Abstract: The uncharacterized alphasatellite of begomovirus associated with a common weed Verbesina encelioides was characterized by using molecular and in silico tools and techniques. Verbesina encelioides leaf curl alphasatellite (HQ631431) shows 87 % nucleotide sequence identity with Sida yellow vein disease associated DNA1 (FN806782). Translated 3-frame of HQ631431 (HQ631431/3-f.pdb) and its homologous 2HWT.pdb had respectively 67 and 77.2% residues in most favorable region of its Ramachandran plot therefore both of these two models cannot be placed in good quality. The 3d2GO server showed the hydrolase activity as possible function for HQ631431/3-f.pdb with 0.58% confidence level. In silico prediction of results can be used to confirm the begomovirus not only in host weeds but also in other crops also.
INTRODUCTION
Medicinal important weeds and other economical important plants can be affected by different type of diseases that shows a wide range of symptoms and the causative agents of these diseases are of biotic or abiotic nature (Simsek et al., 2002). Among the biotic disease agents, viruses can attack all types of plants. Viruses are infectious pathogens that are too small to be seen with a high power microscope and can replicate only inside the living cells of organisms (Koonin et al., 2006).
Geminiviruses are plant viruses that belong to the family Geminiviridae (Goodman, 1977a, b; Fauquet et al., 2005) and characterized by the unique Gemini shape of a fused icosahedral viral particle. Geminiviruses have circular single stranded DNA genomes encapsulated in twinned quasi isometric particles and are responsible for major croplosses worldwide (Moffat, 1999). The family Geminiviridae is divided in to four genera (Mastrevirus, Curtovirus, Topocuvirus and Begomovirus), based on genome structure, type of insect vector and host range (Stanley et al., 2005; Medina-Ramos et al., 2008). Begomovirus is the largest genus of family Geminiviridae (Dhakar et al., 2010) and obligatory transmitted by the whitefly Bemisia tabaci (Funayama et al., 2001; Sidhu et al., 2009; Govindappa et al., 2011) that are common in the tropical and subtropical regions of the world.
The geminivirus genome is organized in one (monopartite) or two (bipartite) covalently closed, circular, ssDNA molecules of about 2.5-2.9 kb. The genes in monopartite and bipartite geminiviruses are arranged in two divergent clusters 280 to 350 nucleotides, each separated by the Intergenic Region (IR) (Lazarowitz and Shepherd, 1992). The single genomic component of monopartite geminiviruses (Mastreviruses and Curtoviruses) contains all the information necessary for virus replication and infectivity (Hanley-Bowdoin et al., 1996). Bipartite Begomoviruses have seven genes distributed in the two genomic components (Fig. 1) designated A and B (Tiendrebeogo et al., 2008). The A component contains genes involved in virus replication and encapsidation and the B component contains the genes involved in virus movement. The A and B components each have a common region which consists of a block of approximately 200 bp within the IR (Sunter and Bisaro, 1991). These common regions are identical in a given bipartite begomovirus but completely different in among the other geminivirus with the small exception of 30 nucleotide stem loop (identified as origin of replication) region. The common region also contains two divergent promoters which differentially regulate the temporal expression of the viral genes.
The Alphasatellite components are satellite-like, circular ssDNA molecules of approximately 1375 nucleotides in length (Fig. 2). They encode a single gene, a rolling circle replication initiator protein and are capable of autonomous replication in plant cells. Closely related to the replication associated protein encoding components of nanoviruses (a second family of plant infecting ssDNA viruses) (Vetten et al., 2004), from which they are believed to have evolved, they require a helper Begomovirus for movement within and between plants (Mansoor et al., 1999; Saunders and Stanley, 1999). Interestingly, alphasatellite components are phenotypically silent; playing no part in the symptoms of the complex and their precise function remains unclear (Tiendrebeogo et al., 2010).
Verbesina encelioides (Astraceae) is an erect annual commonly seen weed up to heights of 1 to 5 feet and distributed not only in India but also several regions of the world (Ball et al., 1951; Robbins et al., 1951; Sindhu et al., 2010). Leaves of Verbesina encelioides are toothed or lobed (Everist, 1957) and have two distinct growth patterns: lower leaves are opposite and triangular, while the upper leaves are alternate and lanceshaped (Wagner et al., 1990). Both upper and lower leaves feature has fine white hairs on the underside (Everist, 1957). These fine white hairs are also present on the stem of Verbesina encelioides which grows from a taproot system (Parker, 1972).
Begomovirus associated symptoms were observed in several Verbesina encelioides growing as common weed in Rajasthan (India). The symptoms of the disease consist of mosaic and curling of leaf, leaf yellowing and stunting of plants. Begomovirus was suspected as a causative pathogen because of observed large population of whitefly (Bemisia tabaci, the vector of Begomovirus (Geminiviridae)) on the plants. DNA isolated from leaves with symptoms and apparently without symptoms by using CTAB (Cetyl Trimethyl Ammonium Bromide) method and subjected to Rolling Circular Amplification (RCA)which uses a high fidelity ø 29 DNA polymerase along with random hexamers to detect the genomes of various Begomoviruses.
| Fig. 1: | Genomic Organization of a Begomovirus distributed in the two genomic components designated A and B |
| Fig. 2: | Structure of alphasatellite component, encodes a single gene in the positive orientation (a replication initiator protein (Rep)) and a region of sequence rich in adenine (a-rich) |
Restriction analysis was performed on positive RCA products in order to detect mixed Begomovirus infections. The digested product was subjected to cloning and partial sequenced by using M13 universal primer and sequence information was submitted in NCBI-bankIt (GenBank accession HQ631431).
At the first stage sequence analysis begin with sequence alignment by using any one of the BLAST programs (BLASTn, BLASTp etc.). After sequence alignment different bioinformatics tools and techniques can be used for homology modeling procedure which can be divided into four sequential steps: template selection, target template alignment, model construction and model assessment (Marti-Renom et al., 2000). Homology modeling is a helpful tool for the production of 3D structure of protein by using in silico methods and by this approach we can overcome the limitations of X-Ray crystallography and NMR. The selected model can be validated by using protein structure checking tools such as Prochek for reliability and 3d2GO for function prediction server.
Bioinformatics tools and techniques based sequence analysis has application of research and development in the field of genomics, transcriptomics and proteomics at gene expression, structural and function production level. The in silico drug designing principles can be apply for in silico antiviral agent designing and that is the new path of research in the field of plant virology. The aim of the study was, characterization of Verbesina encelioides leaf curl alphasatellite (HQ631431) by using molecular (e.g., DNA isolation, Rolling Circular Amplification [RCA], cloning, sequencing) and in silico (e.g., NCBI-bankIt sequence submission, Homology modeling and function prediction by using 3d2GO) techniques.
MATERIALS AND METHODS
Samples collection and DNA extraction: Due to the current realization that weed-infecting begomoviruses may infect crops and those weeds may serve as a reservoir for crop-infecting geminiviruses (Roye et al., 1997) we made an extensive survey for the epidemelogy of begomovirus in 2009-2010. We collected a begomovirus associated with a common weed namely Veresina encelioides across Rajasthan, India. To investigate the possibility the leaf samples from Verbesina sp. exhibiting typical begomovirus symptoms, including yellow and golden mosaic, chlorotic mottling and blistering, were taken for the study. Total DNA from plant samples were extracted by using a CTAB (Cetyl Trimethyl Ammonium Bromide) method (Manen et al., 2005).
Rolling Circular Amplification (RCA): Isolated DNA was subjected to Rolling Circular Amplification (RCA)which uses a high fidelity ø 29 DNA polymerase along with random hexamers to detect the genomes of various Begomoviruses. Genome concatemers generated during amplification were digested with restriction enzymes to release unit-length genomes. A preliminary screening was carried out to identify enzymes capable of linearizing amplification products. After digestion, genomic DNA was gel-purified and cloned into pBS KS (Stratagene).
Restriction analysis was performed on positive RCA products in order to detect mixed Begomovirus infections. Each sample was cleaved with restriction enzymes: EcoRI and SalI essentially. The digested product was taken for cloning and partial sequencing by using M13 universal primer. After sequencing the nucleotide sequence was submitted in NCBI-bankIt (GenBank accession HQ631431).
Retrieval of sequences from NCBI translation: Published information about complete nucleotide sequence of Verbesina encelioides leaf curl alphasatellite (>gi|319428633|gb|HQ631431.1) was retrieved from GenBank-NCBI (http://www.ncbi.nlm.nih.gov/) in FASTA format and was used for BLASTn alignment.
Nucleotide sequence of Verbesina encelioides leaf curl alphasatellite (HQ631431) was translated into protein sequence by using DNA to protein translation server (http://insilico.ehu.es/translate/) (Fig. 3).
Phylogenetic tree construction: Phylogenetic relationship of HQ631431 was determined based on multiple sequence alignment by using the CLC Main Workbench 5.7. The CLC Main Workbench is developed for Windows, MacOSX and Linux. The software for either platform can be obtained from http://www.clcbio.com. The Neighbor Joining algorithms (Saitou and Nei, 1987) were used for construction of phylogenetic trees. Bootstrap values were computed using 100 replicates to evaluate support for the groupings. This analysis clustered each one of the isolates with the other previously sequenced isolate of the respective species.
| Fig. 3: | (a) Different reading frames of complete sequence of Verbesina encelioides leaf curl alphasatellite (HQ631431), created by using DNA to protein translation server (http://insilico.ehu.es/translate/) that have 132 amino acids residues. (b) Amino acid sequence of 3-frame of HQ631431 (HQ631431/3-f) |
Selection of template: The homology modeling requires a query sequence with an unknown 3D structure and the target sequence that have known 3D structure with at least 35% similarity. BLAST (Basic Local Alignment Search Tool) (BLASTp 2.2.24+) was used (Altschul et al., 1997, 2005) to search against the PDB (Protein Databank) to find out the related homologues of the query/template (3-frame of HQ631431 or HQ631431/3-f) sequence (http://blast.ncbi.nlm.nih.gov/Blast.cgi). By the BLASTp search of translated protein HQ631431/3-f against Protein Data Bank (PDB), the Master-Rep protein nuclease domain (2-95) from the Faba Bean Necrotic yellows virus (2HWT) showed highest sequence identity of 40%, positives 64% and gaps were 5%.
The PDB file of the target sequence of 2HWT was downloaded from PDB (http://www.rcsb.org/pdb) and the FASTA format of target sequence (PDB: 2HWT) was mined from GenBank-NCBI as follows:
The 3-frame of HQ631431 (HQ631431/3-f) was find suitable for sufficiently close template in structure database to confidently model of the sequence and rest of translated frames were not find suitable.
Those FASTA sequences of template 3-frame of HQ631431 (HQ631431/3-f) and target (2HWT) were uploaded on the 3D-JIGSAWN (Protein Comparative Modeling Server) for the construction of its PDB files. 3D-JIGSAWN (http://bmm.cancerresearchuk.org/~3djigsaw/) sends the PDB files on the e-mail address that assigned to the modeling server. The PDB file of query and homologous target sequence were further utilized for 3D model energy validation and docking studies (Heinrichs, 2008). The predicted model was validated with the program Procheck and Ramachandran plot statistics was used to evaluate the stability of the model.
Model evaluation and validation: The UCLA-DOE (http://nihserver.mbi.ucla.edu/Verify-3D/) Structure Evaluation server provide a visual analysis of the quality of a putative crystal structure for protein. Verify 3D expects this crystal structure to be submitted in PDB format. The validation for structure models obtained from the three software tools was performed by using PROCHECK (Laskowski et al., 1996). In order to study the energy validation of query and homologous target proteins, we uploaded PDB files of both proteins on structure analysis and validation server (SAVA, http://nihserver.mbi.ucla.edu/SAVES/). SAVA service provides by NIH-MBI laboratory for structural genomics and proteomics (Bowie et al., 1991; Luthy et al., 1992) analysis. In this study the model was checked with Verified-3D (Fig. 4) and Ramachandran plot through Procheck (Fig. 7) server (Prajapat et al., 2011).
PDB file of query (HQ631431/3-f.pdb) utilized for the ribbon representation model construction using 3dLigandSite server Jmol (Fig. 4a) and homologous target protein (2HWT) NMR solution structure that was uploaded from Protein Data Bank (Fig. 4b).
| Fig. 4: | (a) Ribbon representation of model of 3-frame protein sequence of HQ631431/3-f.pdb created by using 3dLigandSite server Jmol. α-helices (Pink) and β-sheets (Yellow) are shown as helices and ribbons (b) 2HWTNMR solution structure of the Master-Rep protein nuclease domain (2-95) from the Faba Bean Necrotic Yellows Virus that uploaded from PDB, as best homologous HQ631431/3-f. α-helices (Green) and β-sheets (Red, Green, Blue and Sky color) are shown as helices and ribbons. |
Function prediction: The 3d2GO server (http://www.sbg.bio.ic.ac.uk/phyre/pfd/html/help.html) was used for the prediction of functions of the predicted model using sequence and structure in the reference of Gene Ontology (GO). 3D2GO predicts the function of protein using some information e.g., overall topological similarity to structures with known function, geometric and residue similarity of predicted functional sites to regions of known structures and sequence homology to functionally annotated sequences. The MAMMOTH structural alignment program was used for full topology search of the model (Ortiz et al., 2002). The MUSCLE program was used for functional site prediction of the predicted model (Edgar, 2004). Functional residue prediction was done using the Jenson-Shannon Divergence (JS Divergence), an information-theory approach to determine relative residue conservation (Capra and Singh, 2007) such conservation is related to the functional importance of residues.
3D-ligand site prediction: The Protein ligand binding residues was predicted by using program 3dLigandSite (http://www.sbg.bio.ic.ac.uk/3dligandsite/) which uses Critical Assessment of protein Structure Prediction experiment (CASP) (Wass and Sternberg, 2009). This was based on the approach to identify binding sites by combining the use of the predicted structure of the targets with both residue conservation and the location of ligands bound to homologues structures.
RESULTS AND DISCUSSION
Veresina encelioides symptoms are of bright yellow spots finds along the midrib which coalesce to give a mosaic, a reduction in leaf size and stunting of the plant. Field survey showed that Veresina encelioides is very common in Rajasthan state (India), it grow in water rich places of the field and show competition with economical important crop plants for nutrition and water and may serve as a reservoir for crop-infecting geminiviruses. Overall results of molecular analysis indicate that there are high infection rate of Begomovirus persisting with yellow mosaic disease of Verbesina encelioides in Rajasthan (India).
Positive RCA products were restriction analyzed in order to detect mixed begomovirus infections and each sample was cleaved with restriction enzymes: EcoRI and SalI essentially. All the samples digested with EcoRI and SalI in RCA, a full-length fragment at approximately 1.3 kb was found.
Complete sequence of Verbesina encelioides leaf curl alphasatellite (HQ631431) was used as query for template search by using BLASTn program. The alignment of HQ631431 by using BLASTn revealed the highest 85% nucleotide sequence identities (Query coverage 100%) with Cyamopsis tetragonoloba leaf curl alphasatellite (GU385877),86% (Query coverage 92%) with Tobacco leaf curl PUSA alphasatellite(HQ180392) also 87% (Query coverage 87%) with Sida yellow vein disease associated DNA 1 (FN806782).
Phylogenetic relationship were determined based on multiple sequence alignment of complete nucleotide sequence of Verbesina encelioides leaf curl alphasatellite (HQ631431), performed using the CLC Main Workbench 5.7. Bootstrap values were computed using 100 replicates to evolution support for the groupings. The Verbesina encelioides leaf curl alphasatellite (HQ631431) placed in monophyletic clusters with 100 bootstrap value of Cyamopsis tetragonoloba leaf curl alphasatellite (GU385877) in Neighbor-Joining tree (Fig. 5).
Complete sequence of Verbesina encelioides leaf curl alphasatellite (HQ631431) was converted into 6 different reading frames through DNA to protein translation server (http://insilico.ehu.es/ translate/). All these frames protein sequences was uploaded on 3D-JIGSAW protein comparative modeling server (bmm.cancerresearchuk.org/~3djigsaw/). Sufficient close template was generated only for first 3-reading protein sequence frame of HQ631431 (HQ631431/3-f), to confidently model this sequence and find unsuccessful with rest of other reading frames.
During BLASTp analysis of query sequence (3-frame protein sequence of HQ631431 or HQ631431/3-f) against PDB (http://ww.rcsb.org) showed 40% identity with Chain A of master rep protein nuclease domain of the Faba Bean Necrotic Yellows Virus (2HWT).The PDB file of 2HWT was uploaded from Protein Data Bank (http://ww.rcsb.org) for further analysis.
Further Verify-3D and Procheck (Laskowski et al., 1993) was used to perform full geometric analysis as well as stereochemical quality of a protein structure by analyzing residue-by-residue geometry and overall structure geometry.
Verify-3D (Eisenberg et al., 1997) works best on proteins with at least 100 residues. Verify3D analyzes the compatibility of an atomic model (3D) with its own amino acid sequence.
The Profile score above zero in the Verify 3D graph corresponds to acceptable environment of the model (Mau Goh et al., 2008). The high score of 0.29 indicate that environment profile of the model is good (Fig. 6).
Homology modeling study is an important method to know the 3D structure of the protein whose structure is not available. After running Procheck, Ramachandran plot of query protein sequence (3-frame of HQ631431 or HQ631431/3-f.pdb) had only 67% residues were in core/favored region, 28.4% in the additional allowed region, 4.5% in the generously allowed region and 0% of the residues in the disallowed region (Fig. 7a) which made this model more acceptable as compared to other predicted models (Table 1). Similar approach was also used in the prediction of homologous 2HWT. Ramachandran plot of 2HWT.pdb had 77.2% in core, 19.0% in allowed, 2.5% in gener and 1.3% disallowed regions (Fig. 7b).
A good quality Ramachandran plot have over 90% in the most favoured or core regions (Prajapat et al., 2010). Ramachandran plot of HQ631431/3-f.pdb had 67% residues in the most favoured region and 2HWT.pdb has 77.2% residues in the most favoured (Table 1), therefore, both of these models cannot place in good quality.
| Fig. 5: | A Neighbor-Joining tree based on complete nucleotide sequence of Verbesina encelioides leaf curl alphasatellite (HQ631431)and other begomovirus sequences available in GenBank. Bootstrap values at major nodes are indicated. Horizontal distances are proportional to the genetic distance between isolates and vertical distances are arbitrary. Scale bar indicates the proportion of sites changing along each branch. The Verbesina encelioides leaf curl alphasatellite (HQ631431) used for comparison: Nanovirus-like particle rep gene (AM930246), Nanovirus-like particle rep gene (AM930247), Nanovirus-like particle, rep gene (AM884370), Cotton leaf curl Burewala alphasatellite (GU992936), Cotton leaf curl Burewala alphasatellite (HM004548), Cotton leaf curl Burewala alphasatellite (FN658728), Nanovirus-like particle rep gene (AJ512956), Nanovirus-like particle rep gene (AJ132345), Nanovirus-like particle rep gene (AJ132344), Nanovirus-like particle rep gene (AJ512955), Cyamopsis tetragonoloba leaf curl alphasatellite (GU385877), Nanovirus-like particle rep gene (AJ512961), Nanovirus-like particle partial rep gene (AJ512962), Nanovirus-like particle partial rep gene (AJ512953), Tobacco leaf curl Yunnan virus (AJ888455), Tobacco curly shoot virus associated DNA 1 rep gene (AJ579351), Tobacco curly shoot virus associated DNA 1 rep gene (FN678900), Tobacco curly shoot virus associated DNA 1 rep gene (AJ579349), Tobacco curly shoot virus associated DNA 1 rep gene (FN678901), Ageratum enation virus associated DNA 1 (FN543100), Nanovirus-like particle (FN794199), Nanovirus-like particle (FN794202), Nanovirus-like particle rep gene (AJ512951), Nanovirus-like particle rep gene (AM930244) |
Protein 2HWT.pdb is more stable then HQ631431/3-f.pdb due to high percentage in core/favored region and less percentage of residues in allowed region and generously allowed region of Ramachandran plot. Phe 6, Gln 49, Glu 47 and Asp 55 present in the generously allowed region, therefore, they may affect the stability of protein (Table 2).
| Fig. 6: | Verified 3D graph/Profile search plot of HQ631431/3-f.pdb (No. of Residues = 104) with score of 0. 29 |
| Fig. 7: | Ramachandran plot of HQ631431/3-f.pdb and its homologous 2HWT.pdb, model predicted using PROCHECK |
| Table 1: | Comparative analysis of Ramachandran statistics in the HQ631431/3-f.pdb and homologous 2HWT.pdb predicted models |
| Table 2: | Residues in generously allowed region that affect the stability of protein |
| Table 3: | Result showing the function prediction of the query modeled protein with 3d2GO (Protein function prediction server) |
| Table 4: | Different ligand clusters information shows that Cluster 1 has ligands and structures (1 each) with the average Mammoth |
| Table 5: | List of amino acid residues observed in cluster 1 of predicted protein with number of contacts of ligand, average distance and JS divergence |
The 3d2GO server uses several methods of function prediction, using sequence and structure, to predict Gene Ontology (GO) terms for the protein (Shankaracharya et al., 2011). Various GO terms, their description and the confidence have been listed in Table 3. Confidence ranges from 0 to 1, with 1 being the most confident prediction. Result show that the predicted query protein HQ631431/3-f.pdb file has functions like hydrolase activity and cation binding with good confidence (Table 3) which confirms its functions.
Hence, the result shows that Tyr60, Gln21, Lys63, Cys61, Tyr23 and Ala57 were more conserved residue in the structure. The predicted site one had Gln21, His45, Tyr23, Ala57, Leu20, Gly22 and in the predicted site two have Tyr60, Gln21, Ala57, Met62, Lys48, Lys53.
Different ligand clusters information shows that Cluster 1 has ligands and structures (1 each) with the average MAMMOTH score (Shankaracharya et al., 2010) of 7.1 (Table 4). Phe74, Gly75 and Glu76 List observed in cluster 1 of predicted protein and there number of contacts, average distance and JS divergence shown in the Table 5.
Prediction of protein 3D structure helps us to find out their active sites, binding site (Amir et al., 2010) etc. The model developed through homology modeling and subsequently the predicted functional characteristics of 3-frame of HQ631431 (HQ631431/3-f) provide a base for structural and functional characteristics of Verbesina encelioides leaf curl alphasatellite (HQ631431/3-f) and suitable role in leaf curl mechanism therefore, provide better control on begomovirus infection not only for Verbesina encelioides but also for other plants species infected by begomovirus in various tropical and subtropical part of the world.
CONCLUSION
Verbesina encelioides (family Asteraceae) is a common weed plant in the fields of Rajasthan, India. In recent years, large numbers of plants were exhibiting typical Begomovirus infection symptoms. Isolated DNA from leaf samples was subjected to Rolling Circular Amplification (RCA)to detect the genomes of various Begomoviruses. All samples were digested with EcoRI and SalI and a full-length fragment of size approximately 1.3 kb was obtained. RCA product was sequenced and submitted to GenBank-NCBI (accession HQ631431). Verbesina encelioides leaf curl alphasatellite (HQ631431) sequence alignment was done by using BLASTn that revealed about 87% nucleotide sequence identity with Sida yellow vein disease associated DNA 1 (FN806782). HQ631431 placed in monophyletic clusters with 100 bootstrap value of Cyamopsis tetragonoloba leaf curl alphasatellite (GU385877) in Neighbor-Joining tree.
Homology modeling and function prediction study of translated 3-frame of HQ631431 (HQ631431/3-f) was performed, thus shown highest homology with 2HWT. HQ631431/3-f.pdb had 67% residues and 2HWT.pdb had 77.2% residues in the most favoured region of its Ramachandran plot, therefore, both of these models cannot place in good quality. The hydrolase activity, cation binding and catalytic activity were predicted for HQ631431/3-f as important functional site of the model with high confidence through 3d2GO (Protein function prediction server). Tyr60, Gln21, Lys63, Cys61, Tyr23 and Ala57 were found as more conserved residue in the structure. According to our best knowledge it is the first report regarding functional characteristics of Verbesina encelioides leaf curl alphasatellite (HQ631431/3-f), that can be used in future for designing in silico antibegomoviral agent by using docking tools and techniques. Results of this study will be contribute in control begomoviral epidemic level and reduce yield loss in crops of tropical and subtropical regions of the world.
ACKNOWLEDGMENT
The authors are thankful to Prof. Shakti Baijal, Dean, FASC, MITS, Rajasthan, India. The authors are also thankful to Department of Biotechnology (DBT), India and Department of Science and Technology, India for financial support for the present studies.