Comparative structural analysis of two proteins belonging to quorum sensing system in Vibrio cholerae

This article was downloaded by: [Institute of Life Sciences] On: 02 July 2012, At: 04:38 Publisher: Taylor & Francis Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK

Journal of Biomolecular Structure and Dynamics Publication details, including instructions for authors and subscription information: http://www.tandfonline.com/loi/tbsd20

Comparative structural analysis of two proteins belonging to quorum sensing system in Vibrio cholerae a

b

c

Mobashar Hussain Urf Turabe Fazil , Sunil Kumar , Naidu Subba Rao , Chandrabose d

d

e

Selvaraj , Sanjeev Kumar Singh , Haushila Prasad Pandey & Durg Vijai Singh

a

a

Infectious Disease Biology, Institute of Life Sciences, Nalco Square, Bhubaneswar, 751023, India b

Bioinformatics Centre, Institute of Life Sciences, Nalco Square, Bhubaneswar, 751023, India c

School of Computational and Integrative Sciences, Jawaharlal Nehru University, New Delhi, 110067, India d

Computer Aided Drug Design and Molecular Modeling Laboratory, Department of Bioinformatics, Alagappa University, Karaikudi, 630003, Tamil Nadu, India e

Faculty of Science, Department of Biochemistry, Banaras Hindu University, Varanasi, 221005, India Version of record first published: 26 Jun 2012

To cite this article: Mobashar Hussain Urf Turabe Fazil, Sunil Kumar, Naidu Subba Rao, Chandrabose Selvaraj, Sanjeev Kumar Singh, Haushila Prasad Pandey & Durg Vijai Singh (2012): Comparative structural analysis of two proteins belonging to quorum sensing system in Vibrio cholerae , Journal of Biomolecular Structure and Dynamics, DOI:10.1080/07391102.2012.687523 To link to this article: http://dx.doi.org/10.1080/07391102.2012.687523

PLEASE SCROLL DOWN FOR ARTICLE Full terms and conditions of use: http://www.tandfonline.com/page/terms-and-conditions This article may be used for research, teaching, and private study purposes. Any substantial or systematic reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any form to anyone is expressly forbidden. The publisher does not give any warranty express or implied or make any representation that the contents will be complete or accurate or up to date. The accuracy of any instructions, formulae, and drug doses should be independently verified with primary sources. The publisher shall not be liable for any loss, actions, claims, proceedings, demand, or costs or damages whatsoever or howsoever caused arising directly or indirectly in connection with or arising out of the use of this material.

Journal of Biomolecular Structure and Dynamics iFirst, 2012, 1–11

Comparative structural analysis of two proteins belonging to quorum sensing system in Vibrio cholerae Mobashar Hussain Urf Turabe Fazila, Sunil Kumarb, Naidu Subba Raoc, Chandrabose Selvarajd, Sanjeev Kumar Singhd, Haushila Prasad Pandeye and Durg Vijai Singha* a Infectious Disease Biology, Institute of Life Sciences, Nalco Square, Bhubaneswar 751023, India; bBioinformatics Centre, Institute of Life Sciences, Nalco Square, Bhubaneswar 751023, India; cSchool of Computational and Integrative Sciences, Jawaharlal Nehru University, New Delhi 110067, India; dComputer Aided Drug Design and Molecular Modeling Laboratory, Department of Bioinformatics, Alagappa University, Karaikudi 630003, Tamil Nadu, India; eFaculty of Science, Department of Biochemistry, Banaras Hindu University, Varanasi 221005, India

Downloaded by [Institute of Life Sciences] at 04:38 02 July 2012

Communicated by Ramaswamy H. Sarma (Received 2 November 2011; final version received 14 March 2012) Vibrio cholerae uses quorum sensing communication system to interact with other bacteria and for gauzing environmental parameters. This organism dwells equally well in both human host and aquatic environments. Quorum sensing regulates multitude of activities and is one of the lucrative targets presently pursued for drug design in bacteria to encounter virulence. Histidine phosphotransfer protein LuxU and response regulator LuxO of V. cholerae are known to play important roles in biofilms and virulence machinery. In the present study, we used computational methods to model LuxU and LuxO and simulated the interactions of LuxO and LuxU. Since no structural details of the proteins were available, we employed homology modeling to construct the three-dimensional structures and then performed molecular dynamics simulations to study dynamic behavior of the LuxO and LuxU from V. cholerae. The modeled proteins were validated and subjected to molecular docking analyses. This allowed us to predict the binding modes of the proteins to elucidate probable sites of interference. Keywords: Vibrio cholerae; two component signaling system; molecular dynamics; protein modeling; drug targets

Introduction Vibrio cholerae is a natural inhabitant of brackish water and estuarine systems and causative agent of infectious diarrheal disease cholera (Colwell, Kaper, & Joseph, 1977). Biofilm formation and viable but nonculturable state are important characteristics in long-term survival of vibrios in the environment (Alam et al., 2007). Environmental clones of V. cholerae, possessing characteristics of pathogenic isolates are known to have been associated with clinical isolates causing cholera outbreaks (Singh et al., 2001). Bacterial survival depends on integration of multicellular responses and acclimatizing to changes that occur in the surrounding environment. This is accomplished through a bacterial cell–cell communication process called quorum sensing. Quorum sensing in V. cholerae controls virulence and biofilm formation through synthesis, release, and subsequent detection of signaling mole-

*Corresponding author. Email: [email protected] ISSN 0739-1102 print/ISSN 1538-0254 online Copyright Ó 2012 Taylor & Francis http://dx.doi.org/10.1080/07391102.2012.687523 http://www.tandfonline.com

cules called autoinducers (Hammer & Bassler, 2003; Zhu et al., 2002). At least three distinct sensory pathways are known to effect regulation of virulence gene expression (Figure 1), biofilm formation, and protease production (Miller, Skorupski, Lenz, Taylor, & Bassler, 2002). Quorum sensing system-1 is composed of the CAI-1 autoinducer and a two-component sensor kinase Cqs. System 2 is composed of AI-2 (a furanosyl borate diester autoinducer), the periplasmic binding protein LuxP, and twocomponent sensor kinase LuxQ (Chen et al., 2002). The sensory information from both of these systems is conveyed through a phosphorelay mechanism mediated by a phosphorelay protein LuxU. Autophosphorylation of sensor histidine kinases of the respective quorum sensing circuitries upon activation leads to transfer of phosphate to phosphotransfer protein LuxU. LuxU interacts with various proteins and acts both as a phospho donor and acceptor. LuxU transfers

2

M.H. Fazil et al.

Downloaded by [Institute of Life Sciences] at 04:38 02 July 2012

Material and methods

Figure 1. Three parallel systems of quorum sensing involved in virulence gene regulation in V. cholerae.

the phosphate to LuxO. Phospho-LuxO triggers the transcription of genes encoding sRNA’s that destabilize mRNA encoding transcription factor HapR, the terminal effector (Jobling & Holmes, 1997; Lenz et al., 2004). Mutations in conserved aspartate residue of LuxO receiver domain severely inhibit the terminal regulator activities and acts as a central switch in coordinate regulation of virulence-related phenotypes such as protease production and biofilm formation (Vance, Zhu, & Mekalanos, 2003). A luxO mutant is severely defective in colonizing small intestine and production of cholera toxin (Zhu et al., 2002). The ability of this pathogen to transpose between expression of virulence traits and biofilm formation are necessary to its infection cycle and survival. Variable antibiotic resistance had been found in V. cholerae strains isolated from different cholera epidemics varying in space and time. An increase in multidrug resistant phenotype of V. cholerae strains had been observed by continuous monitoring of epidemic strains (Ramamurthy, 2008). Since bacterial signal transduction systems including quorum sensing circuit has no counterpart in the human host, components of this machinery are considered for development of drugs and vaccines against human pathogens. Quorum sensing proteins in other enteropathogenic bacteria have been used successfully to evaluate the effectiveness of microbial inhibitors (Rasko et al., 2008). Designs of quorum sensing antagonists or agonists require understanding of quorum sensing regulation through structural and chemical properties of regulators. In the absence of structural data, a model built on known three-dimensional (3D) structures of homologous proteins is the only reliable approach. Therefore, in the present study, we built insilico 3D structures of V. cholerae LuxU and LuxO analyzed their molecular interactions through the model and evaluated physical basis of activity.

Template search and sequence alignment The amino acid sequence of LuxO and LuxU of V. cholerae (O1 biotype El Tor strain N16961) was retrieved from the sequence database of NCBI (www.ncbi.nlm.nih. gov) (ID: LuxO: NP_230666.1, LuxU: NP_230667.1). The 3D structures of these proteins were not available in Protein Data Bank (PDB). Hence, the present exercise of developing the 3D models of the LuxO and LuxU of V. cholerae was undertaken. To find suitable templates for homology modeling, BLASTP (Altschul, Gish, Miller, Myers, & Lipman, 1990) search was performed against the Brookhaven PDB (Berman et al., 2000) with the default parameters. Based on the maximum identity with high score and lower e-value crystal structures of the joined N-terminal regulatory and central ATPase domains (NtrC1RC) of NtrC1 from Aquifex aeolicus (1NY5), an N-terminal domain deletion of the ZraR from Salmonella typhimurium (1OJL), a putative LuxO repressor protein from Vibrio parahaemolyticus (3CFY) for LuxO and LuxU and from Vibrio harveyi (1Y6D) for LuxU were used as templates for building 3D structures. The sequence alignments of LuxO with 1NY5, 1OJL, and 3CFY and that of LuxU with 1Y6D were carried out using the CLUSTAL W (http://www.ebi.ac.uk/clustalw) program (Thompson, Higgins, & Gibson, 1994). The quality of the structure employed for model generation is 2.4 Å (1NY5), 3.0 Å (1OJL), and 2.5 Å (3CFY), of which 1Y6D was nuclear magnetic resonance (NMR) structure. The default settings used in BLAST were 10 (expect threshold), 3 (word size), BLOSUM62 (Scoring matrix), existence 11, extension 1 (Gap Costs), and conditional compositional scoring matrix adjustment. The ClustalW default parameters were Gonnet (protein weight matrix), 10 (gap open), 0.2 (gap extension), 5 (gap distance), 1 (numiter), and NJ (clustering) with no end gaps and interaction. The output options for ClustalW were Aln w/number (format) and aligned (Order). 3D structure generation The academic version of MODELLER9v6 (http://salilab. org/modeller/) was used for 3D structure generation based on the information obtained from sequence alignment (Sali & Blundell, 1993). Out of 20 models generated by MODELLER for both the proteins, one model central to cluster was selected and subjected to stereochemical check to find the deviations from normal bond length, dihedrals, and nonbonded atom–atom distances. Each model (LuxO and LuxU) with the highest G-score of PROCHECK (Laskoswki, MacArthur, Moss, & Thornton, 1993) and VERIFY3D (Eisenberg, Luthy, & Bowie, 1997) profile was subjected to energy minimization. The energy minimization was started with side chains and then applied to main chain of Cα backbone.

Two Proteins Belonging to Quorum Sensing System

Downloaded by [Institute of Life Sciences] at 04:38 02 July 2012

All calculations were performed by using ACCELRYS DS modeling 2.5 (Accelrys Inc. San Diego, CA, USA) software suite. STRIDE (Heing & Frishman, 2004) was used in prediction of secondary structure of the modeled LuxO and LuxU. PROSA was used for calculating Zscores (Wiederstein & Sippl, 2007). The weighted root mean square deviation (RMSD) of the modeled protein was calculated using the combinatorial extension (CE) algorithm (Shindyalov & Bourne, 1998). The modeled structure was then superimposed on the crystal template without altering the coordinate systems of atomic position in the template. The residue profiles of the 3D models were further checked using VERIFY3D. PROCHECK analysis was performed to assess the stereochemical properties of the 3D models and Ramachandran plots. Molecular dynamics simulations Desmond was used to perform molecular dynamics simulations (MDS) on the modeled proteins (LuxU and LuxO) and complexes (LuxO–ATP, LuxU–LuxO, and LuxU–LuxO–ATP) (Maestro version, 2008). We used OPLS-AA force field in combination with the TIP4P water model to describe the interactions between the atoms (Bowers et al., 2006; Kaminski, Friesner, TiradoRives, & Jorgensen, 2001). An orthorhombic box 10 × 10 × 10 Å was used for simulations. RESPA multiple time step integrator (Tuckerman, Berne, & Martyna, 1992) was employed with a 2 fs time step and relaxation time of 2.0 ps without restraints. The system was neutralized by Na+ and Cl counter ions at 0.5 M concentration. Particle-mesh Ewald method was used to calculate longrange electrostatic interactions with Ewald tolerance of 1e 9 (Essmann et al., 1995). Van der Waals and shortrange electrostatic interactions were smoothly truncated at 9.0 Å. Nose–Hoover thermostats were utilized to maintain constant simulation temperature and the Martina–Tobias–Klein barostat method was used to control pressure (Martyna, Klein, & Tuckerman, 1992; Martyna, Tobias, & Klein, 1994). The system was minimized using a hybrid method of steepest decent and limitedmemory Broyden–Fletcher–Goldfarb–Shanno (LBFGS) algorithms and the simulation was started directly from the initial configuration with constant number of atoms, pressure and temperature (NPT) relaxation protocol. Following equilibration, a molecular simulation of 10 ns and configuration recording interval of one picosecond were run. The whole system was subjected to MD simulations at temperature of 300 K, pressure 1.0 bar with constant area coupling style. Positional restraints, keeping the whole protein as part of selection with a force constant (100.0 kcal/mol/Å2), were applied for this calculation as described previously (Guo et al., 2010; Shivakumar et al., 2010). The final MD calculations were performed for 10 ns under the same conditions, except that the posi-

3

tion restraints were removed. Dyndom was used for analysis of motion in proteins after final MDS run (http:// fizz.cmp.uea.ac.uk/dyndom/). Protein–ligand interaction The chemical structures of ATP molecule were extracted from pubchem database (http://pubchem.ncbi.nlm.nih. gov). Structure of the ligand was retrieved into twodimensional MDL/SDF format and 3D coordinates were generated using the CORINA program (Tetko et al., 2005). The molecule was then read into INSIGHT II for further treatment of energy minimization for 100 steps with consistent valence force field (CVFF) (Hagler, Huler, & Lifson, 1974). Genetic optimization for ligand docking (GOLD) version 4.1.1 (Cambridge Crystallographic Data Centre, Cambridge, UK) was used for docking ATP for 50 times in the standard default settings (Jones, Willett, Glen, Leach, & Taylor, 1997). For ligand–protein binding, 10 docking conformations (poses) were tested and the best GOLD score was selected for studies. GOLD uses a genetic algorithm to explore the full range of ligand conformational flexibility with partial flexibility of the protein (Jones et al., 1997). The binding affinity between the protein and ligand was estimated by using the consensus scoring function XScore V2.1 (Wang, Lai, & Wang, 2002). SILVER was used to predict the interactions of LuxO–ATP (Jones et al., 1997). The ligand showing maximum interactions with the protein was plotted using the program LIGPLOT (Wallace, Laskowski, & Thornton, 1995). Hydrogen bond interactions were double-checked with the software GETNEARES, available with the program DOCK (Ewing, Makino, Skillman, & Kuntz, 2001). The complex was subjected to MD simulations postdocking. Protein–protein docking of LuxO and LuxU A geometry-based molecular docking algorithm called PatchDock (http://bioinfo3d.cs.tau.ac.il/PatchDock) was used to dock the predicted 3D models of LuxO and LuxU. The complexes were further optimized by another docking program FireDock (http://bioinfo3d.cs.tau.ac.il/ FireDock). The theoretical binding energies of the docked complexes predicted by DCOMPLEX server (http:// sparks.informatics.iupui.edu/song/complex.html) were useful in verifying the stability of the complex. A default value of 4 Å was used for clustering and redundant solutions were discarded by RMSD clustering. The PatchDock output generates the geometric score, desolvation energy, interface area size, and the actual rigid transformation of the solutions. Twenty solutions, out of about 60 predicted LuxO–LuxU complexes, were sorted according to their geometric shape. The complementarity scores were analyzed for identifying the residues involved in the protein–protein interface.

4

M.H. Fazil et al.

Downloaded by [Institute of Life Sciences] at 04:38 02 July 2012

Results Model building High-resolution structure of LuxO and LuxU proteins in V. cholerae has not yet been determined experimentally (X-ray or NMR). We built a model following homology modeling protocol. BLASTP search was performed against PDB with default parameters to find suitable templates for homology modeling. Based on the maximum identity with high score and lower evalue 1NY5 (Identity = 42%, Similarity = 62%), 1OJL (Identity = 42%, Similarity = 58%), and 3CFY (Identity = 46%, Similarity = 66%), for LuxO and 1Y6D (Identity = 51%, Similarity = 72%), for LuxU were used as templates for homology modeling (Figure S1(a) and (b)). Sequence alignment between target and template was done using ClustalW program, characterized by some insertions and deletions in the loop regions. Because the last 70 residues at C-terminal (residues 385–455) in LuxO did not have the corresponding equivalent region in 1NY5, 1OJL, and 3CFY, the modeling was carried out from 1st to the 384th residues followed by a rigorous refinement of the model by means of molecular dynamics and energy minimization. Molecular dynamics simulations Explicit solvent MDS of LuxU and LuxO modeled protein showed good stability in the simulation point. Desmond working panel obtained the stability of the protein from maestro – simulation quality analysis. RMSD of Cα backbone in LuxO is 0.7 Å and in LuxU is 2.3 Å from time 0 to the end of 10 ns MDS run (Figure 2). Dyndom, domain motion analysis software (http://fizz/ cmp.uea.ac.uk/dyndom/) was used to assess domain flexibility. The architecture of the proteins remained relatively similar after simulation. The final stable protein structures after simulation event are represented as ribbon models in Figure 3(a) and (b).

Figure 2. time.

Figure 3. PyMOL representation of modeled LuxO (1–384 residues lacking complete DNA binding domain) (a) and LuxU (b). The α-helices and β-sheets are shown as helices and ribbons in red and yellow, respectively. The rest are shown as loops in green.

Protein structure validation To validate the homology modeled LuxO and LuxU structure, a Ramachandran plot was drawn. The structure was analyzed by PROCHECK, a well-known protein structure checking program. Procheck Ramachandran

Observed RMSD for modeled proteins and complexes after Desmond simulation event analysis from 0 to 10 ns simulation

Downloaded by [Institute of Life Sciences] at 04:38 02 July 2012

Two Proteins Belonging to Quorum Sensing System plot analysis of LuxO shows 94.1% of residues in the core region, 5.3% in the allowed region, 0.6% in the generously allowed region, and none of the residues lied in the disallowed conformations (Figure S2(a)) and LuxU had 87.6% of residues in the core region, 9.5% in the allowed region, 2.9% in the generously allowed region, and none of the residues lied in the disallowed conformations (Figure S2(b)). The observed G-factor of dihedrals model was 0.02 for the LuxO and 0.19 for LuxU. An overall decrease was observed in the G-factor after MD simulation. These observations thus indicate that the modeled structure had an increase in the number of bad dihedral angles. This may be due to MD simulation causing an unfavorable dihedral angle, allowing the protein to overcome high-energy barriers. The energetic architecture, as predicted by PROSA score, was negative ( 9.16 and 4.14) for the modeled proteins (LuxO and LuxU), quite similar to template, ( 10.14 1NY5, 7.16 1OJL, 6.73 3CFY, and 3.55 for 1Y6D), indicating the reliability of the model. The structural superimposition of Cα trace of the LuxO model over template structure (1NY5, 1OJL, and 3CFY) resulted in RMSD of 2.2, 1.7, and 1.8 Å, respectively. Using the CE program (http://cl.sdsc.edu/ce.html) RMSD between the LuxU model and its template structure (1Y6D) was found to be 0.3 Å indicating valid structure of the model. Secondary structures were predicted using STRIDE for modeled LuxO and LuxU (Figure S3(a) and (b)). Protein–protein docking The modeled structures of LuxO and LuxU after optimization were docked to understand the mode of interaction between these two proteins using Patchdock program (http://bioinfo3d.cs.tau.ac.il/PatchDock) (Table S1). The top ranking 20 solutions (protein complexes) were further analyzed to calculate the hydrogen bonds, hydrophobic contacts, and nonbonded contacts by using LIGPLOT program. Optimization of the docked complexes was done by FireDock program (http://bioinfo3d.cs.tau.ac.il/FireDock) (Table S2). The scores representing, predicted binding free energy for the top ranked solutions, number of structurally aligned residues appearing within the distance cut-off of 2.25–3.6 Å and the number of hot spot residues in the protein–protein interface were analyzed by DCOMPLEX server. LuxO and LuxU complex best in terms of global energy and atomic contact energy of the complex had six hydrogen bonds, 100 hydrophobic contacts, and 294 nonbonded contacts. The hydrogen bonds and hydrophobic contacts are shown in Ligplot (Figure 4). The residues present at the interface are D121, D123, G134, Y143, Y315, and V330 contributed by LuxO and E32, D35, K48, S65, and F66 by LuxU. Although we included these residues for interaction sites in protein–protein interactions, we did not see direct involvement of D47

5

residue in stabilizing the complex either through hydrogen bonds or hydrophobic interactions. Sampling of protein–protein interaction results obtained based on energy minima indicate that H57 of LuxU interacts with G131 in LuxO (Figure 5). All residues at the interface were conserved at sequence level demonstrated by sequence alignment and conserved domain search at NCBI (Freeman & Bassler, 1999; Raychaudhuri, Jain, & Dongre, 2006; Vance et al., 2003). The surface representation of LuxU and LuxO complex showed good shape complementarity. Protein–ligand interaction Studies on other response regulators (NtrC1) carrying an AAA+ATPase site showed that the residues involved in various conserved motifs have a direct role in defining transcription activity associated with the regulator (Lee et al., 2003). Therefore, docking with ATP was carried out on the binding site of LuxO using GOLD software applying the parameters of standard default setting with 50 genetic algorithm runs, filtering poses based on GOLD fitness function. The residues E161, S162, G163, T164, G165, K166, E167, V168, D231, and N273 were taken for interaction studies with ATP, based on sequence homology in LuxO of V. cholerae. These residues taken for binding site of ATP were based on homology studies of various response regulators (Marchler-Bauer et al., 2011). The binding pocket of each protein had regions corresponding to variable amino acid sequence and the 3D structure interacting with ligands in a specific manner. We, therefore, defined the binding site of LuxO having these residues and neighboring residues within 4 Å for docking with ATP. To substantiate the estimations done by the GOLD program, we used consensus scoring program X-Score. The scoring schema used in the software X-Score computes a binding score for a given protein–ligand complex structure. The predicted binding energy for the docked complex was 5.51 kcal/mol and predicted average log Kd was 3.77, found using X-Score program. This complex was then subjected to MD simulations for 10 ns. Discussion Typical histidine phosphotransfer proteins of bacteria are α-helical bundles (Ulrich, Kojetin, Bassler, Cavanagh, & Loria, 2005). Similarly, V. cholerae LuxU also showed four α-helices, much in accord with the template used (Figure S1(b)). LuxU mediates phospho-transfer signaling in at least two of the three predicted parallel systems working in V. cholerae (Miller et al., 2002). Although there is little sequence homology between V. cholerae LuxU and other known bacterial phosphotransfer proteins, the conserved active sites were intact of which histidine residue involved in phosphotransfer resides at

Downloaded by [Institute of Life Sciences] at 04:38 02 July 2012

6

M.H. Fazil et al.

Figure 4.

Ligplot for LuxO (first surface in purple) and LuxU (second surface in yellow) interaction using LIGPLOT software.

position 57. The positively charged residues that help in phosphate binding in the active site region reside at positions 53 and 60, respectively. His-57 was in the center of second α-helix flanked by lysines 53 and 60, providing the necessary charge at the active site. NMR studies of V. harveyi LuxU predicted the presence of histidine residue in the terminal helix, important for activity (Varughese et al., 1998). However, V. cholerae LuxU model showed two additional histidine residues (His105 and His113) in the terminal helix that might have unknown function assigned to the secondary structure. MD simulations in LuxU stretch the α-helical components to β turns. This is not unexpected as α-β transitions are common among α-helical proteins. These deformations are established to have important connectors between proteins mechanical strength and 3D structure (Buehler & Yung, 2009).

Response regulators are usually multidomain proteins consisting of a conserved N-terminal regulatory (or receiver) domain and a variable C-terminal effector domain that elicits the specific output response of the system, most commonly the transcriptional regulation (Gao, Mack, & Stock, 2007). Bacterial response regulators possess a conserved structure and shares similar conformational activation. LuxO belongs to the NtrC family of response regulators (Lilley & Bassler, 2000). The defining feature of response regulators is the presence of structurally conserved α/β domain, referred to as a regulatory or receiver domain. This consists of five-stranded parallel β sheet surrounded by five amphipathic helices (Neuwald, Aravind, Spouge, & Koonin, 1999). Interestingly a majority of response regulators studied so far, had sequence similarities of 20–30%, harboring the conserved receiver domain consisting of five β sheets

Downloaded by [Institute of Life Sciences] at 04:38 02 July 2012

Two Proteins Belonging to Quorum Sensing System

Figure 5. Protein–protein interactions between LuxO and LuxU generated by PDBSum (red: polar acidic; green: polar neutral; gray: aliphatic; purple: aromatic; orange: glycine; and yellow: cysteine).

surrounded by as many amphipathic α helices (Stock, Robinson, & Goudreau, 2000). But in the case of V. cholerae O1 El Tor (Strain N16961), we found a unique α/β domain with three central β sheets and five surrounding α helices, though the sequence identity with the templates was more than 40% (Figure S1(a)). This may be due to high sequence variability at critical positions of amino acids which defines the α/β receiver subdomain of this response regulator. The receiver domain also encompasses a dimerization interface that plays a

7

significant role in effectors functions of the response regulators (Bachhawat, Swapna, Montelione, & Stock, 2005). Overall variation observed in domain structure would lead to assumption that the mode of activation in this protein might vary compared to other known response regulators. Phosphorylation at the conserved site has been known to induce conformational changes exposing the hydrophobic surface of the protein, thus helping in activation. A conserved Aspartate residue, the site of phosphorylation, occurs at position 47. The NtrC/DctD subfamily of response regulator’s is characterized by the presence of a central AAA+ATPase domain located between the receiver domain and the helix-turn-helix motif (Lee et al., 2003; Sallai & Tucker, 2005). The role of the AAA+ATPase domain is to induce open complex formation in σ54 (Sigma factor 54) containing RNA polymerase in an ATP dependent manner. Our modeled structure predicts that V. cholerae LuxO also harbors an AAA+ATPase domain. The characteristic features of this domain include the nucleotide binding walker A and B motifs and an arginine finger motif. Walker A motif comprises of the consensus sequence GXXGXGK (where X is any amino acid). This sequence is found in a loop that facilitates phosphate moieties binding to the nucleotide. This region of specificity in this protein is observed at positions 160–167. The walker B motif is important for magnesium ion binding and thus in stabilizing the nucleotide complex. It consists of a conserved consensus sequence that contains four aliphatic amino acid residues followed by two negatively charged residues. This motif (TLFLDE) is detected at positions 228 through 233 in V. cholerae LuxO. An arginine finger is a specific motif comprising of a conserved arginine residue that is required for facilitating ATP binding and hydrolysis. The conserved Arginine of the Arginine finger is observed at position 292. To validate the domain and to check the interactions of the given domain with its ligand, we employed docking simulations of LuxO protein with ATP. The binding pocket was defined according to known interactions in template, sequence homology, and molecular biology studies. The GOLD-docking score for 1OJL–ATP complex is 72.7750, for 1NY5–ATP complex is 75.9085, and for LuxO–ATP complex is 71.4160. In this regard, we assume that the interactive capabilities of ATP within the modeled domain are satisfactory. The ratio of intra-domain to inter-domain displacement, due to the presence of ATP, in LuxO is 1.85. A rotation angle of 50.1° observed in LuxO after ATP docking and residues 116–134 act as hinge for domain motion. An interaction with ribose moiety of ATP in 1OJL is stabilized by Arg residue at 29 position whereas in case of 1NY5, it is shared by Glu137 and Lys 360. Analogously in LuxO, Leu 196 and Asp 195 stabilize these interactions. Glycine residue is invariably involved

Downloaded by [Institute of Life Sciences] at 04:38 02 July 2012

8

M.H. Fazil et al.

Figure 6.

Schrodinger view of interaction between ATP and LuxO protein.

in interactions with triphosphate or diphosphate groups in 1OJL (Gly 172) and in 1NY5 (Gly 172 and Gly 170), respectively. In LuxO, this glycine residue is mapped at position 257 (Figure 6). Although predictions of DNA binding domains indicated presence of helix-turn-helix motif from amino acid sequence positions 421–442 (data not shown), we were unable to extrapolate the model to the DNA binding domain because of lack of proper template that could give a stable structure at the C-terminus in LuxO. Keeping in view the importance of signal propagation in this circuit to the organism, protein–protein interactions between LuxU and LuxO are vital to the understanding of molecular basis of transduction and to define prospective sites of interference. The protein–protein interaction studies reveal that the molecular interface of the two proteins is stabilized by six hydrogen bonds and many hydrophobic and nonbonded contacts. The results also indicate that the inter domain regions connecting the receiver and AAA+ATPase domain of LuxO (Leu118, Gly131, Tyr315, Glu171, Lys65, Tyr143, Lys 178, Asp181, Gly180, and Arg331) are involved in binding with residues Lys48, His57 Ser62, Lys53, Ala68, Thr44, and Glu16 of helix 1 and 2 of LuxU. This protein–protein interaction produces a rotation of 19.5 degree due to motion in amino acid residues from 116 to

147 in LuxO. The existing model of phosphorelay indicates that the phosphorylation of LuxO by LuxU triggers oligomerization which in turn would undergo ATP hydrolysis to effectively promote its action by binding to DNA transcribing region (Hammer & Bassler, 2003; Miller et al., 2002; Vance et al., 2003). Interestingly, the interactions of ATP with LuxO vary in the presence of LuxU. It was observed that Asp237 and Glu198 stabilized the phosphate moiety, whereas adenine ring interacts with Ala190. However, after 10 ns of MD simulations, Lys194 was found interacting with the adenine ring of ATP (Figure 7). The percentage of domain closure due to the presence of LuxU was 41.7% compared to initial complex. The residues involved in domain closure were Lys117, Leu118, Lys119, and Asn120 (percentage progress in bending is 0.3, 66.6, 114.6, and 15.5, respectively). Because of their importance in phosphorelay, targeted disruption of the LuxO– ATP interfaces could result in probable loss of function in the quorum sensors similar to effect observed by the targeted disruption of the LuxO–LuxU interfaces. These results indicate variations to earlier predicted mode of action for LuxO and hence protein–protein interactions among LuxO and LuxU proteins. The resultant modulations shown by LuxO could possibly suggest either multifunctional role or unique mode of action in this protein.

Downloaded by [Institute of Life Sciences] at 04:38 02 July 2012

Two Proteins Belonging to Quorum Sensing System

9

Figure 7. Interactions of ATP–LuxO in LuxU–LuxO ATP complex. Pink color indicates initial pose of ATP docking, whereas white frame indicates post MD simulation status.

Because of their importance in environmental survival, biofilm formation, and virulence gene regulation these models can serve as a valuable reference for structural studies, drug design, and validation or vaccine until further biophysical and structural studies on these proteins.

sensors, which have no counterparts in human host. Their precise role in virulence inhibition and decreasing the trend of antibiotic resistance in pathogens, these proteins can be effectively targeted for drug and vaccine design. Abbreviations

Conclusion Understanding the molecular mechanisms underlying signal transduction is imperative to advance our knowledge against increasing drug resistant phenotypes in bacteria. In silico structural elucidation of quorum sensing regulator, LuxO in V. cholerae reveals structural variation at conserved architecture. Although LuxU is similar in predicted conformation to other known phosphotransfer proteins, histidine residues falling at the terminal helix require biochemical investigation to establish function. Autoinducers acting through quorum sensing components are known to repress the expression of virulence factors in V. cholerae. The existing model of quorum sensing regulated virulence gene expression suggests that high autoinducer concentration causes repression of pathogenesis through inactivation of LuxO (Camara, Hardman, Williams, & Milton, 2002; Miller et al., 2002). Targeted disruption of positions involved in signal propagation and activation of LuxO could result in probable loss of function in these quorum sensors. Apart from structural description of LuxU, elaboration of binding sites for ATP and phosphorylation site in LuxO provides us with an opportunity to utilize binding pockets in targeted inactivation of these response regulators, and thus repression of virulence. Recent explorations of bacterial pathogenesis enlighten the importance of targeting quorum

nm ps MDS

nano meters pico second molecular dynamics simulations

Acknowledgments This work in part was supported by the funds contributed by the Department of Biotechnology, New Delhi, to the Institute of Life Sciences. Senior Research Fellowship awarded by the Indian Council of Medical Research, New Delhi, India, to M. H.U. Turabe Fazil is gratefully acknowledged.

Supplementary material The supplementary material for this paper is available online at http://dx.doi.10.1080/07391102.2012.687523. References Alam, M., Sultana, M., Nair, G.B., Siddique, A.K., Hasan, N. A., Sack, R.B., … Colwell, R.R. (2007). Viable but nonculturable Vibrio cholerae O1 in biofilms in the aquatic environment and their role in cholera transmission. Proceedings of the National Academy of Sciences of the USA, 104, 17801–17806. Altschul, S.F., Gish, W., Miller, W., Myers, E.W., & Lipman, D.J. (1990). Basic local alignment search tool. Journal of Molecular Biology, 215, 403–410.

Downloaded by [Institute of Life Sciences] at 04:38 02 July 2012

10

M.H. Fazil et al.

Bachhawat, P., Swapna, G.V.T., Montelione, G.T., & Stock, A. M. (2005). Mechanism of activation for transcription factor PhoB suggested by different modes of dimerization in the inactive and active states. Structure, 13, 1353–1363. Berman, H.M., Westbrook, J., Feng, Z., Gilliland, G., Bhat, T. N., Weissig, H., … Bourne, P.E. (2000). The protein data bank. Nucleic Acids Research, 28, 235–242. Bowers, K.J., Chow, E., Xu, H., Dror, R.O., Eastwood, M.P., Gregersen, B.A., … Shaw, D.E. (2006). Scalable algorithms for molecular dynamics simulations on commodity clusters. In SC ‘06 Proceedings of the 2006 ACM/IEEE conference on Supercomputing, 11–17 November, Tampa, FL, USA. New York: ACM Press. Buehler, M.J., & Yung, Y.C. (2009). Deformation and failure of protein materials in physiologically extreme conditions and disease. Nature Materials, 8, 175–188. Camara, M., Hardman, A., Williams, P., & Milton, D. (2002). Quorum sensing in Vibrio cholerae. Nature Genetics, 32, 217–2181. Chen, X., Schauder, S., Potier, N., Van Dorsselaer, A., Pelczer, I., Bassler, B.L., & Hughson, F.M. (2002). Structural identification of a bacterial quorum-sensing signal containing boron. Nature, 415, 545–549. Colwell, R.R., Kaper, J., & Joseph, J.W. (1977). Vibrio cholerae, Vibrio parahaemolyticus and other vibrios: Occurrence and distribution in Chesapeake Bay. Science, 198, 394–396. Eisenberg, D., Luthy, R., & Bowie, J.U. (1997). VERIFY3D: Assessment of protein models with three-dimensional profiles. Methods in Enzymology, 277, 396–404. Essmann, U., Perera, L., Berkowitz, M., Darden, T., Lee, H., & Pedersen, L.J. (1995). Smooth particles mesh Ewald method. Journal of Chemical Physics. doi:101063/1470117 Ewing, T.J., Makino, S., Skillman, A.G., & Kuntz, I.D. (2001). DOCK 4.0: Search strategies for automated molecular docking of flexible molecule database. Journal of Computer-Aided Molecular Design, 15, 411–428. Freeman, J.A., & Bassler, B.L. (1999). A genetic analysis of the function of LuxO, a two-component response regulator involved in quorum sensing in Vibrio harveyi. Molecular Microbiology, 31, 665–677. Gao, R., Mack, T.R., & Stock, A.M. (2007). Bacterial response regulators: Versatile regulatory strategies from common domains. Trends in Biochemical Sciences, 32, 225–234. Guo, Z., Mohanty, U., Noehre, J., Sawyer, T.K., Sherman, W., & Krilov, G. (2010). Probing the α-helical structural stability of stapled p53 peptides: Molecular dynamics simulations and analysis. Chemical Biology & Drug Design, 75, 348–359. Hagler, A.T., Huler, E., & Lifson, S. (1974). Energy functions for peptides and proteins: I derivation of a consistent force field including the Hydrogen bond from amide crystals. Journal of the American Chemical Society, 96, 5319–5327. Hammer, B.K., & Bassler, B.L. (2003). Quorum sensing controls biofilm formation in Vibrio cholerae. Molecular Microbiology, 50, 101–104. Heing, M., & Frishman, D. (2004). STRIDE: A web server for secondary structure assignment from known atomic coordinates of proteins. Nucleic Acid Research, 32, W500–W502. Jobling, M.G., & Holmes, R.K. (1997). Characterization of hapR a positive regulator of the Vibrio cholerae HA/protease gene hap and its identification as a functional homologue of the Vibrio harveyi luxR gene. Molecular Microbiology, 26, 1023–1034.

Jones, G., Willett, P., Glen, R.C., Leach, A.R., & Taylor, R. (1997). Development and validation of a genetic algorithm for flexible docking. Journal of Molecular Biology, 267, 727–748. Kaminski, G.A., Friesner, R.A., Tirado-Rives, J., & Jorgensen, W.L. (2001). Evaluation and reparametrization of the OPLS-AA force field for proteins via comparison with accurate quantum chemical calculations on peptides. Journal of Physical Chemistry B, 105, 6474–6487. Laskoswki, R.A., MacArthur, M.W., Moss, D.S., & Thornton, J.M. (1993). PROCHECK: A program to check the stereochemical quality of protein structures. Journal of Applied Crystallography, 26, 283–291. Lee, S.Y., De La Torre, A., Yan, D., Kustu, S., Nixon, B.T., & Wemmer, D.E. (2003). Regulation of the transcriptional activator NtrC1: Structural studies of the regulatory and AAA+ ATPase domains. Genes & Development, 17, 2552– 2563. Lenz, D.H., Mok, K.C., Lilley, B.N., Kulkarni, R.V., Wingreen, N.S., & Bassler, B.L. (2004). The small RNA chaperone Hfq and multiple small RNAs control quorum sensing in Vibrio harveyi and Vibrio cholerae. Cell, 118, 69–82. Lilley, B.N., & Bassler, B.L. (2000). Regulation of quorum sensing in Vibrio harveyi by LuxO and sigma-54. Molecular Microbiology, 36, 940–954. Maestro version 8.5. (2008). New York, NY: Schrodinger Inc. Marchler-Bauer, A., Lu, S., Anderson, J.B., Chitsaz, F., Derbyshire, M.K., Deweese-Scott, C., … Bryant, S.H. (2011). CDD: A conserved domain database for the functional annotation of proteins. Nucleic Acids Research, (Database issue), 39, D225–D229. Martyna, G.J., Klein, M.L., & Tuckerman, M. (1992). Nosé– Hoover chains: The canonical ensemble via continuous dynamics. Journal of Chemical Physics, 97, 2635–2643. Martyna, G.J., Tobias, D.J., & Klein, M.L. (1994). Constant pressure molecular dynamics algorithms. Journal of Chemical Physics, 101, 4177–4189. Miller, M.B., Skorupski, K., Lenz, D.H., Taylor, R.K., & Bassler, B.L. (2002). Parallel quorum sensing systems converge to regulate virulence in Vibrio cholerae. Cell, 110, 303– 314. Neuwald, A.F., Aravind, L., Spouge, J.L., & Koonin, E.V. (1999). AAA+: A class of chaperone-like ATPases associated with the assembly operation and disassembly of protein complexes. Genome Research, 9, 27–43. Ramamurthy, T. (2008). Antibiotic resistance in Vibrio cholerae. In S.M. Faruque & G.B. Nair (Eds.), Vibrio cholerae: Genomics and molecular biology (pp. 191–207). London: Caister Academic Press. Rasko, D.A., Moreira, C.G., Li de, R., Reading, N.C., Ritchie, J.M., Waldor, M.K., … Sperandio, V. (2008). Targeting QseC signaling and virulence for antibiotic development. Science, 321, 1078–1080. Raychaudhuri, S., Jain, V., & Dongre, M. (2006). Identification of a constitutively active variant of LuxO that affects production of HA/protease and biofilm development in a nonO1, non-O139 Vibrio cholerae O110. Gene, 369, 126–133. Sali, A., & Blundell, T.L. (1993). Comparative protein modeling by satisfaction of spatial restraints. Journal of Molecular Biology, 234, 779–815. Sallai, L., & Tucker, P.A. (2005). Crystal structure of the central and C-terminal domain of the sigma (54)-activator ZraR. Journal of Structural Biology, 151, 160–170.

Downloaded by [Institute of Life Sciences] at 04:38 02 July 2012

Two Proteins Belonging to Quorum Sensing System Shindyalov, I.N., & Bourne, P.E. (1998). Protein structure alignment by incremental combinatorial extension (CE) of the optimal path. Protein Engineering, 11, 739–747. Shivakumar, D., Williams, J., Wu, Y., Damm, W., Shelley, J., & Sherman, W. (2010). Prediction of absolute solvation free energies using molecular dynamics free energy perturbation and the OPLS force field. Journal of Chemical Theory and Computation, 6, 1509–1519. Singh, D.V., Matte, M.H., Matte, G.R., Jiang, S., Sabeena, F., Shukla, B.N., … Colwell, R.R. (2001). Molecular analysis of Vibrio cholerae O1, O139, non-O1 and non-O139 strains: Clonal relationships between clinical and environmental isolates. Applied and Environment Microbiology, 67, 910–921. Stock, A.M., Robinson, V.L., & Goudreau, P.N. (2000). Twocomponent signal transduction. Annual Review of Biochemistry, 69, 183–215. Tetko, I.V., Gasteiger, J., Todeschini, R., Mauri, A., Livingstone, D., Ertl, P., … Prokopenko, V.V. (2005). Virtual computational chemistry laboratory – design and description. Journal of Computer-Aided Molecular Design, 19, 453–463. Thompson, J.D., Higgins, D.G., & Gibson, T.J. (1994). CLUSTALW: Improving the sensitivity of progressive multiple sequence alignment through sequence weighting positionspecific gap penalties and weight matrix choice. Nucleic Acids Research, 22, 4673–4680. Tuckerman, M., Berne, B.J., & Martyna, G.J. (1992). Reversible multiple time scale molecular dynamics. Journal of Chemical Physics, 97, 1990–2001.

11

Ulrich, D.L., Kojetin, D., Bassler, B.L., Cavanagh, J., & Loria, J.P. (2005). Solution structure and dynamics of LuxU from Vibrio harveyi a phospho-transferase protein involved in bacterial quorum sensing. Journal of Molecular Biology, 347, 297–307. Vance, R.E., Zhu, J., & Mekalanos, J.J. (2003). A constitutively active variant of the quorum-sensing regulator LuxO affects protease production and biofilm formation in Vibrio cholerae. Infection and Immunity, 71, 2571–2576. Varughese, K.I., Madhusudan, X.Z., Whiteley, J.M., & Hoch, J. A. (1998). Formation of a novel four-helix bundle and molecular recognition sites by dimerization of a response regulator phosphotransferase. Molecular Cell, 2, 485–493. Wallace, A.C., Laskowski, R.A., & Thornton, J.M. (1995). LIGPLOT: A program to generate schematic diagrams of protein–ligand interaction. Protein Engineering, 8, 127–134. Wang, R., Lai, L., & Wang, S. (2002). Further development and validation of empirical scoring functions for structure – based binding affinity prediction. Journal of ComputerAided Molecular Design, 16, 11–26. Wiederstein, M., & Sippl, M.J. (2007). ProSA-web: Interactive web service for the recognition of errors in three-dimensional structures of proteins. Nucleic Acids Research, 35, 407–410. Zhu, J., Miller, M.B., Vance, R.E., Dziejman, M., Bassler, B. L., & Mekalanos, J.J. (2002). Quorum-sensing regulators control virulence gene expression in Vibrio cholerae. Proceedings of the National Academy of Sciences of the USA, 99, 3129–3134.