS) based on hydrophobic pockets with only few interaction sites. For protein ligand clashes and intra ligand clashes, default values of maximum allowed overlap volume (2.9 cubic Å) and a clash factor (0.6) was used. For each iteration and fragmentation, 200 solutions were generated. Top 10 best poses for the ligand were generated for each docking calculation and conformations with highest docking score (more negative) and maximum number of interactions were used for further analysis. The output score is based on Bohm's function which involves a number of parameters (Böhm, 1994) and is calculated as:
2.1. Ligand preparation Ciprofloxacin (Pubchem Compund Id: 2764) and nalidixic acid (Pubchem Compund Id: 4421) 2D structures (SDF format) as shown in Fig. 1 were retrieved from Pubchem Compound database (Bolton et al., 2008). The structures were prepared by assigning appropriate ionization in physiological pH range 6.5–8.5 because at pH greater than 6.09, the carboxylic acid group will be primarily dissociated and at pH less than 8.74, the nitrogen will be primarily protonated. A conformational search for prepared ligands was performed in order to find a set of low-energy conformers. For each structure, a maximum of 10 tautomers and stereoisomers were generated. The prepared structures were energy minimized by Smart Minimizer Algorithm using CHARMM force field. 2.2. Homology modeling GyrA amino acid sequence (875aa) for ETEC strain E24377A (E. coli O78:H11) was retrieved from Uniprot database with Id A7ZP49. ETEC gyrA structure was modeled in Discovery Studio using E. coli gyrA (PDB Id: 1AB4) as template which showed 91.2% identity. Initial homology models were built using DS. Out of the 5 models generated by DS, the one with the lowest Discrete Optimized Protein Energy (DOPE)
X ΔGbind ¼ ΔG0 þ ΔGhb h−bonds f ðΔR; ΔαÞ X þ ΔGionic ionic interaction f ðΔR; ΔαÞ þ ΔGlipo Alipo þ ΔGrot NROT where ΔG values on the right side of equation are all constants, ΔG0 is contribution to the binding energy that does not depend directly on any specific interaction with the protein, hydrogen bonding and ionic terms depending on the geometry of interaction, with deviations from ideal distance R and ideal angle α being penalized, lipophilic term
Table 1 Docking scores and hydrogen bond residues involved in nalidixic acid and ciprofloxacin binding with WT and MT gyrA using LeadIt. Mutation
WT Ser83Leu Ser83Leu/Asp87Asn Ser83Ala Asp87Glu
Nalidixic acid
Ciprofloxacin
Score
No. of hydrogen bonds (residues involved)
Score
No. of hydrogen bonds (residues involved)
−12.51 −11.91 −10.55 −8.34 −11.75
2 (Arg 91, Gln 94) 2 (His 45, Arg 91) 2 (Arg 91, Gln 94) 2 (Arg 91, Gln 94) 2 (Arg 91, Gln 94)
−25.50 −21.23 −18.90 −20.36 −23.17
3 (Lys 42, Arg 91, Asp 87) 3 (Lys 42, Arg 91, Asp 87) 3 (Lys 42, Arg 91, Asn 87) 4 (Arg 32, Lys 42, HIS 45, Arg 91) 4 (Lys 42, Arg 91, Glu 87)
Please cite this article as: Mehla, K., Ramana, J., Structural signature of Ser83Leu and Asp87Asn mutations in DNA gyrase from enterotoxigenic Escherichia coli and impact on quinolone resistance, Gene (2015), http://dx.doi.org/10.1016/j.gene.2015.09.063
K. Mehla, J. Ramana / Gene xxx (2015) xxx–xxx
| Alipo | proportional to the contact surface area between non-polar atoms of protein and ligand, entropy term is directly proportional to the number of rotatable bonds in ligand (NROT) and represents the penalty associated with freezing internal rotations of the ligand.
Table 2 Hydrogen bonds involved in gyrA-ciprofloxacin interaction. Mutations Wild type
3. Results and discussion: 3.1. Model building and structure validation As there was no crystal structure available for ETEC gyrA, it was determined using a homology modeling protocol. The structure was built using E. coli gyrA as template which showed 91.2% identity. The sequence alignment generated using a Praline program (Simossis and Heringa, 2005) is shown in Fig. 2. The modeled structure shows a typical Rossmann fold composed of six parallel beta strands linked to two pairs of alpha helices (Fig. 3). The model with lowest DOPE score was subjected to structure validation in terms of the stereo-chemical properties using PROCHECK and Verify3D. It was observed that 99.8% residues were in the allowed regions with 67.3% residues in the most favored regions (Online Resource 1). Verify3D also predicted it to be a high quality model with 99.87% of the residues with score ≥0.2. The structural superimposition of the backbone of the modeled structure with the template (Fig. 4) resulted in a root mean square deviation (RMSD) of 1.3 Å with Zscore 51, indicating that the structure is a reasonable good quality model. High sequence identity of the modeled structure with the template ensures correct connectivity of the secondary structure elements. 3.2. Molecular docking and binding affinity calculations We used LeadIt for binding energy calculations of nalidixic acid and ciprofloxacin. FlexX incremental algorithm is aimed at assisting correct orientation of the ligands into the active site by using the SIS approach.
5
Receptor residue
Lys 42 Arg 91 Arg 91 Asp 87 Ser83Leu Lys 42 Arg 91 Asp 87 Ser83Leu/Asp87Asn Lys 42 Lys 42 Arg 91 Asn 87 Ser83Ala Arg 32 Lys 42 His 45 Arg 91 Asp87Glu Lys 42 Lys 42 Glu 87 Arg 91
Receptor atom
Ligand atom
Bond lengths (Å)
NZ NH1 NH1 OD2 NZ NH1 OD2 NZ NZ NH1 OD1 O HZ3 HE2 HH12 HZ3 HZ3 OE1 HH12
O3 O4 O2 H43 O3 O2 H43 O3 O4 O2 H42 H42 O2 O4 O3 O3 O4 H42 O2
3.04 2.79 3.01 2.76 2.74 2.75 2.80 2.96 2.73 2.75 3.25 1.98 2.15 1.98 2.06 2.13 2.10 1.73 1.67
The resulting poses were ranked based on docking score. Ciprofloxacin interacting with WT gyrA through hydrogen bonds with residues Lys42, Asp87, and Arg91 predicted a docking score of − 25.50. Fig. 5 captures the binding of ciprofloxacin with WT gyrA. The abilities to make hydrogen bonding and hydrophobic contacts are two crucial factors that decide whether ligand fits appropriately into the binding site. After introducing mutations in the WT gyrA, docking score decreased reasonably (Table 1). In MT Ser83Leu and double mutant Ser83Leu/ Asp87Asn docking score was observed to be −21.2327 and −18.8967 involving hydrogen bonds with Lys42, Asp87, and Arg91 (Fig. 6). In
Fig. 6. Interactions involved in ciprofloxacin binding with A.1) wild type A.2) mutated S83L and A.3) S83L/D87N QRDR of gyrA. Similarly interactions involved in QRDR complex with nalidixic acid in B.1) wild type B.2) mutated S83L and B.3) S83L/D87N. Images are generated by a PoseView program in LeadIt.
Please cite this article as: Mehla, K., Ramana, J., Structural signature of Ser83Leu and Asp87Asn mutations in DNA gyrase from enterotoxigenic Escherichia coli and impact on quinolone resistance, Gene (2015), http://dx.doi.org/10.1016/j.gene.2015.09.063
6
K. Mehla, J. Ramana / Gene xxx (2015) xxx–xxx
Table 3 Contribution of various energy terms in ciprofloxacin and nalidixic acid binding with WT and MT forms of gyrA. Mutations
Match score
Lipo score
Ambig score
Clash score
Rot score
Ciprofloxacin Wild type Ser83Leu Ser83Leu/Asp87Asn Ser83Ala Asp87Glu
−23.37 −21.88 −16.82 −18.90 −21.03
−5.61 −5.34 −6.47 −7.71 −5.69
−5.78 −4.04 −5.08 −5.00 −4.90
2.45 3.24 2.67 4.45 1.66
1.4 1.4 1.4 1.4 1.4
Nalidixic acid Wild type Ser83Leu Ser83Leu/Asp87Asn Ser83Ala Asp87Glu
−12.69 −13.48 −11.95 −17.09 −18.61
−4.50 −3.03 −3.90 −2.15 −2.65
−3.24 −3.23 −3.34 −1.61 −1.28
1.1 1.02 1.85 1.23 2.19
1.4 1.4 1.4 1.4 1.4
WT gyrA, Arg91 bonded with ciprofloxacin atoms O2 and O4 while in the MT gyrA hydrogen bond with ciprofloxacin O4 was lost (Table 2), thereby contributing to a decrease in the docking score. Asp87 being directly hydrogen bonded to the ciprofloxacin H43 atom, plays a crucial role in the quinolone binding with gyrA. In the double mutant hydrogen bond length for residue 87 was increased from 2.76 Å in WT gyrA to 3.25 Å in double mutant, significantly decreasing the bond energy. In MT Ser83Ala and Asp87Glu docking score against ciprofloxacin was observed to be − 20.17 and − 23.17 involving ionic interactions with Arg 32, Lys 42, His 45, Arg 91 and Lys 42, Glu 87, Arg 91 respectively. Loss of interaction with Asp87 which is considered as hotspot amino acid in gyrA and fluoroquinolone interaction is assumed to be major cause of reduced binding affinity for Ser83Ala. Ser83Leu despite having bulkier side chain has relatively better binding score owing to the fact
that interaction with Asp 87 was conserved. As Glu and Asp both have negatively charged side chains, replacement of Asp with Glu at 87 has a less pronounced effect on ciprofloxacin binding as compared to other mutant forms. Other than hydrogen bonds, hydrophobic interactions also contribute to the docking score. Fluorine atom of ciprofloxacin made close contacts with Asp87. In WT gyrA complexed with ciprofloxacin, close atom–atom contact score (clash) was 2.45 lower as compared to Ser83Leu and Ser83Leu/Asp87Asn in which it was observed to be 3.24 and 2.67 (Table 3). In Ser83Leu clash score was highest because leucine being a bulkier residue, causes steric hindrance due to its side chain. In Ser83Leu lipophilic atom–atom contact score was also lowest (−5.34) which was − 5.61 and − 6.47 in WT gyrA and double mutant respectively. The lipophilic term is proportional to the contact surface area between the ligand and the protein involving non-polar terms and it contributes positively to the binding affinity. As a result binding affinity is reduced with a decrease in the lipophilic contacts. Hydrogen bonded interactions of ciprofloxacin with both the mutants are shown in Online Resource 1. We can discern from these findings that substitution of Ser with Leu at position 83 and Asp with Asn at position 87 in gyrA has a pronounced effect on ciprofloxacin binding and have been reported frequently. It is evident that mutation of both Ser83 and Asp 87 resulted in a relatively low docking score suggesting the critical role of these residues in ciprofloxacin binding with gyrA. This decreasing trend in docking score accounts for increased resistance in the MT gyrA, observed to be highest in the double mutant. Nalidixic acid interacts with WT gyrA through hydrogen bonds with residues Arg91 and Gln94 and hydrophobic interaction with neighboring residues Asp87, Phe96 and Ser111 with a high docking score of −12.51 (Table 1). Substitution of Ser with Leu at 83 resulted in loss of hydrogen bond with Gln94 residue and formation of new hydrogen
Fig. 7. Hydrogen bonds involved in binding of ciprofloxacin with A.1) wild type A.2) mutated S83L and A.3) S83L/D87N QRDR of gyrA. Similarly hydrogen bonds involved in QRDR complex with nalidixic acid in B.1) wild type B.2) mutated S83L and B.3) S83L/D87N.
Please cite this article as: Mehla, K., Ramana, J., Structural signature of Ser83Leu and Asp87Asn mutations in DNA gyrase from enterotoxigenic Escherichia coli and impact on quinolone resistance, Gene (2015), http://dx.doi.org/10.1016/j.gene.2015.09.063
K. Mehla, J. Ramana / Gene xxx (2015) xxx–xxx
bond with His45 and Arg91. As shown in Fig. 2, Arg91 shared weaker interactions with nalidixic acid atoms O1 and O3, with bond lengths being 3.19 Å and 2.90 Å respectively. In the double mutant, interactions were with Arg91 and Gln94 as in the WT gyrA (Fig. 7), but bond lengths increased as shown in Table 4, significantly reducing the docking score. Loss of one hydrogen bond with Gln94 in Ser83Leu mutant suggests a critical role of this residue in nalidixic acid binding with gyrA. Hydrogen bonded interactions of nalidixic acid with both the WT and MT are shown in Online Resource 1. In WT gyrA, close atom–atom contact score (clash) was 1.1, lower as compared to Ser83Leu/Asp87Asn in which it was observed to be 1.85. In Ser83Leu, clash score was almost similar to WT gyrA but the lipophilic contact score was much lower (− 3.03) as compared to WT gyrA (− 4.50). In Ser83Leu/Asp87Asn, lipophilic atom–atom contact score was − 3.90 (Table 3). Nalidixic acid interacts with MT Ser83Ala and Asp87Glu through weaker hydrogen bonds with residues Arg 91 and Gln 94 having relatively low docking score of − 8.35 and − 11.75 as compared to WT gyrA. MT Ser83Ala and Asp87Glu follow the similar trend for nalidixic acid as for ciprofloxacin. It is apparent that substitution of both Ser83 and Asp87 accounts for a decrease in the docking score and less efficient binding which ultimately leads to intermediate to high level drug resistance in the strains previously sensitive to drugs. The docking score is relatively high for ciprofloxacin because of the additional interactions with the fluorine atom, answering why resistance to the first generation nalidixic acid is higher as compared to the second generation drug ciprofloxacin. Our in silico docking results show good agreement with previous experimental studies on these mutations (Vila et al., 2000; Mendez Arancibia et al., 2009; Pazhani et al., 2011). Through molecular modeling studies we can derive a possible explanation for increased resistance in pathogens, but we cannot firmly establish the predictive powers of this approach since other mechanisms as well can contribute to resistance such as plasmid mediated resistance, or the presence of integrons and gene cassettes. Binding affinity is a function of the stability of ligand-target complex. Protein stability and interactions are important biological aspects that can be influenced by amino acid substitutions. The results presented in this paper provide an in silico based approach to seek insights into the effect of mutations at structural level and guide the pharmacologists to introduce necessary modifications in the existing drugs and paving a way for counteracting the problem of antibiotic resistance. 4. Conclusion A number of mutations localized in QRDR region of DNA gyrase have been reported that lead to quinolone resistance. Our computational molecular docking study shows that mutations at both Ser83 and Asp87 are primarily responsible for reduced affinity of these drugs towards gyrA and provides a possible explanation for observed quinolone resistance. This study provides interesting insights into molecular level mechanisms of antibiotic resistance due to mutations in quinolones, a major Table 4 Hydrogen bonds involved in gyrA-nalidixic acid interaction. Mutations Wild type
Receptor residue
Arg 91 Gln 94 Ser83Leu Arg 91 Arg 91 His 45 Ser83Leu/Asp87Asn Arg 91 Gln 94 Ser83Ala Arg 91 Gln 94 Asp87Glu Arg 91 Gln 94
Receptor atom
Ligand atom
Bond lengths (Å)
NH1 NE2 NH1 NH1 NE2 NH1 NE2 HH12 HE21 HH12 HE21
O2 N5 O1 O3 O2 O3 N5 O3 N5 O2 N5
2.71 2.74 3.19 2.90 2.68 2.88 3.12 1.82 2.17 2.09 1.85
7
class of antibiotics against diarrheagenic E. coli. Additionally, gaining knowledge about possible disease associations of mutations is another major research problem. The SNPs data is enormous, yet it is still not clear what biological effect do these mutations have. By studying the reported mutations at structural level, these studies open up new avenues for designing novel inhibitors against DNA gyrase and can be valuable in guiding the necessary modifications in the existing drugs to counteract drug resistance. Compliance with ethical standards Competing interests We confirm that there are no conflicts of interest associated with this publication. Acknowledgments This research was supported by DST (Department of Science and Technology), Ministry of Science and Technology, India under grant number SB/FT/LS-278/2012. The authors would like to thank the anonymous reviewers for their insightful comments that helped improve the content of the manuscript. Appendix A. Supplementary data Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.gene.2015.09.063. References Andersson, M.I., MacGowan, A.P., 2003. Development of the quinolones. J. Antimicrob. Chemother. 51 (Suppl 1), 1–11. http://dx.doi.org/10.1093/jac/dkg212. Böhm, H.J., 1994. The development of a simple empirical scoring function to estimate the binding constant for a protein-ligand complex of known three-dimensional structure. J. Comput. Aided Mol. Des. 8, 243–256. Bolton, E.E., Wang, Y., Thiessen, P.A., Bryant, S.H., 2008. Chapter 12 — PubChem: integrated platform of small molecules and biological activities. Annu. Rep. Comput. Chem. 217–241. Champoux, J.J., 2001. DNA topoisomerases: structure, function, and mechanism. Annu. Rev. Biochem. 70, 369–413. http://dx.doi.org/10.1146/annurev.biochem.70.1.369. Chen, J.Y., Siu, L.K., Chen, Y.H., et al., 2001. Molecular epidemiology and mutations at gyrA and parC genes of ciprofloxacin-resistant Escherichia coli isolates from a Taiwan medical center. Microb. Drug Resist. 7, 47–53. http://dx.doi.org/10.1089/ 107662901750152783. Conrad, S., Oethinger, M., Kaifel, K., et al., 1996. GyrA mutations in high-level fluoroquinolone-resistant clinical isolates of Escherichia coli. J. Antimicrob. Chemother. 38, 443–455. De la Cabada, B.J., Dupont, H.L., 2011. New developments in traveler's diarrhea. Gastroenterol. Hepatol. (N Y) 7, 88–95. Eisenberg, D., Lüthy, R., Bowie, J.U., 1997. VERIFY3D: assessment of protein models with three-dimensional profiles. Methods Enzymol. 277, 396–404. Ericsson, C.D., Johnson, P.C., Dupont, H.L., et al., 1987. Ciprofloxacin or trimethoprim–sulfamethoxazole as initial therapy for travelers' diarrhea. A placebo-controlled, randomized trial. Ann. Intern. Med. 106, 216–220. Fàbrega, A., Madurga, S., Giralt, E., Vila, J., 2009. Mechanism of action of and resistance to quinolones. Microb. Biotechnol. 2, 40–61. http://dx.doi.org/10.1111/j.1751-7915. 2008.00063.x. Frishman, D., Argos, P., 1995. Knowledge-based protein secondary structure assignment. Proteins 23, 566–579. http://dx.doi.org/10.1002/prot.340230412. Gellert, M., 1981. DNA topoisomerases. Annu. Rev. Biochem. 50, 879–910. http://dx.doi. org/10.1146/annurev.bi.50.070181.004311. Hawkey, P.M., 2003. Mechanisms of quinolone action and microbial response. J. Antimicrob. Chemother. 51 (Suppl 1), 29–35. http://dx.doi.org/10.1093/jac/dkg207. Isenbarger, D.W., Hoge, C.W., Srijan, A., et al., 2002. Comparative antibiotic resistance of diarrheal pathogens from Vietnam and Thailand, 1996–1999. Emerg. Infect. Dis. 8, 175–180. http://dx.doi.org/10.3201/eid0802.010145. Jiang, Z.D., Mathewson, J.J., Ericsson, C.D., et al., 2000. Characterization of enterotoxigenic Escherichia coli strains in patients with travelers' diarrhea acquired in Guadalajara, Mexico, 1992–1997. J. Infect. Dis. 181, 779–782. http://dx.doi.org/10.1086/315272. Kaper, J.B., Nataro, J.P., Mobley, H.L., 2004. Pathogenic Escherichia coli. Nat. Rev. Microbiol. 2, 123–140. http://dx.doi.org/10.1038/nrmicro818. Laskowski, R.A., MacArthur, M.W., Moss, D.S., Thornton, J.M., 1993. PROCHECK: a program to check the stereochemical quality of protein structures. J. Appl. Crystallogr. 26, 283–291. http://dx.doi.org/10.1107/S0021889892009944. Mendez Arancibia, E., Pitart, C., Ruiz, J., et al., 2009. Evolution of antimicrobial resistance in enteroaggregative Escherichia coli and enterotoxigenic Escherichia coli causing
Please cite this article as: Mehla, K., Ramana, J., Structural signature of Ser83Leu and Asp87Asn mutations in DNA gyrase from enterotoxigenic Escherichia coli and impact on quinolone resistance, Gene (2015), http://dx.doi.org/10.1016/j.gene.2015.09.063
8
K. Mehla, J. Ramana / Gene xxx (2015) xxx–xxx
traveller's diarrhoea. J. Antimicrob. Chemothe.r 64, 343–347. http://dx.doi.org/10. 1093/jac/dkp178. Nguyen, T.V., Le, P.V., Le, C.H., Weintraub, A., 2005. Antibiotic resistance in diarrheagenic Escherichia coli and Shigella strains isolated from children in Hanoi, Vietnam. Antimicrob. Agents Chemother. 49, 816–819. http://dx.doi.org/10.1128/AAC.49.2. 816-819.2005. PATH, bvgh, 2011. The Case for Investment in Enterotoxigenic Escherichia coli Vaccines. Pazhani, G.P., Chakraborty, S., Fujihara, K., et al., 2011. QRDR mutations, efflux system & antimicrobial resistance genes in enterotoxigenic Escherichia coli isolated from an outbreak of diarrhoea in Ahmedabad, India. Indian J. Med. Res. 134, 214–223. Reece, R.J., Maxwell, A., 1991. DNA gyrase: structure and function. Crit. Rev. Biochem. Mol. Biol. 26, 335–375. http://dx.doi.org/10.3109/10409239109114072. Sáenz, Y., Zarazaga, M., Briñas, L., et al., 2003. Mutations in gyrA and parC genes in nalidixic acid-resistant Escherichia coli strains from food products, humans and animals. J. Antimicrob. Chemother. 51, 1001–1005. http://dx.doi.org/10.1093/jac/ dkg168. Schneider, N., Hindle, S., Lange, G., et al., 2012. Substantial improvements in large-scale redocking and screening using the novel HYDE scoring function. J. Comput. Aided Mol. Des. 26, 701–723. http://dx.doi.org/10.1007/s10822-011-9531-0. Shindyalov, I.N., Bourne, P.E., 1998. Protein structure alignment by incremental combinatorial extension (CE) of the optimal path. Protein Eng. 11, 739–747. Simossis, V.A., Heringa, J., 2005. PRALINE: a multiple sequence alignment toolbox that integrates homology-extended and secondary structure information. Nucleic Acids Res. 33, W289–W294. http://dx.doi.org/10.1093/nar/gki390.
Sizemore, D.R., Roland, K.L., Ryan, U.S., 2004. Enterotoxigenic Escherichia coli virulence factors and vaccine approaches. Expert Rev. Vaccines 3, 585–595. http://dx.doi.org/ 10.1586/14760584.3.5.585. Taylor, D.N., Sanchez, J.L., Candler, W., et al., 1991. Treatment of travelers' diarrhea: ciprofloxacin plus loperamide compared with ciprofloxacin alone. A placebo-controlled, randomized trial. Ann. Intern. Med. 114, 731–734. Vila, J., Ruiz, J., Marco, F., et al., 1994. Association between double mutation in gyrA gene of ciprofloxacin-resistant clinical isolates of Escherichia coli and MICs. Antimicrob. Agents Chemother. 38, 2477–2479. Vila, J., Vargas, M., Casals, C., et al., 1999. Antimicrobial resistance of diarrheagenic Escherichia coli isolated from children under the age of 5 years from Ifakara, Tanzania. Antimicrob. Agents Chemother. 43, 3022–3024. Vila, J., Vargas, M., Ruiz, J., et al., 2000. Quinolone resistance in enterotoxigenic Escherichia coli causing diarrhea in travelers to India in comparison with other geographical areas. Antimicrob. Agents Chemother. 44, 1731–1733. WHO, 2006. Future directions for research on enterotoxigenic Escherichia coli vaccines for developing countries. Wkly Epidemiol. Rec. 81, 97–104. Woodford, N., Ellington, M.J., 2007. The emergence of antibiotic resistance by mutation. Clin. Microbiol. Infect. 13, 5–18. http://dx.doi.org/10.1111/j.1469-0691.2006.01492.x. Yoshida, H., Bogaki, M., Nakamura, M., Nakamura, S., 1990. Quinolone resistance-determining region in the DNA gyrase gyrA gene of Escherichia coli. Antimicrob. Agents Chemother. 34, 1271–1272.
Please cite this article as: Mehla, K., Ramana, J., Structural signature of Ser83Leu and Asp87Asn mutations in DNA gyrase from enterotoxigenic Escherichia coli and impact on quinolone resistance, Gene (2015), http://dx.doi.org/10.1016/j.gene.2015.09.063
Contents lists available at ScienceDirect
Gene journal homepage: www.elsevier.com/locate/gene
Research paper
Structural signature of Ser83Leu and Asp87Asn mutations in DNA gyrase from enterotoxigenic Escherichia coli and impact on quinolone resistance Kusum Mehla, Jayashree Ramana ⁎ Department of Biotechnology and Bioinformatics, Jaypee University of Information Technology, Solan PIN-173234, Himachal Pradesh, India
a r t i c l e
i n f o
Article history: Received 10 March 2015 Received in revised form 22 September 2015 Accepted 23 September 2015 Available online xxxx Keywords: Diarrhea Drug resistance Amino acid substitutions Homology modeling Binding affinity
a b s t r a c t Enterotoxigenic Escherichia coli (ETEC) is among the most frequent microorganisms causing traveler's diarrhea (TD). Quinolones are potent antimicrobial agents used for the treatment of TD. Resistance to quinolones is typically caused by substitutions in QRDR region of gyrA subunit of DNA gyrase. The aim of this study was to seek insights into the effect of these substitutions at structural level and their association with observed quinolone resistance. Majority of the ETEC strains have gyrA mutations at amino acid position 83 and 87. To understand the quinolone resistance mechanism at molecular level, we have studied the interaction of wild type and mutant forms of ETEC gyrA with nalidixic acid and ciprofloxacin by molecular modeling using Discovery Studio and LeadIt. All the mutants had reduced affinity towards both ciprofloxacin and nalidixic acid relative to the wild type due to the mutations introduced in gyrA. Besides Ser83 and Asp87, for nalidixic acid binding Arg91 and His45 residues were observed to be critical while in ciprofloxacin binding Lys42 and Arg91 residues played a significant role. Amino acid substitutions contribute to the emergence of drug resistance in sensitive strains by causing structural alterations leading to reduced affinity of the drug towards receptor. Analysis of the effect of amino acid substitutions at structural level is of utmost importance to establish possible associations between mutations and the diseases. These studies accelerate the identification of pharmaceutical targets for relevant treatments and could also be helpful in guiding the design of further experimental research. © 2015 Elsevier B.V. All rights reserved.
1. Introduction Progressive increase in antimicrobial resistance among enteric pathogens has become a critical area of public concern. Enterotoxigenic Escherichia coli (ETEC) is the leading bacterial cause of diarrhea in developing world where inadequate clean water and poor sanitation facilities are prevailing and a predominant cause of TD (Sizemore et al., 2004). Enteroaggregative E. coli, Shigella spp. and Salmonella spp. are other common bacterial agents, while Campylobacter, Yersinia, Aeromonas, and Plesiomonas spp. cases are less frequently found (De la Cabada and Dupont, 2011). ETEC's diarrheal incidence rate is estimated to be 280–400 million cases per year in children under 5 years of age (WHO, 2006). Furthermore there are currently no vaccines licensed against ETEC (Sizemore et al., 2004) though several vaccine candidates are in various phases of development (WHO, 2006). LT/ST-based antitoxin vaccine, fimbrial tip adhesin vaccine candidate, EtpA glycoprotein Abbreviations: ETEC, enterotoxigenic Escherichia coli; TD, traveler's diarrhea; LT/ST, heat labile/heat stable toxin; gyrA, gyraseA; gyrB, gyraseB; QRDR, quinolone resistance determining region; PCR, polymerase chain reaction; DS, discovery studio software; SDF, structure data file; DOPE, discrete optimized protein energy; WT, wild type; MT, mutant type; SIS, single interaction scan. ⁎ Corresponding author. E-mail address: [email protected] (J. Ramana).
vaccine candidate are in preclinical stage; inactivated ETEC vaccine candidate, vectored ETEC vaccine candidate (attenuated Salmonella typhi and attenuated Shigella) and transgenic plant vaccine candidates are in phase I; attenuated ETEC vaccine candidate is in its second phase of clinical trials. Dukoral®, the inactivated whole-cell cholera vaccine has shown short-term protection against ETEC but with limited success, and does not have significant impact, particularly in developing countries (PATH and bvgh, 2011). ETEC bacteria colonize the mucosal surface of the small intestine mediated by fimbrial adhesins followed by elaboration of diarrheagenic enterotoxin(s) (Kaper et al., 2004). ETEC makes two toxins, heat-labile (LT) and heat-stable (ST), which cause intestinal epithelial cells to secrete excess fluid. Use of antimicrobials in ETEC diarrhea is problematic because clinical laboratories do not perform toxin detection tests; hence accurate diagnosis cannot be made rapidly between various E. coli pathotypes. Moreover various studies witness a continuous increase in resistance against frequent antimicrobials including trimethoprim–sulfamethoxazole ampicillin, chloramphenicol, and tetracycline in different parts of the world (Vila et al., 1999; Jiang et al., 2000; Isenbarger et al., 2002). Fluoroquinolones have been found to be effective in traveler's diarrhea (Taylor et al., 1991; Ericsson et al., 1987) but the growing body of evidence has reported varying degree of reduced susceptibility or resistance to these antibiotics in different pathogens like
http://dx.doi.org/10.1016/j.gene.2015.09.063 0378-1119/© 2015 Elsevier B.V. All rights reserved.
Please cite this article as: Mehla, K., Ramana, J., Structural signature of Ser83Leu and Asp87Asn mutations in DNA gyrase from enterotoxigenic Escherichia coli and impact on quinolone resistance, Gene (2015), http://dx.doi.org/10.1016/j.gene.2015.09.063
2
K. Mehla, J. Ramana / Gene xxx (2015) xxx–xxx
Fig. 1. 2D chemical structures of marketed quinolones nalidixic acid and ciprofloxacin.
Shigella and ETEC (Vila et al., 2000; Nguyen et al., 2005). Quinolone resistance in E. coli is mainly associated with mutations in DNA topoisomerase II (Vila et al., 1994). DNA topoisomerase II is a tetramer of A and B subunits encoded by gyrA and gyrB genes, which associate to form an A2B2 complex in the active enzyme (Champoux, 2001). GyrA has a unique feature that it introduces negative supercoils into DNA, thereby relieving the torsional stress arising from transcription and replication complexes that leave the DNA in an energetically activated state (Gellert, 1981). Quinolones, a family of entirely synthetic broad spectrum drugs (Andersson and MacGowan, 2003) bind to DNA-gyrase complex forming a quinolonegyrase-DNA complex. Gyrase enzyme cleaves the DNA and quinolone
precludes the broken DNA strands from re-ligation (Hawkey, 2003). Nalidixic acid, the first of its class has a narrow spectrum while ciprofloxacin is a more potent and evolved member of the quinolone class of drug (Fàbrega et al., 2009). Quinolones unlike many other antibiotics are not subjected to transferrable plasmid mediated resistance; rather resistance is due to chromosomal mutations. The mutations generally fall in a small region between amino acids 67 and 106 of subunit A, termed as QRDR (Quinolone Resistance Determining Region) (Reece and Maxwell, 1991; Yoshida et al., 1990). Mutations are the major force behind evolution which marks significant phenotypic changes. Mutational resistance has special clinical relevance when considering resistance to quinolones (Woodford and Ellington, 2007). Mutations
Fig. 2. Sequence alignments for ETEC gyrA and template structure of gyrA from E. coli showing 91.2% identity.
Please cite this article as: Mehla, K., Ramana, J., Structural signature of Ser83Leu and Asp87Asn mutations in DNA gyrase from enterotoxigenic Escherichia coli and impact on quinolone resistance, Gene (2015), http://dx.doi.org/10.1016/j.gene.2015.09.063
K. Mehla, J. Ramana / Gene xxx (2015) xxx–xxx
3
Fig. 3. The secondary structure elements assignment in the corresponding sequence of the modeled protein structure.
at amino acid residues Ser83 and Asp87 implicated in quinolone resistance in E. coli have been widely reported in the literature (Conrad et al., 1996; Sáenz et al., 2003; Chen et al., 2001; Mendez Arancibia et al., 2009; Vila et al., 2000; Pazhani et al., 2011). The present study was inspired from the experimental work of Mendez Arancibia et al. (2009a); Pazhani et al. (2011); Vila et al. (2000) who focused their study on clinical isolates of diarrhea from patients who had traveled to North Africa and India. Minimum inhibitory concentration (MIC) of the clinical isolates was determined for various
antimicrobial agents like ampicillin, chloramphenicol, nalidixic acid, tetracycline, trimethoprim/sulfamethoxazole, ciprofloxacin and amoxicillin/clavulanic acid. All the isolates were analyzed for mutations in the QRDR region of gyrA and/or parC genes using PCR and DNA sequencing techniques. It was found that the strains which developed intermediate to high level of resistance harbored mutations in the gyrA and parC genes. In the work presented here, in silico docking studies have been exploited with a goal to provide a direct link between deleterious mutations in gyrA of ETEC and resistance to quinolones at structural level.
Fig. 4. Superimposition of ETEC gyrA and the template (PDB ID: 1AB4) produced by structural modeling with color coding scheme for helix, sheet and loop as red, yellow, green in template structure and cyan, magenta, salman in the modeled structure. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
Please cite this article as: Mehla, K., Ramana, J., Structural signature of Ser83Leu and Asp87Asn mutations in DNA gyrase from enterotoxigenic Escherichia coli and impact on quinolone resistance, Gene (2015), http://dx.doi.org/10.1016/j.gene.2015.09.063
4
K. Mehla, J. Ramana / Gene xxx (2015) xxx–xxx
Fig. 5. Highest scoring pose of ciprofloxacin at the active site of ETEC WT gyrA. Interacting residues and atoms are labeled. Receptor is shown as cartoons. Ligand is shown in stick mode, gyrA interacting residues are shown in ball and stick mode. Binding interactions are shown as dashed lines indicating salt bridge and hydrogen bonding interactions.
score was selected. DOPE score is a statistical measure for assessment of the quality of homology modeled structure as a whole. The residue profiles of the model were further verified for stereo-chemical properties by subjecting it to Verify3D (Eisenberg et al., 1997) and PROCHECK (Laskowski et al., 1993). The modeled native and mutant structures were prepared and energy minimized in Discovery Studio. Four mutant type structures (MT) were obtained from wild type (WT) by introducing amino acid substitutions at position 83 and 87, generating three mutants with single mutation Ser83Leu, Ser83Ala, Asp87Glu and a double mutant with Ser83Leu/ Asp87Asn. The receptor structures were prepared by optimizing the hydrogen network, inserting missing loops based on SEQRES data in PDB, optimizing short and medium loop regions with a looper algorithm and minimizing remaining loop regions, assigning appropriate ionization states at physiological pH 7.4. All structures were then energy minimized to RMS gradient 0.1 with 200 steps of minimization, so as to remove steric clashes, by a Smart Minimizer Algorithm with CHARMM force field. The secondary structure of the modeled gyrA was predicted using STRIDE (Frishman and Argos, 1995). The modeled structure was superimposed on the template crystal structure without changing the coordinate system of atoms in the template using a jCE algorithm in the Combinatorial Extension method (Shindyalov and Bourne, 1998).
2. Material and methods
2.3. Molecular docking
All the docking calculations were carried out in Accelrys Discovery Studio (DS) version 4.1 and LeadIt version 2.1.6 (Schneider et al., 2012). DS Client 4.1 was used for docking preparation; LeadIt was used for binding energy calculations. LeadIt uses the FlexX algorithm which is a fast, flexible docking method that uses an incremental construction algorithm for ligand placement into an active site.
Amino acids at position 83 and 87 in the receptor were selected to define a sphere center and amino acids within 10 Å radius of this sphere and specify the binding site to allow the ligand to rotate freely even in its fully extended conformation. The minimized structures of ligands and receptors were used for docking in LeadIt using the FlexX incremental algorithm which decomposes ligand into fragments and then selects some base fragments and uses them as anchors for systematic conformational analysis of remainder of the ligand using a variety of placement strategies. For present work, we employed an entropy approach which utilizes single interaction scan (SIS) based on hydrophobic pockets with only few interaction sites. For protein ligand clashes and intra ligand clashes, default values of maximum allowed overlap volume (2.9 cubic Å) and a clash factor (0.6) was used. For each iteration and fragmentation, 200 solutions were generated. Top 10 best poses for the ligand were generated for each docking calculation and conformations with highest docking score (more negative) and maximum number of interactions were used for further analysis. The output score is based on Bohm's function which involves a number of parameters (Böhm, 1994) and is calculated as:
2.1. Ligand preparation Ciprofloxacin (Pubchem Compund Id: 2764) and nalidixic acid (Pubchem Compund Id: 4421) 2D structures (SDF format) as shown in Fig. 1 were retrieved from Pubchem Compound database (Bolton et al., 2008). The structures were prepared by assigning appropriate ionization in physiological pH range 6.5–8.5 because at pH greater than 6.09, the carboxylic acid group will be primarily dissociated and at pH less than 8.74, the nitrogen will be primarily protonated. A conformational search for prepared ligands was performed in order to find a set of low-energy conformers. For each structure, a maximum of 10 tautomers and stereoisomers were generated. The prepared structures were energy minimized by Smart Minimizer Algorithm using CHARMM force field. 2.2. Homology modeling GyrA amino acid sequence (875aa) for ETEC strain E24377A (E. coli O78:H11) was retrieved from Uniprot database with Id A7ZP49. ETEC gyrA structure was modeled in Discovery Studio using E. coli gyrA (PDB Id: 1AB4) as template which showed 91.2% identity. Initial homology models were built using DS. Out of the 5 models generated by DS, the one with the lowest Discrete Optimized Protein Energy (DOPE)
X ΔGbind ¼ ΔG0 þ ΔGhb h−bonds f ðΔR; ΔαÞ X þ ΔGionic ionic interaction f ðΔR; ΔαÞ þ ΔGlipo Alipo þ ΔGrot NROT where ΔG values on the right side of equation are all constants, ΔG0 is contribution to the binding energy that does not depend directly on any specific interaction with the protein, hydrogen bonding and ionic terms depending on the geometry of interaction, with deviations from ideal distance R and ideal angle α being penalized, lipophilic term
Table 1 Docking scores and hydrogen bond residues involved in nalidixic acid and ciprofloxacin binding with WT and MT gyrA using LeadIt. Mutation
WT Ser83Leu Ser83Leu/Asp87Asn Ser83Ala Asp87Glu
Nalidixic acid
Ciprofloxacin
Score
No. of hydrogen bonds (residues involved)
Score
No. of hydrogen bonds (residues involved)
−12.51 −11.91 −10.55 −8.34 −11.75
2 (Arg 91, Gln 94) 2 (His 45, Arg 91) 2 (Arg 91, Gln 94) 2 (Arg 91, Gln 94) 2 (Arg 91, Gln 94)
−25.50 −21.23 −18.90 −20.36 −23.17
3 (Lys 42, Arg 91, Asp 87) 3 (Lys 42, Arg 91, Asp 87) 3 (Lys 42, Arg 91, Asn 87) 4 (Arg 32, Lys 42, HIS 45, Arg 91) 4 (Lys 42, Arg 91, Glu 87)
Please cite this article as: Mehla, K., Ramana, J., Structural signature of Ser83Leu and Asp87Asn mutations in DNA gyrase from enterotoxigenic Escherichia coli and impact on quinolone resistance, Gene (2015), http://dx.doi.org/10.1016/j.gene.2015.09.063
K. Mehla, J. Ramana / Gene xxx (2015) xxx–xxx
| Alipo | proportional to the contact surface area between non-polar atoms of protein and ligand, entropy term is directly proportional to the number of rotatable bonds in ligand (NROT) and represents the penalty associated with freezing internal rotations of the ligand.
Table 2 Hydrogen bonds involved in gyrA-ciprofloxacin interaction. Mutations Wild type
3. Results and discussion: 3.1. Model building and structure validation As there was no crystal structure available for ETEC gyrA, it was determined using a homology modeling protocol. The structure was built using E. coli gyrA as template which showed 91.2% identity. The sequence alignment generated using a Praline program (Simossis and Heringa, 2005) is shown in Fig. 2. The modeled structure shows a typical Rossmann fold composed of six parallel beta strands linked to two pairs of alpha helices (Fig. 3). The model with lowest DOPE score was subjected to structure validation in terms of the stereo-chemical properties using PROCHECK and Verify3D. It was observed that 99.8% residues were in the allowed regions with 67.3% residues in the most favored regions (Online Resource 1). Verify3D also predicted it to be a high quality model with 99.87% of the residues with score ≥0.2. The structural superimposition of the backbone of the modeled structure with the template (Fig. 4) resulted in a root mean square deviation (RMSD) of 1.3 Å with Zscore 51, indicating that the structure is a reasonable good quality model. High sequence identity of the modeled structure with the template ensures correct connectivity of the secondary structure elements. 3.2. Molecular docking and binding affinity calculations We used LeadIt for binding energy calculations of nalidixic acid and ciprofloxacin. FlexX incremental algorithm is aimed at assisting correct orientation of the ligands into the active site by using the SIS approach.
5
Receptor residue
Lys 42 Arg 91 Arg 91 Asp 87 Ser83Leu Lys 42 Arg 91 Asp 87 Ser83Leu/Asp87Asn Lys 42 Lys 42 Arg 91 Asn 87 Ser83Ala Arg 32 Lys 42 His 45 Arg 91 Asp87Glu Lys 42 Lys 42 Glu 87 Arg 91
Receptor atom
Ligand atom
Bond lengths (Å)
NZ NH1 NH1 OD2 NZ NH1 OD2 NZ NZ NH1 OD1 O HZ3 HE2 HH12 HZ3 HZ3 OE1 HH12
O3 O4 O2 H43 O3 O2 H43 O3 O4 O2 H42 H42 O2 O4 O3 O3 O4 H42 O2
3.04 2.79 3.01 2.76 2.74 2.75 2.80 2.96 2.73 2.75 3.25 1.98 2.15 1.98 2.06 2.13 2.10 1.73 1.67
The resulting poses were ranked based on docking score. Ciprofloxacin interacting with WT gyrA through hydrogen bonds with residues Lys42, Asp87, and Arg91 predicted a docking score of − 25.50. Fig. 5 captures the binding of ciprofloxacin with WT gyrA. The abilities to make hydrogen bonding and hydrophobic contacts are two crucial factors that decide whether ligand fits appropriately into the binding site. After introducing mutations in the WT gyrA, docking score decreased reasonably (Table 1). In MT Ser83Leu and double mutant Ser83Leu/ Asp87Asn docking score was observed to be −21.2327 and −18.8967 involving hydrogen bonds with Lys42, Asp87, and Arg91 (Fig. 6). In
Fig. 6. Interactions involved in ciprofloxacin binding with A.1) wild type A.2) mutated S83L and A.3) S83L/D87N QRDR of gyrA. Similarly interactions involved in QRDR complex with nalidixic acid in B.1) wild type B.2) mutated S83L and B.3) S83L/D87N. Images are generated by a PoseView program in LeadIt.
Please cite this article as: Mehla, K., Ramana, J., Structural signature of Ser83Leu and Asp87Asn mutations in DNA gyrase from enterotoxigenic Escherichia coli and impact on quinolone resistance, Gene (2015), http://dx.doi.org/10.1016/j.gene.2015.09.063
6
K. Mehla, J. Ramana / Gene xxx (2015) xxx–xxx
Table 3 Contribution of various energy terms in ciprofloxacin and nalidixic acid binding with WT and MT forms of gyrA. Mutations
Match score
Lipo score
Ambig score
Clash score
Rot score
Ciprofloxacin Wild type Ser83Leu Ser83Leu/Asp87Asn Ser83Ala Asp87Glu
−23.37 −21.88 −16.82 −18.90 −21.03
−5.61 −5.34 −6.47 −7.71 −5.69
−5.78 −4.04 −5.08 −5.00 −4.90
2.45 3.24 2.67 4.45 1.66
1.4 1.4 1.4 1.4 1.4
Nalidixic acid Wild type Ser83Leu Ser83Leu/Asp87Asn Ser83Ala Asp87Glu
−12.69 −13.48 −11.95 −17.09 −18.61
−4.50 −3.03 −3.90 −2.15 −2.65
−3.24 −3.23 −3.34 −1.61 −1.28
1.1 1.02 1.85 1.23 2.19
1.4 1.4 1.4 1.4 1.4
WT gyrA, Arg91 bonded with ciprofloxacin atoms O2 and O4 while in the MT gyrA hydrogen bond with ciprofloxacin O4 was lost (Table 2), thereby contributing to a decrease in the docking score. Asp87 being directly hydrogen bonded to the ciprofloxacin H43 atom, plays a crucial role in the quinolone binding with gyrA. In the double mutant hydrogen bond length for residue 87 was increased from 2.76 Å in WT gyrA to 3.25 Å in double mutant, significantly decreasing the bond energy. In MT Ser83Ala and Asp87Glu docking score against ciprofloxacin was observed to be − 20.17 and − 23.17 involving ionic interactions with Arg 32, Lys 42, His 45, Arg 91 and Lys 42, Glu 87, Arg 91 respectively. Loss of interaction with Asp87 which is considered as hotspot amino acid in gyrA and fluoroquinolone interaction is assumed to be major cause of reduced binding affinity for Ser83Ala. Ser83Leu despite having bulkier side chain has relatively better binding score owing to the fact
that interaction with Asp 87 was conserved. As Glu and Asp both have negatively charged side chains, replacement of Asp with Glu at 87 has a less pronounced effect on ciprofloxacin binding as compared to other mutant forms. Other than hydrogen bonds, hydrophobic interactions also contribute to the docking score. Fluorine atom of ciprofloxacin made close contacts with Asp87. In WT gyrA complexed with ciprofloxacin, close atom–atom contact score (clash) was 2.45 lower as compared to Ser83Leu and Ser83Leu/Asp87Asn in which it was observed to be 3.24 and 2.67 (Table 3). In Ser83Leu clash score was highest because leucine being a bulkier residue, causes steric hindrance due to its side chain. In Ser83Leu lipophilic atom–atom contact score was also lowest (−5.34) which was − 5.61 and − 6.47 in WT gyrA and double mutant respectively. The lipophilic term is proportional to the contact surface area between the ligand and the protein involving non-polar terms and it contributes positively to the binding affinity. As a result binding affinity is reduced with a decrease in the lipophilic contacts. Hydrogen bonded interactions of ciprofloxacin with both the mutants are shown in Online Resource 1. We can discern from these findings that substitution of Ser with Leu at position 83 and Asp with Asn at position 87 in gyrA has a pronounced effect on ciprofloxacin binding and have been reported frequently. It is evident that mutation of both Ser83 and Asp 87 resulted in a relatively low docking score suggesting the critical role of these residues in ciprofloxacin binding with gyrA. This decreasing trend in docking score accounts for increased resistance in the MT gyrA, observed to be highest in the double mutant. Nalidixic acid interacts with WT gyrA through hydrogen bonds with residues Arg91 and Gln94 and hydrophobic interaction with neighboring residues Asp87, Phe96 and Ser111 with a high docking score of −12.51 (Table 1). Substitution of Ser with Leu at 83 resulted in loss of hydrogen bond with Gln94 residue and formation of new hydrogen
Fig. 7. Hydrogen bonds involved in binding of ciprofloxacin with A.1) wild type A.2) mutated S83L and A.3) S83L/D87N QRDR of gyrA. Similarly hydrogen bonds involved in QRDR complex with nalidixic acid in B.1) wild type B.2) mutated S83L and B.3) S83L/D87N.
Please cite this article as: Mehla, K., Ramana, J., Structural signature of Ser83Leu and Asp87Asn mutations in DNA gyrase from enterotoxigenic Escherichia coli and impact on quinolone resistance, Gene (2015), http://dx.doi.org/10.1016/j.gene.2015.09.063
K. Mehla, J. Ramana / Gene xxx (2015) xxx–xxx
bond with His45 and Arg91. As shown in Fig. 2, Arg91 shared weaker interactions with nalidixic acid atoms O1 and O3, with bond lengths being 3.19 Å and 2.90 Å respectively. In the double mutant, interactions were with Arg91 and Gln94 as in the WT gyrA (Fig. 7), but bond lengths increased as shown in Table 4, significantly reducing the docking score. Loss of one hydrogen bond with Gln94 in Ser83Leu mutant suggests a critical role of this residue in nalidixic acid binding with gyrA. Hydrogen bonded interactions of nalidixic acid with both the WT and MT are shown in Online Resource 1. In WT gyrA, close atom–atom contact score (clash) was 1.1, lower as compared to Ser83Leu/Asp87Asn in which it was observed to be 1.85. In Ser83Leu, clash score was almost similar to WT gyrA but the lipophilic contact score was much lower (− 3.03) as compared to WT gyrA (− 4.50). In Ser83Leu/Asp87Asn, lipophilic atom–atom contact score was − 3.90 (Table 3). Nalidixic acid interacts with MT Ser83Ala and Asp87Glu through weaker hydrogen bonds with residues Arg 91 and Gln 94 having relatively low docking score of − 8.35 and − 11.75 as compared to WT gyrA. MT Ser83Ala and Asp87Glu follow the similar trend for nalidixic acid as for ciprofloxacin. It is apparent that substitution of both Ser83 and Asp87 accounts for a decrease in the docking score and less efficient binding which ultimately leads to intermediate to high level drug resistance in the strains previously sensitive to drugs. The docking score is relatively high for ciprofloxacin because of the additional interactions with the fluorine atom, answering why resistance to the first generation nalidixic acid is higher as compared to the second generation drug ciprofloxacin. Our in silico docking results show good agreement with previous experimental studies on these mutations (Vila et al., 2000; Mendez Arancibia et al., 2009; Pazhani et al., 2011). Through molecular modeling studies we can derive a possible explanation for increased resistance in pathogens, but we cannot firmly establish the predictive powers of this approach since other mechanisms as well can contribute to resistance such as plasmid mediated resistance, or the presence of integrons and gene cassettes. Binding affinity is a function of the stability of ligand-target complex. Protein stability and interactions are important biological aspects that can be influenced by amino acid substitutions. The results presented in this paper provide an in silico based approach to seek insights into the effect of mutations at structural level and guide the pharmacologists to introduce necessary modifications in the existing drugs and paving a way for counteracting the problem of antibiotic resistance. 4. Conclusion A number of mutations localized in QRDR region of DNA gyrase have been reported that lead to quinolone resistance. Our computational molecular docking study shows that mutations at both Ser83 and Asp87 are primarily responsible for reduced affinity of these drugs towards gyrA and provides a possible explanation for observed quinolone resistance. This study provides interesting insights into molecular level mechanisms of antibiotic resistance due to mutations in quinolones, a major Table 4 Hydrogen bonds involved in gyrA-nalidixic acid interaction. Mutations Wild type
Receptor residue
Arg 91 Gln 94 Ser83Leu Arg 91 Arg 91 His 45 Ser83Leu/Asp87Asn Arg 91 Gln 94 Ser83Ala Arg 91 Gln 94 Asp87Glu Arg 91 Gln 94
Receptor atom
Ligand atom
Bond lengths (Å)
NH1 NE2 NH1 NH1 NE2 NH1 NE2 HH12 HE21 HH12 HE21
O2 N5 O1 O3 O2 O3 N5 O3 N5 O2 N5
2.71 2.74 3.19 2.90 2.68 2.88 3.12 1.82 2.17 2.09 1.85
7
class of antibiotics against diarrheagenic E. coli. Additionally, gaining knowledge about possible disease associations of mutations is another major research problem. The SNPs data is enormous, yet it is still not clear what biological effect do these mutations have. By studying the reported mutations at structural level, these studies open up new avenues for designing novel inhibitors against DNA gyrase and can be valuable in guiding the necessary modifications in the existing drugs to counteract drug resistance. Compliance with ethical standards Competing interests We confirm that there are no conflicts of interest associated with this publication. Acknowledgments This research was supported by DST (Department of Science and Technology), Ministry of Science and Technology, India under grant number SB/FT/LS-278/2012. The authors would like to thank the anonymous reviewers for their insightful comments that helped improve the content of the manuscript. Appendix A. Supplementary data Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.gene.2015.09.063. References Andersson, M.I., MacGowan, A.P., 2003. Development of the quinolones. J. Antimicrob. Chemother. 51 (Suppl 1), 1–11. http://dx.doi.org/10.1093/jac/dkg212. Böhm, H.J., 1994. The development of a simple empirical scoring function to estimate the binding constant for a protein-ligand complex of known three-dimensional structure. J. Comput. Aided Mol. Des. 8, 243–256. Bolton, E.E., Wang, Y., Thiessen, P.A., Bryant, S.H., 2008. Chapter 12 — PubChem: integrated platform of small molecules and biological activities. Annu. Rep. Comput. Chem. 217–241. Champoux, J.J., 2001. DNA topoisomerases: structure, function, and mechanism. Annu. Rev. Biochem. 70, 369–413. http://dx.doi.org/10.1146/annurev.biochem.70.1.369. Chen, J.Y., Siu, L.K., Chen, Y.H., et al., 2001. Molecular epidemiology and mutations at gyrA and parC genes of ciprofloxacin-resistant Escherichia coli isolates from a Taiwan medical center. Microb. Drug Resist. 7, 47–53. http://dx.doi.org/10.1089/ 107662901750152783. Conrad, S., Oethinger, M., Kaifel, K., et al., 1996. GyrA mutations in high-level fluoroquinolone-resistant clinical isolates of Escherichia coli. J. Antimicrob. Chemother. 38, 443–455. De la Cabada, B.J., Dupont, H.L., 2011. New developments in traveler's diarrhea. Gastroenterol. Hepatol. (N Y) 7, 88–95. Eisenberg, D., Lüthy, R., Bowie, J.U., 1997. VERIFY3D: assessment of protein models with three-dimensional profiles. Methods Enzymol. 277, 396–404. Ericsson, C.D., Johnson, P.C., Dupont, H.L., et al., 1987. Ciprofloxacin or trimethoprim–sulfamethoxazole as initial therapy for travelers' diarrhea. A placebo-controlled, randomized trial. Ann. Intern. Med. 106, 216–220. Fàbrega, A., Madurga, S., Giralt, E., Vila, J., 2009. Mechanism of action of and resistance to quinolones. Microb. Biotechnol. 2, 40–61. http://dx.doi.org/10.1111/j.1751-7915. 2008.00063.x. Frishman, D., Argos, P., 1995. Knowledge-based protein secondary structure assignment. Proteins 23, 566–579. http://dx.doi.org/10.1002/prot.340230412. Gellert, M., 1981. DNA topoisomerases. Annu. Rev. Biochem. 50, 879–910. http://dx.doi. org/10.1146/annurev.bi.50.070181.004311. Hawkey, P.M., 2003. Mechanisms of quinolone action and microbial response. J. Antimicrob. Chemother. 51 (Suppl 1), 29–35. http://dx.doi.org/10.1093/jac/dkg207. Isenbarger, D.W., Hoge, C.W., Srijan, A., et al., 2002. Comparative antibiotic resistance of diarrheal pathogens from Vietnam and Thailand, 1996–1999. Emerg. Infect. Dis. 8, 175–180. http://dx.doi.org/10.3201/eid0802.010145. Jiang, Z.D., Mathewson, J.J., Ericsson, C.D., et al., 2000. Characterization of enterotoxigenic Escherichia coli strains in patients with travelers' diarrhea acquired in Guadalajara, Mexico, 1992–1997. J. Infect. Dis. 181, 779–782. http://dx.doi.org/10.1086/315272. Kaper, J.B., Nataro, J.P., Mobley, H.L., 2004. Pathogenic Escherichia coli. Nat. Rev. Microbiol. 2, 123–140. http://dx.doi.org/10.1038/nrmicro818. Laskowski, R.A., MacArthur, M.W., Moss, D.S., Thornton, J.M., 1993. PROCHECK: a program to check the stereochemical quality of protein structures. J. Appl. Crystallogr. 26, 283–291. http://dx.doi.org/10.1107/S0021889892009944. Mendez Arancibia, E., Pitart, C., Ruiz, J., et al., 2009. Evolution of antimicrobial resistance in enteroaggregative Escherichia coli and enterotoxigenic Escherichia coli causing
Please cite this article as: Mehla, K., Ramana, J., Structural signature of Ser83Leu and Asp87Asn mutations in DNA gyrase from enterotoxigenic Escherichia coli and impact on quinolone resistance, Gene (2015), http://dx.doi.org/10.1016/j.gene.2015.09.063
8
K. Mehla, J. Ramana / Gene xxx (2015) xxx–xxx
traveller's diarrhoea. J. Antimicrob. Chemothe.r 64, 343–347. http://dx.doi.org/10. 1093/jac/dkp178. Nguyen, T.V., Le, P.V., Le, C.H., Weintraub, A., 2005. Antibiotic resistance in diarrheagenic Escherichia coli and Shigella strains isolated from children in Hanoi, Vietnam. Antimicrob. Agents Chemother. 49, 816–819. http://dx.doi.org/10.1128/AAC.49.2. 816-819.2005. PATH, bvgh, 2011. The Case for Investment in Enterotoxigenic Escherichia coli Vaccines. Pazhani, G.P., Chakraborty, S., Fujihara, K., et al., 2011. QRDR mutations, efflux system & antimicrobial resistance genes in enterotoxigenic Escherichia coli isolated from an outbreak of diarrhoea in Ahmedabad, India. Indian J. Med. Res. 134, 214–223. Reece, R.J., Maxwell, A., 1991. DNA gyrase: structure and function. Crit. Rev. Biochem. Mol. Biol. 26, 335–375. http://dx.doi.org/10.3109/10409239109114072. Sáenz, Y., Zarazaga, M., Briñas, L., et al., 2003. Mutations in gyrA and parC genes in nalidixic acid-resistant Escherichia coli strains from food products, humans and animals. J. Antimicrob. Chemother. 51, 1001–1005. http://dx.doi.org/10.1093/jac/ dkg168. Schneider, N., Hindle, S., Lange, G., et al., 2012. Substantial improvements in large-scale redocking and screening using the novel HYDE scoring function. J. Comput. Aided Mol. Des. 26, 701–723. http://dx.doi.org/10.1007/s10822-011-9531-0. Shindyalov, I.N., Bourne, P.E., 1998. Protein structure alignment by incremental combinatorial extension (CE) of the optimal path. Protein Eng. 11, 739–747. Simossis, V.A., Heringa, J., 2005. PRALINE: a multiple sequence alignment toolbox that integrates homology-extended and secondary structure information. Nucleic Acids Res. 33, W289–W294. http://dx.doi.org/10.1093/nar/gki390.
Sizemore, D.R., Roland, K.L., Ryan, U.S., 2004. Enterotoxigenic Escherichia coli virulence factors and vaccine approaches. Expert Rev. Vaccines 3, 585–595. http://dx.doi.org/ 10.1586/14760584.3.5.585. Taylor, D.N., Sanchez, J.L., Candler, W., et al., 1991. Treatment of travelers' diarrhea: ciprofloxacin plus loperamide compared with ciprofloxacin alone. A placebo-controlled, randomized trial. Ann. Intern. Med. 114, 731–734. Vila, J., Ruiz, J., Marco, F., et al., 1994. Association between double mutation in gyrA gene of ciprofloxacin-resistant clinical isolates of Escherichia coli and MICs. Antimicrob. Agents Chemother. 38, 2477–2479. Vila, J., Vargas, M., Casals, C., et al., 1999. Antimicrobial resistance of diarrheagenic Escherichia coli isolated from children under the age of 5 years from Ifakara, Tanzania. Antimicrob. Agents Chemother. 43, 3022–3024. Vila, J., Vargas, M., Ruiz, J., et al., 2000. Quinolone resistance in enterotoxigenic Escherichia coli causing diarrhea in travelers to India in comparison with other geographical areas. Antimicrob. Agents Chemother. 44, 1731–1733. WHO, 2006. Future directions for research on enterotoxigenic Escherichia coli vaccines for developing countries. Wkly Epidemiol. Rec. 81, 97–104. Woodford, N., Ellington, M.J., 2007. The emergence of antibiotic resistance by mutation. Clin. Microbiol. Infect. 13, 5–18. http://dx.doi.org/10.1111/j.1469-0691.2006.01492.x. Yoshida, H., Bogaki, M., Nakamura, M., Nakamura, S., 1990. Quinolone resistance-determining region in the DNA gyrase gyrA gene of Escherichia coli. Antimicrob. Agents Chemother. 34, 1271–1272.
Please cite this article as: Mehla, K., Ramana, J., Structural signature of Ser83Leu and Asp87Asn mutations in DNA gyrase from enterotoxigenic Escherichia coli and impact on quinolone resistance, Gene (2015), http://dx.doi.org/10.1016/j.gene.2015.09.063