<%BANNER%>

Detection, Characterization, and Distribution of Begomoviruses Infecting Tomatoes in Venezuela


PAGE 1

DETECTION, CHARACTERIZATION, AND DISTRIBUTION OF BEGOMOVIRUSES INFECTING TOMATOES IN VENEZUELA By ALBA RUTH NAVA A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2003

PAGE 2

Copyright 2003 by Alba Ruth Nava

PAGE 3

ACKNOWLEDGMENTS I thank Dr. Jane Polston (chairman of the supervisory committee) for her support and guidance during my graduate studies. I also thank Drs. Ernest Hiebert, Gail Wisler, Susan Webb and Maria Gallo-Meagher (who served on the supervisory committee) for their suggestions and contributions. Special thanks go to Christian Patte, Tracy Sherwood, and Kristin Beckham for their technical assistance and friendship. I thank my friends Yolanda, Denise, Abby, Zenaida, Maritza, Belkys, Francisco, Olimpia, Adriana, Marlene, Juliana, Gustavo, Jorge, Basma, Abdulwahid, Hamed, Vicente, Julia, Renato, Manjunath, Lucious, and Eugene for their support and constant encouragement during the last 4 years. I thank the Universidad del Zulia, for providing financial support to my studies. iii

PAGE 4

TABLE OF CONTENTS Page ACKNOWLEDGMENTS.................................................................................................iii LIST OF TABLES............................................................................................................vii LIST OF FIGURES...........................................................................................................ix ABSTRACT.......................................................................................................................xi CHAPTER 1 LITERATURE REVIEW.............................................................................................1 Begomovirus Description.............................................................................................1 Begomovirus Genome Structure and Viral-Encoded Proteins.....................................1 Criteria for Species Demarcation..................................................................................6 Impact of Begomoviruses on Agriculture.....................................................................6 Management of Begomoviruses...................................................................................8 Detection and Discrimination of Begomoviruses.......................................................10 Variability of Begomovirus Species...........................................................................10 Variability of African cassava mosaic virus and Tomato yellow mosaic virus in the Old World..................................................................................................11 Variability of Potato yellow mosaic virus in the New World.............................12 Begomoviruses in Tomato Crops in Venezuela.........................................................13 Background and Objectives........................................................................................14 2 DETECTION AND VARIABILITY OF BEGOMOVIRUSES IN TOMATO-PRODUCTION AREAS OF ANDEAN STATES, VENEZUELA...........................16 Introduction.................................................................................................................16 Materials and Methods...............................................................................................18 Survey..................................................................................................................18 DNA xtraction.....................................................................................................19 Polymerase Chain Reaction (PCR).....................................................................21 Restriction Enzyme Analysis of Purified PCR Products.....................................22 Cloning and DNA Sequence Determination.......................................................22 Comparison of DNA Sequence...........................................................................23 Phylogenetic Analysis.........................................................................................23 iv

PAGE 5

Results.........................................................................................................................24 Comparison of DNA Extraction Protocols..........................................................24 PCR Amplification of Begomovirus Sequences from Field Samples.................25 Restriction Enzyme Analysis of Purified PCR Product......................................28 DNA Sequence Comparison................................................................................30 Phylogenetic Analysis.........................................................................................35 Discussion...................................................................................................................38 3 CHARACTERIZATION OF A NEW BEGOMOVIRUS FROM THE STATE OF TRUJILLO, VENEZUELA........................................................................................43 Introduction.................................................................................................................43 Materials and Methods...............................................................................................44 Plant Sample and Extraction of Genomic DNA..................................................44 Obtaining Full-Length Sequences.......................................................................45 Obtaining Infectious Clones................................................................................47 Comparison of DNA and Amino Acid Sequences..............................................48 Phylogenetic Analysis.........................................................................................48 Results.........................................................................................................................51 Full-Length Sequencing......................................................................................51 Comparison of DNA and Amino Acid Sequences..............................................51 Phylogenetic analysis..........................................................................................53 Discussion...................................................................................................................54 4 CHARACTERIZATION OF A NEW BEGOMOVIRUS FROM THE STATE OF MERIDA, VENEZUELA...........................................................................................61 Introduction.................................................................................................................61 Materials and Methods...............................................................................................63 Plant Sample and Extraction of Genomic DNA..................................................63 Obtaining Full-Length Sequences.......................................................................63 Obtaining Full-Length Clones.............................................................................65 Comparison of DNA and Amino Acid Sequences..............................................67 Phylogenetic Analysis.........................................................................................67 Results.........................................................................................................................68 Full-Length Sequencing and Cloning..................................................................68 Comparison of DNA and Amino Acid Sequences..............................................70 Phylogenetic Analysis.........................................................................................71 Discussion...................................................................................................................72 5 DISTRIBUTION OF TWO BEGOMOVIRUSES IN TOMATO-PRODUCTION AREAS OF VENEZUELA........................................................................................79 Introduction.................................................................................................................79 Materials and Methods...............................................................................................81 Survey..................................................................................................................81 DNA Extraction, Begomovirus Detection by PCR, and Hybridization..............82 v

PAGE 6

PCR Probe...........................................................................................................83 Hybridization Conditions....................................................................................84 Confirmation of Hybridization Results by PCR and Restriction Analysis.........85 Southern Blot Analysis, Cloning and Sequence Determination..........................86 Results.........................................................................................................................87 Confirmation of Samples with Hybridization Signal to the 2.9-v Probe by PCR and Restriction Analysis..................................................................................87 Confirmation of Samples with Hybridization Signal to the 57-v Probe by PCR and Restriction Analysis..................................................................................87 Southern Blot Analysis, Sequence comparison...................................................88 Discussion...................................................................................................................89 6 CONCLUSIONS........................................................................................................96 LIST OF REFERENCES.................................................................................................100 BIOGRAPHICAL SKETCH...........................................................................................108 vi

PAGE 7

LIST OF TABLES Table page 2-1. Acronyms and accession numbers of known begomoviruses used in the phylogenetic analysis of partial sequences of samples from Andean states............24 2-2. Comparison of two DNA extraction protocols through PCR results of samples from Andean states, Venezuela using 4 primer sets to amplify partial sequence of A and B components of begomoviruses..............................................................26 2-3. Polymerase chain reaction results using degenerate primers to amplify A and B components of begomoviruses using Doyle and Doyle DNA extraction of Venezuelan samples from tomato-production areas................................................27 2-4. Expected restriction-fragment sizes (bp) in the fragment amplified by primer set PAR1c496/PAL1v1978 of the A component of known begomoviruses.................28 2-5. Restriction-fragment sizes in the fragment amplified by primer set PAR1c496/PAL1v1978 in the A component of begomoviruses from Andean states, in Venezuela..................................................................................................29 2-6. Expected restriction-fragment sizes in the fragment amplified by primer set PBL1v2040/PCRc154 of the B component of known begomoviruses....................30 2-7. Restriction-fragment sizes in the fragment amplified by PBL1v2040/PCRc154 of the B component of begomoviruses from Andean states, in Venezuela..................31 2-8. Nucleotide sequence identity (%) between partial sequences from PCR using primer set PAR1c496/PAL1v1978 of the A component of begomoviruses from Andean states and four known begomoviruses........................................................34 2-9. Motifs of the intergenic region and Iteron-related domain of Replication protein (Rep IRD) of cloned sequences from Andean states and eight known begomoviruses..........................................................................................................36 2-10. Nucleotide sequence identity (%) between partial sequences from PCR using primer set PBL1v2040/PCRc154 of the B component of begomoviruses from Andean states and four known begomoviruses........................................................37 3-1. Specific primers designed to determine the full-length sequence of a new begomovirus from Trujillo, Venezuela....................................................................47 vii

PAGE 8

3-2. Acronyms and accession numbers of known begomoviruses used for nucleic acid and amino acid sequence comparison and phylogenetic analysis of 2.9-v virus.....49 3-3. Percent of nucleic acid identity between 2.9-v virus sequence and 15 known begomoviruses..........................................................................................................52 3-4. Percent of similarity of amino acid sequences between proteins of 2.9-v virus and proteins of 15 known begomoviruses.......................................................................53 4-1. Specific primers designed to determine the full-length sequence of 57-v virus from Mrida, Venezuela...................................................................................................66 4 -2. Acronyms and accession numbers of known begomoviruses used for nucleic acid and amino acid sequence comparison and phylogenetic analysis of 57-v virus......68 4-3. Percent of nucleotide sequence identity between 57-v virus sequence and 14-known begomoviruses........................................................................................71 4-4. Percent of similarity of amino acid sequences between proteins of 57-v virus and proteins of 14-known begomoviruses......................................................................72 5-1. PCR products and digestion fragments of samples with hybridization signal with 2.9-v probe...............................................................................................................91 5-2. PCR products and digestion fragments of samples with hybridization signal with 57-v probe................................................................................................................92 5-3. Nucleotide sequence identity (%) of four partial sequences of the B component with 57-v probe........................................................................................................94 viii

PAGE 9

LIST OF FIGURES Figure page 1-1. Genome structure of begomoviruses............................................................................3 1-2. Hypothetical model for the functional organization of the replication origin in begomoviruses............................................................................................................5 2-1. Map of Venezuela indicating the states where plant tissue samples were collected...................................................................................................................20 2-2. Phylogenetic tree constructed based on the nucleotide sequence of the entire fragment sequenced from amplification using primer set PAR1c496/PAL1v1978 for the A component.................................................................................................39 2-3. Phylogenetic tree constructed based on the nucleotide sequence of the entire fragment sequenced from amplification using primer set PBL1v2040/PCRc154 for the B component.................................................................................................40 3-1. Symptoms of sample 2.9-v collected in Trujillo state, Venezuela.............................45 3-2. Genome organization of 2.9-v sequence a bipartite begomovirus from Trujillo state, Venezuela........................................................................................................50 3-3. Phylogenetic tree based on the complete nucleotide sequence of the A component of the 2.9-v sample from Trujillo state and fifteen bipartite begomoviruses...........55 3-4. Phylogenetic tree based on the complete nucleotide sequence of the B component of the 2.9-v sample from Trujillo state and fifteen bipartite begomoviruses...........56 3-5. Phylogenetic tree based on the complete coat protein nucleotide sequence of the 2.9-v sample from Trujillo state and fifteen bipartite begomoviruses.....................57 3-6. Phylogenetic tree based on the complete replication-associated protein nucleotide sequence of the 2.9-v sample from Trujillo state and fifteen bipartite begomoviruses..........................................................................................................58 4-1. Symptoms of sample 57-v collected in Mrida state, Venezuela...............................64 4-2. Genome organization of 57-v sequence a bipartite begomovirus from Mrida state, Venezuela........................................................................................................69 ix

PAGE 10

4-3. Phylogenetic tree based on the complete nucleotide sequence of the A component of the 57-v sample from Mrida state and fourteen bipartite begomoviruses..........73 4-4. Phylogenetic tree based on the complete nucleotide sequence of the B component of the 57-v sample from Mrida state and fourteen bipartite begomoviruses..........74 4-5. Phylogenetic tree based on the complete coat protein nucleotide sequence of the 57-v sample from Mrida state and fourteen bipartite begomoviruses....................75 4-6. Phylogenetic tree constructed based on the complete replication-associated protein nucleotide sequence of the 57-v sample from Mrida state and fourteen bipartite begomoviruses..........................................................................................................76 5-1. Map of Venezuela indicating the states where tomato samples were collected for this research and the localities where 2.9-v and 57-v viruses were detected using specific probe labeled with 32 P.............................................................................90 5-2. Detection of 57-v virus in digested-PCR product from primers PBL1v2040/PCRc154 of samples from four Venezuelan states: 103-v (Lara), 117-v (Zulia), 261-v (Aragua), 297-v and 307-v (Gurico)..............94 x

PAGE 11

Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy DETECTION, CHARACTERIZATION, AND DISTRIBUTION OF BEGOMOVIRUSES INFECTING TOMATOES IN VENEZUELA By Alba Ruth Nava December 2003 Chair: Jane E. Polston Major Department: Plant Pathology Begomoviruses are members of the geminiviridae family. They are monoor bipartite single-stranded DNA plant viruses that are transmitted by whiteflies. This genus belongs to the geminivirudae family. Begomoviruses are a limiting factor in the production of vegetables and other crops in the tropics and subtropics worldwide. To identify begomoviruses in tomatoes in Venezuela, symptomatic leaves were collected in ten states from 1993 to 1998. Detection of Begomovirus was performed by polymerase chain reaction (PCR) with four sets of primers. Begomoviruses were detected in 50% of the samples. Samples from Andean states (Trujillo, Mrida, and Tchira) were selected to examine variability of begomovirus. Resulting sequences were placed in four groups based on BLAST, GAP, and phylogenetic analyses. Two groups were considered new begomoviruses based on low (<89%) nucleotide sequence identity and phylogenetic analyses. The viruses were designated 2.9-v (from Trujillo) and 57-v virus (from Mrida). The full-length sequences of the A and B component of both viruses were xi

PAGE 12

obtained and compared with known begomoviruses. The 2.9v virus was closely related to Dicliptera yellow mottle virus (DiYMoV). However, the 57-v virus was a unique virus, distantly, but most closely related, to DiYMoV. The hypervariable regions of the B components of both viruses were used as specific probes to determine the distribution of these viruses. The 2.9-v virus was detected in samples from Trujillo, Lara, and Zulia. The 57-v virus was detected in samples from Mrida, Gurico, and Aragua. The distribution of these viruses in such distant states could be explained by movement of infected transplants or by the whitefly vector. We confirmed the high variability of begomoviruses in the Andean states. We expect to find more distinct begomoviruses in samples from other states included in the survey. The genomic characterization of these two new begomoviruses and the generation of specific probes will allow the monitoring of the current prevalence of these two begomoviruses in tomato commercial areas of Venezuela. This information will also be used to develop a virus resistance program against these new begomoviruses using genetic engineering approaches. xii

PAGE 13

CHAPTER 1 LITERATURE REVIEW Begomovirus Description Begomoviruses are small (ca. 18-30 nm) plant viruses with single-stranded circular DNA genomes that are encapsidated in twinned quasi-icosahedral particles. They belong to the Geminiviridae family. Begomoviruses are transmitted by whiteflies, and infect dicotyledonous plants; their genomes can be monoor bipartite (Lazarowitz, 1992). They cause significant and often total yield losses of important food and industrial crops in tropical and subtropical regions of the Western and Eastern Hemispheres (Morales and Anderson, 2001; Navas-Castillo et al. 1998; Polston and Anderson, 1997). High incidences of begomoviruses are associated with high populations of whiteflies and serious losses in several crops in the Americas and the Caribbean Basin (Brown and Bird, 1992; Morales and Anderson, 2001; Polston and Anderson, 1997). Begomovirus Genome Structure and Viral-Encoded Proteins Begomoviruses are mostly bipartite, but some Old World begomoviruses are monopartite. Bipartite begomoviruses have two components, designated A and B. Each component has ~2,600 nt. The genes on the A component are involved in encapsidation and replication, whereas the genes on the B component are involved in the movement of virus through the plant, host range, and symptom expression (Figure 1-1) (Gafni and Epel, 2002; Lazarowitz, 1992). One of the five genes in the A component, the coat protein (CP) gene, is transcribed in the viral sense or clockwise direction. The other four 1

PAGE 14

2 genes replication-associated protein (Rep), transcriptional activator protein (TrAP), replication enhancer (REn), and AC4 are transcribed in complementary sense or counterclockwise direction (Lazarowitz, 1992). The two sets of genes overlap and are separated by an intergenic region (IR), which begins with the start codon of the Rep and ends with the start codon of the CP. This region does not encode any protein and its sequence varies widely among begomoviruses, except that there is a conserved GC-rich inverted repeat sequence, which has the potential to form a stem-loop structure (~30 nt) with the invariant nanomeric TAATATT()AC sequence or loop of the stem-loop structure. The nanomeric sequence contains the initiation site () of rolling circle DNA replication (Gutierrez, 2000; Laufs et al. 1995a), the TATA box, and the forward and inverted repeats. In bipartite begomoviruses, the IR also contains an identical sequence of ~200 nt in the A and the B components called the common region (CR) (Lazarowitz, 1992). The CR sequence is different among different begomoviruses and is used to identify the A and B components of the same virus. The CP is required for encapsidation of progeny virions, vector transmission, virion structure, and host specificity. For bipartite begomoviruses, the CP is not required for either local or systemic viral spread. In contrast, in all monopartite begomoviruses, the CP is essential for viral spread (Gafni and Epel, 2002). The Rep is the only gene essential for replication, being required for transcription of both A and B components (Argello-Astorga et al. 1994). Begomoviruses replicate in the nucleus of infected cells through a double-stranded DNA intermediate via a rolling circle mechanism. The Rep protein has two functional targets or DNA elements in the begomovirus origin of replication: 1) The nanomeric sequence, where the Rep introduces

PAGE 15

3 A B A A R R e e p p T T r r A A P P R R E E n n C C P P C C R R A A A A C C 4 4 ~ ~ 2 2 6 6 0 0 0 0 n n t t B B ~ ~ 2 2 6 6 0 0 0 0 n n t t C C R R B B N N S S P P M M P P H H V V R R R R e e T T r r A A P P R R E E n n C C P P I I R R C C 4 4 ~ ~ 2 2 8 8 0 0 0 0 n n t t V V 2 2 Figure 1-1. Genome structure of begomoviruses. A) Bipartite begomovirus. B) Monopartite bigomoviruses. IR = Intergenic region, CRA = Common region A, CRB = Common region B, CP = Coat protein, TrAP = Transcriptionaactivator protein, REn = Replication-enhancer protein, AC4 = AC4 protein, Rep = Replicationassociated protein, C4 = Symptoms expression V2 = Movement protein MP = Movement protein, NSP = Nuclearshuttle protein.

PAGE 16

4 a site-specific nick to initiate virus replication for the rolling circle mechanism, and 2) a tandemly repeated motif located at variable distances from the conserved hairpin sequence, which is bound specifically by its cognate Rep protein. This motif functions as a major recognition element of the replication origin in begomoviruses. These cis-acting elements belong to a series of iterate DNA motifs called iterons (Argello-Astorga and Ruiz-Medrano, 2001). A functional organization of the replication origin in begomoviruses has been hypothesized (Figure 1-2) (Argello-Astorga et al. 1994). The Rep protein binds to the iterons associated with the TATA box, where a TATA-binding protein was previously bound. A transcription factor binds to a cis-regulatory element, associated with the 5 border of the stem-loop sequence, and creates a nucleosome-free region in its neighborhood. The transcriptional factor interacts with the TATA binding protein, by means of its activation domain, looping the intervening DNA. This event would place the stem-loop structure in an accessible position so that the Rep complex can nick the viral (+) strand in the loop of the hairpin structure. The stem-loop structure may acquire a cruciform structure as a consequence of the interactions with the transcription factor or/and Rep (Argello-Astorga et al. 1994). TrAP is a transactivator of the expression of the CP and the nuclear shuttle protein (NSP) genes (Sunter et al. 1990; Sunter and Bisaro, 1991). The TrAP along with the two proteins encoded by the B component are indirectly involved in the systemic movement of the virus through the plant (Gafni and Epel, 2002). The REn is not essential for viral replication. However, viral DNA replicates at higher levels when REn is present (Sunter

PAGE 17

5 et al. 1990). The AC4 gene is involved in symptom expression of monopartite begomoviruses, but to date does not appear to have a function in bipartite begomoviruses. Figure 1-2. Hypothetical model for the functional organization of the replication origin in begomoviruses. Rep Compl = Replication complex, TBP Complex= TATA-binding protein complex, TF = Transcription factor, S.L = Stem loop (Source: Argello-Astorga et al. 1994). The B component has two genes: the NSP gene, which is transcribed from the viral-sense strand, and a movement protein (MP), which is transcribed from the complementary-sense strand. The NSP is implicated in nuclear shuttling of the viral genome, and MP is involved in cell-to-cell movement of the virus via plasmodesmata (Gafni and Epel, 2002). The MP appears to be a symptom-inducing element or a determinant of pathogenicity of bipartite begomoviruses. Mutation studies suggest that

PAGE 18

6 the 3 region of the MP gene is associated with symptom development (Gafni and Epel, 2002). Bipartite begomoviruses often spontaneously produce approximately half-sized defective DNA B components that function as defective interfering (DI) DNA. The DI DNA may have a biological role during infection to reduce the severity of the disease by competing with the genomic components for cellular resources (Stanley et al. 1990). Monopartite begomoviruses have small circular single-stranded DNA satellites, named DNA These depend on begomoviruses for their proliferation and, in turn, they affect the accumulation and symptom expression of begomovirus (Mansoor et al. 2003). Criteria for Species Demarcation Several taxonomic criteria for demarcating species of begomoviruses have been proposed by the International Committee on Taxonomy of Viruses (ICTV) based on the reliability and applicability of these criteria to the large number of characterized begomoviruses (Fauquet et al. 2003). Nucleotide sequence comparison plays a much greater role in determining taxonomic status. Thus, for comparative analyses, only full-length DNA A sequences were considered, based on recombination events that readily occur among begomoviruses (Fauquet et al. 2003; Pita et al. 2001). A cut-off value of 89% of nucleotide sequence identity (NSI) of the A component was established to distinguish different species from strains (Fauquet et al. 2003). Impact of Begomoviruses on Agriculture Proliferation and rapid dissemination of begomoviruses that infect important food and industrial crops in Latin America have been the consequence of drastic changes in traditional cropping systems (Morales and Anderson, 2001), along with the introduction

PAGE 19

7 of the B biotype of Bemisia tabaci beginning in the mid 1980s (Polston and Anderson, 1997). The B biotype of B. tabaci has displaced many indigenous biotypes, because of its broader host range, higher fecundity, dispersal capacity, virus-transmission efficiency, and resistance to insecticides traditionally used against whiteflies (Brown et al. 1995). Begomoviruses have been reported as limiting factors in the production of several crops in the Americas such as cotton, common bean, tomato, pepper, and cucurbits, among others. (Morales and Anderson, 2001; Polston and Anderson, 1997). In the 1990s, Cassava mosaic begomoviruses caused a major regional pandemic of Cassava mosaic disease (CMD) (affecting parts of at least five countries in Africa) that led to massive economic losses and destabilization of food security (Legg and Thresh, 2000). A key factor in the genesis and spread of the pandemic was the recombination of two distinct cassava mosaic begomoviruses to produce a novel and more virulent hybrid (Pita et al. 2001). Resistance was developed originally in Tanzania, providing effective CMD control in current pandemic-affected areas of East Africa. (Legg and Thresh, 2000). Tomato yellow leaf curl virus (TYLCV) is another example of an emerging virus that causes epidemics worldwide with frequent losses of up to 100% (Moriones and Navas-Castillo, 2000). In tomato, the virus causes prominent upward curling of leaflet margins, reduction of leaflet area, yellowing of young leaves, and stunting of plants. These symptoms were first reported in tomato crops from Israel in the late 1930s, then in Middle Eastern countries from the 1960s to the present. Damage to tomato crops attributed to TYLCV has been reported in the Middle East and Far East, Africa, Europe, Caribbean Islands, Central America, Mexico, and the United States of America (Moriones and Navas-Castillo, 2000).

PAGE 20

8 Management of Begomoviruses Begomovirus management strategies have been implemented in several locations in Central America and the Caribbean after the occurrence of serious agricultural and economic crises caused by begomovirus infection of many crops in the mid 1980s (Hilje, 2002). In the first decade of implementation, area-wide plant-protection campaigns were initiated. These involved quarantine regulations and host-free periods in the Dominican Republic, Mexico, and Cuba. Cultural practices (such as production of seedlings under netting and the use of living ground covers in production fields) are the most novel contributions of this action plan (Hilje, 2002). Several practices have been implemented to control begomoviruses in tomato crops, such as the destruction of infected crops at the end of the production cycle using herbicide combined with oil to kill plants and whiteflies, and then burning the plants; use of virus-free transplants; and removal of infected plants at the first sign of begomovirus symptoms (Schuster and Polston, 1999). In greenhouses, the use of ultraviolet (UV)-absorbing plastic sheets or (UV)-absorbing screens of 50-mesh density (Antignus, 2000) (which interferes with the visual behavior of B. tabaci) has been shown to reduce begomovirus transmission and crop losses. The use of a 50-mesh screen prevents vector entry, but it produces poor ventilation and overheating of the closed structures (Morriones and NavasCastillo, 2000). A good practice is to eradicate plants that can be sources of inoculum, making it possible to reduce the primary spread of the virus (Kashina et al. 2002; Schuster and Polston, 1999). One practice more accepted by growers in many locations is chemical control of the insect vector, but it has a detrimental environmental impact; and whitefly vector populations rapidly develop resistance to insecticides. Insecticides that are not toxic to nontarget species reduce the impact on

PAGE 21

9 important natural enemies such as Eretmocerus sp. and Diglyphus sp. parasitoids compared to conventional insecticides (Hanafi et al. 2002). Host resistance to begomoviruses is another means of management. Development of resistant cultivars by classical plant breeding or genetic engineering requires time, a good scale for evaluating symptom severity, inoculation protocols, and constant adjustment of the resistance due to changes in begomovirus populations (Lapidot and Friedmann, 2002). Some genes for resistance to certain begomoviruses that infect tomato have been identified in wild species of Lycopersicon and have been transferred to cultivated tomato. Thus, resistant cultivars and breeding lines have been generated. Recently, immunity to infection to TYLCV was obtained from L. hirsutum. This immunity was shown to be controlled by three additive recessive genes (Vidavsky and Czosnek, 1998). Another option to obtain resistance is using genetic engineering via pathogen-derived resistance approaches, such as CP-mediated resistance (Kunik et al. 1994; Sinisterra et al. 1999); MP-mediated resistance (Duan et al. 1997; Hou et al. 2000), defective interfering viral DNA (Stanley et al. 1990); Rep gene in antisense orientation (Bendahmane and Gronenborn, 1997); and expression of truncated viral Rep protein (Brunetti et al. 2001; Polston et al. 2001). The two first approaches involve expression of the CP and MP to inhibit viral proliferation. The last three approaches inhibit viral replication by disrupting the activity of the Rep gene. Currently for optimal control of begomoviruses in tomato it is necessary to use several practices simultaneously. Development of new and improved methods of control for begomoviruses depend on our understanding of the mechanisms involved in virus

PAGE 22

10 vector and virus-host plant recognition, and knowledge of the variant forms of the virus in natural populations. Detection and Discrimination of Begomoviruses Begomoviruses have been detected in plants or insects by different techniques, such as visualization of nuclear inclusion bodies by light microscopy, ultrastructural localization of virions in plant cell by transmission electron microscopy, serological assays using polyclonal or monoclonal antibodies (Hunter et al. 1998; Konate et al. 1995; Pico et al. 1999; Polston et al. 1989), DNA hybridization assays (Lotrakul et al. 1998), Polymerase chain reaction (PCR) (Deng et al. 1994; Ghanim et al. 1998; Lotrakul et al. 1998; Mehta-Prem et al. 1994; Pico et al. 1999; Rosell et al. 1999), immunocapture PCR (Rampersad and Umaharan, 2003), and print-PCR (Navas-Castillo et al. 1998) among others. Molecular cloning and DNA sequencing of viral genomes have become the tools of choice, allowing virus identification and evaluation of relationships with other virus isolates (Brown et al. 2001; Padidam et al. 1995; Paximadis et al. 1999; Rybicki, 1994). Variability of Begomovirus Species High diversity among begomovirus species is associated with mixed infections, in which recombination and pseudorecombination events may explain the frequent emergence of new begomoviruses. Recombination is the exchange of DNA between similar DNA components, and pseudorecombination is the exchange of DNA components (Polston and Anderson, 1997). Both events have been demonstrated in the laboratory (Hill et al. 1998; Garrido-Ramirez et al. 2000) and under natural conditions (Pita et al. 2001). Some isolates of African cassava mosaic virus (ACMV), TYLCV and Potato yellow mosaic virus (PYMV) are good examples of recombination and pseudorecombination (Monci et al. 2002; Pita et al. 2001; Umaharan et al. 1998).

PAGE 23

11 Variability of African cassava mosaic virus and Tomato yellow mosaic virus in the Old World The situation of African cassava mosaic disease (ACMD) in Africa is increasing in complexity due to the number and types of begomovirus isolates from different locations within the continent. Four different species, Indian cassava mosaic virus, ACMV, East African cassava mosaic virus (EACMV), and South African cassava mosaic virus, can cause ACMD. A comparison of sequences of African cassava begomoviruses revealed that all the isolates of ACMV (irrespective of their geographical origin) were clustered together with little or no variation in their genomic sequence. However, the sequences of EACMV isolates were more genetically diverse than those of ACMV due to the frequent occurrence of recombination among equivalent components of different strains and species. Variation among EACMV isolates is so high that their classification is becoming problematic (Pita et al. 2001). In addition, a synergistic interaction between ACMV and EACMV has been reported in three countries in Africa: Uganda, Cameroon, and Ivory Coast. This interaction has led to very severe symptomatology and yield losses (Fondong et al. 2000; Pita et al. 2001). The yellow leaf curl disease in tomato is causing important economic losses to this crop worldwide. Phylogenetic analysis of the CP from 23 accessions of TYLCV revealed the presence of seven groups of sequences now classified as seven species: TYLCV-Israel, TYLCV-Sardinia, TYLCV-Thailand, TYLCV-China, TYLCV-Tanzania, TYLCVNigeria, TYLCV-Saudi Arabia (Moriones and Navas-Castillo, 2000). Recombination also contributes to the genetic diversity of emerging begomovirus populations. In southern Spain, Tomato yellow leaf curl Sardinia virus (TYLCSV) and TYLCV are distinct begomovirus species that co-exist in the field and contribute to the

PAGE 24

12 epidemic of tomato yellow leaf curl disease. A natural recombinant between TYLCSV and TYLCV has been detected and an infectious clone of a recombinant isolate (ES421/99) was obtained and characterized. Analysis of its genome showed that recombination sites are located in the intergenic region in which a conserved stem-loop structure occurs at the 3'-end of the REn gene. The ES421/99 exhibited a wider host range than TYLCSV and TYLCV, which might provide it with a selective advantage over the parental genotypes. Field studies revealed that the recombinant strain is becoming the predominant strain in the region in which it was detected (Monci et al. 2002). Variability of Potato yellow mosaic virus in the New World A relatively high variability of PYMV isolates has been reported from the Caribbean, and from Central and South America: Four full-length and five partial sequences of PYMV strains have been reported. Potato yellow mosaic virus-Venezuela (PYMV-VE) was first reported in 1963 in Venezuela (Debrot et al. 1963). A bipartite begomovirus closely related to PYMV-VE was reported to infect tomato in Panama (Engel et al. 1998). The virus was molecularly characterized and called Potato yellow Panama mosaic virus (PYMPV) (formerly named Tomato leaf curl virus-Panama). High amino acid sequence similarity between PYMPV and PYVM-VE was determined for all the open reading frames with the exception of the AC4 gene product (Engel et al. 1998). Another bipartite begomovirus was reported to infect pepper, sweet pepper, okra, beans, and several weeds in different locations in Trinidad (Umaharan et al. 1998). The virus was fully sequenced and named Potato yellow mosaic Trinidad virusTrinidad & Tobago (PYMTV-TT). The full A component of PYMTV-TT has 85% nucleotide sequence identity (NSI) with PYMV-VE; and was proposed to be a recombinant between either

PAGE 25

13 PYMV-VE or PYMPV and Sida golden mosaic Honduras virus (SiGMHV) (Umaharan et al. 1998). Potato yellow mosaic virus-[Guadeloupe] (PYMV-[GP]) was first reported in 1998 as a strain of PYMV-VE (Polston and Bois, 1998). It was responsible for epidemics of virus-like symptoms in tomato in Guadeloupe, Martinique, and Puerto Rico (Polston and Bois, 1998). It was suggested that PYMV-[GP] could be a recent introduction based on the high value of NSI between begomovirus sequences from distant locations (Polston and Bois, 1998). Phylogenetic analysis of the A component of PYMTV-[GP] revealed that this virus is closely related to PYMV-VE, and PYMTV-TT (Urbino et al. in press). In addition, several partial sequences of PYMV-VE strains have been reported such as PYMV-Martinique, PYMV-Puerto Rico, PYMV-Dominican Republic, PYMV-VE strain tomato, and Tomato yellow mosaic virus (ToYMV) (Guzman et al. 1997; Morales et al. 2001; Polston and Bois, 1998). Begomoviruses in Tomato Crops in Venezuela Venezuela was one of the first places worldwide where tomato begomoviruses were identified. In 1963, Debrot and colleagues reported the presence of Tomato yellow mosaic virus (ToYMV), the first begomovirus affecting tomato in Venezuela (Debrot et al. 1963); and the second begomovirus of tomato known in the New World. Whitefly transmission and physical properties of ToYMV were published by Uzctegui and Lastra (1978). Later, a yellow mosaic disease affecting potato crops was reported in Venezuela (Debrot and Centeno, 1985a). The causal agent of this disease was a bipartite begomovirus named Potato yellow mosaic virus-Venezuela (PYMV-VE) (Roberts et al. 1988). The complete nucleotide sequence of PYMV-VE was generated (Coutts et al. 1991).

PAGE 26

14 More recently, other begomoviruses in tomato in Venezuela have been reported, based on partial sequences. One, from the state of Aragua, was considered an isolate or strain of PYMV-VE (PYMV-VE strain tomato) (Guzman et al. 1997). The other (from the states of Monagas, Gurico, and Portuguesa) was considered an undescribed virus and tentatively named Tomato Venezuela virus (ToVEV) (Guzman et al. 1997). Recently the genome of ToYMV was partially sequenced and found to be related to PYMV due to a 95.7% NSI between the two genomes. Thus, it was proposed that PYMV is not a distinct species, but synonym of ToYMV (Morales et al. 2001). Therefore, it appears that there is some variability of begomoviruses in tomato in Venezuela. Background and Objectives Tomato is one of the most important vegetable crops in Venezuela. Its production is concentrated in the following states: Gurico, Lara, Aragua, Trujillo, Tchira and Portuguesa. Since 1975, a yellow mosaic disease has been associated with annual losses of up to 100% in Aragua and Lara states (Lastra and Uzctegui. 1975). Since the mid 1980s, there has been a lack of information regarding the importance of begomoviruses in tomato fields in Venezuela. In the last 23 years, only three partial sequences have been generated (Guzman et al. 1997: Morales et al. 2001). A survey of tomato-growing areas was conducted from 1993 to 1998 with the purpose of identifying and determining the incidence of RNA viruses in tomato in Venezuela. Samples (334) were collected from 10 states: Aragua, Barinas, Cojedes, Gurico, Lara, Mrida, Portuguesa, Tchira, Trujillo, and Zulia. Tomato aspermy virus, Tobacco mosaic virus, Potato virus Y and Cucumber mosaic virus were detected in less than 18% of the samples (Nava et al. 1996, 1997, 1998a, 1998b). To study the begomovirus situation for tomato crops in Venezuela, the

PAGE 27

15 samples collected from this survey were used to address the following general and specific objectives General objective: establish the identity, diversity and distribution of begomoviruses that affect tomato crops in Venezuela. Specific objectives: Detect begomoviruses in genomic DNA extracted from desiccated tomato tissues using PCR. Analyze the variability of begomovirus sequences from Andean samples using partial sequences of the A and B components amplified by PCR using degenerate primers. Characterize two begomovirus, designed 2.9-v and 57-v viruses, which were found in tomato in the Andean states of Trujillo and Mrida. Generate full-length sequences of the A and B components of 2.9-v and 57-v viruses. Determine the distribution of 57-v and 2.9-v viruses in tomato samples from ten states in Venezuela by hybridization with specific probes.

PAGE 28

CHAPTER 2 DETECTION AND VARIABILITY OF BEGOMOVIRUSES IN TOMATO-PRODUCTION AREAS OF ANDEAN STATES, VENEZUELA Introduction The first report of a whitefly-transmitted virus affecting tomatoes in Venezuela was published by Debrot et al. (1963). The tomato plants showed yellow mosaic, stunting, upward cupping, and leaf deformation. In some fields these symptoms were observed in more than 30% of the tomato crops in the state of Aragua. The disease was called tomato yellow mosaic (TYM) and was shown to be caused by the begomovirus Tomato yellow mosaic virus (ToYMV) (Debrot et al. 1963). Transmission and physical properties of ToYMV have been described (Uzctegui and Lastra, 1978). A yellow mosaic disease was also reported in potato in the state of Aragua in 1985 (Debrot and Centeno, 1985a). The causal agent of this disease was thought to be ToYMV and the virus was characterized as a geminivirus. The disease was transmitted by grafting, mechanically, and by whitefly to potato, Lycopersicon esculentum var. cerasiforme (a common weed in Venezuela), and several other hosts of ToYMV (Debrot and Centeno, 1985a). Later the causal agent of a yellow mosaic on potato was characterized and named Potato yellow mosaic virusVenezuela (PYMV-VE) (Roberts et al. 1986; 1988). The full sequence of the A and the B components of PYMV, as well as infectious clones, have been generated (Coutts et al. 1991; Roberts et al. 1988); and the nucleic acid sequence was shown to be closely related to other begomoviruses from the New World, especially in the coat protein (CP) gene regions (Coutts et al. 1991). 16

PAGE 29

17 High incidences of TYM in tomato and other crops in Venezuela have been correlated with high populations of the whitefly, Bemisia tabaci (Anzola and Lastra, 1985; Arnal et al.1993a; 1993b; Debrot et al. 1963; Debrot and Centeno, 1985a, 1985b; Debrot and Ordosgoitti, 1975). Annual rainfall patterns appear to be a major factor that affect whitefly populations, with rapid population increases at the end of the rainy season (November-December). The greatest number of whiteflies was captured at a height of 10 to 60 cm above ground, which coincided with the height of tomato plants (Arnal et al. 1993a). Recently the biotype B has been identified by ramdom amplified polymorphism DNA (RAPD) analysis (Salas and Arnal, 2001). There is evidence for other begomoviruses in tomato in Venezuela. Two partial sequences of begomoviruses affecting tomato in Venezuela have been reported. One, from the state of Aragua, was considered an isolate or strain of PYMV-VE (PYMV-VE strain tomato) (Guzman et al. 1997). The other, from the states of Monagas, Gurico, and Portuguesa, was considered an undescribed virus and named Tomato Venezuela virus (ToVEV) (Guzman et al. 1997). Recently the genome of ToYMV was partially sequenced and found to be related to PYMV based on 95.7% nucleotide sequence identity (NSI). Thus it was proposed that PYMV is not a distinct species, but synonym of ToYMV (Morales et al. 2001). There appears to be great variability of PYMV in Venezuela. Much of what is known about begomoviruses in tomato is based on only very few samples (selected from different states) which were collected and sent to specialists outside of Venezuela for characterization and sequence analysis. Tomato is an important crop in Venezuela and is produced commercially in 13 of the 23 states in Venezuela. A more thorough study of the presence and diversity of begomoviruses is

PAGE 30

18 needed that uses larger numbers of samples and includes samples from all the major tomato-production areas in Venezuela. In addition to begomoviruses, other viruses have been found to infect tomato in Venezuela. A survey conducted in 1975 reported four viruses that were affecting tomatoes in Aragua and Lara states: ToYMV, Tobacco etch virus, Tobacco mosaic virus, and Cucumber mosaic virus. At flowering, all the plants were shown to be infected with ToYMV alone or in combination with one of the other three viruses mentioned above (Lastra and Uzctegui, 1975). Tobacco etch virus was found infecting tomato crops in Yaracuy and Aragua states (Debrot, 1976). A survey of tomato fields in 10 Venezuelan states from 1992 to 1998 reported that, in addition to the viruses reported previously, tomato plants were infected with Potato virus Y, Potato virus X, Tomato aspermy virus, Tomato ringspot virus, Tomato spotted wilt virus, Tobacco streak virus, and Zucchini yellow mosaic virus (Nava, 1999; Nava et al. 1996, 1997, 1998a, 1998a). Objectives of this research were Determine a reliable DNA extraction protocol for amplification of DNA from desiccated tomato tissue of samples from the Andean states Detect begomoviruses in genomic DNA extracted from desiccated tomato tissue using the polymerase chain reaction (PCR) Analyze the variability of begomovirus sequences from Andean samples using partial sequences of the A and B components amplified by PCR using degenerate primers. Materials and Methods Survey Samples of leaf tissue from tomato plants showing symptoms of viral diseases were collected from the main tomato-growing states in Venezuela from 1993 to 1998 (Nava et al. 1997, 1996, 1998a, 1998b). In general, mosaic, curly leaf, yellowing, vein

PAGE 31

19 clearing, stunting, upward cupping, reduced leaf size, and chlorotic leaf margins were the symptoms observed in the tomato plants in the fields. The samples were collected from plants approximately 2 months after transplant to the field. The total number of samples collected was 334. The samples were dried at room temperature for 2 weeks. Then they were cut, rolled in a piece of tissue paper, placed into a vial that was 1/3 filled with silica gel, and then stored at o C. The survey covered 10 states: Aragua, Barinas, Cojedes, Gurico, Lara, Mrida, Portuguesa, Tchira, Trujillo, and Zulia (Figure 2-1). Aragua and Lara have large fields and continuous tomato production. The Andean states (Mrida, Tchira and Trujillo) have small fields and continuous tomato production. The states of Barinas, Cojedes, Gurico, Portuguesa, and Zulia have large fields and one season of tomato production per year. DNA xtraction Genomic DNA was extracted from each sample using the protocol described by Doyle and Doyle (1987). This protocol resulted in a DNA pellet with sign of oxidation from a large number of the samples. Sixteen samples from the Andean states were selected to compare Doyle and Doyle (1987) and another protocol developed to obtain tanninand polysaccharide-free genomic DNA from mature tissue of plants from genera belonging to the Dipterocarpaceae (Rath et al. 1998).

PAGE 32

20 Figure 2-1. Map of Venezuela indicating the states where plant tissue samples were collected. Samples from the Andean states were selected for more in-depth studies because this region was likely to have the greatest variability of begomovirus sequences. The higher diversity of sequences is expected because of the continuous (year-round) production of tomatoes in this region; and because of some growers from the state of Trujillo routinely obtain tomato transplants from the state of Lara, where several begomoviruses have already been reported (Guzman et al. 1997; Lastra and Uzctegui, 1975). Some modifications were made to the Rath protocol: desiccated tissue was rehydrated for 10 min in 2X CTAB (cetyltrimethylammonium bromide) buffer before extraction; and phenol:chloroform:isoamyl in 25:24:1 (v:v:v) was used in place of

PAGE 33

21 phenol:chloroform in 1:1 (v:v). The DNA extracted by both protocols was stored at o C. Polymerase Chain Reaction (PCR) Two sets of primers were used for each component of the viral genome. Primers PAR1c496 and PAL1v1978 (Rojas et al. 1993) amplify ~1100 bp of the A component. This includes ~690 bp of the replication associated protein (Rep), ~300 bp of the region between the beginning of the Rep and the beginning of the coat protein (CP), and ~120 bp of the CP. Primers PCRv181 and PAR1c496 (Rojas et al. 1993) amplify ~300 bp of the A component that contains ~180 bp of region between the beginning of the loop and the beginning of the CP and ~120 bp of the CP. Primers PVL1v2040 and PCRc154 (Rojas et al. 1993) amplify ~600 bp of the B component, which consists of part of the nuclear shuttle protein (NSP) (~160 bp), the entire hypervariable region and part of the common region (CR). The primers JAP 58 (5 -TTCAGTGCCGAAGACCGAAG-3 ) and JAP 59 (5 -ACGGGAAATGGGAGAGGAAG-3 ) are designed for PYMV-VE and amplify a fragment of ~750 bp of the B component, which comprises ~ 420 bp of the NSP, ~20 bp of sequence between the end of the NSP and the movement protein (MP), and ~ 330 bp of the MP. The PCR reactions for primer pairs PAR1c496 and PAL1v1978, and PVL1v2040 and PCRc154 were carried out as previously described (Rojas et al. 1993). For primers PCRv181 and PAR1c496, DNA amplification parameters were 35 cycles of denaturation for 1 min at 94 o C, primer annealing for 20 s at 60 o C, and primer extension for 30 s at 72 o C, with an initial denaturation at 94 o C for 2 mins and a final extension for 7 min at 72 o C. The DNA amplification parameters for primers JAP58 and JAP59 were 35 cycles

PAGE 34

22 of denaturation for 30 s at 94 o C, primer annealing for 30 s at 60 o C and primer extension for 45 s at 75 o C, with an initial denaturing at 94 o C for 5 min and a final extension for 5 min at 72 o C. The PCR reactions were carried out in a Gene Amp PCR system 9700 (PE Applied Biosystems, Foster City, CA) thermocycler. All amplifications were performed in volumes of 25 L containing 10 mM Tris-HCl (pH 9), 50 mM KCl and 1% Triton X-100, 2.5 mM MgCl 2 except for primers JAP58 and JAP59, which contained 2 mM MgCl 2 250 M dNTPs, 0.5 mM spermidine, 1 M of each primer, 100 ng of genomic DNA, and 2.5 U of Taq polymerase. PCR products were electrophoresed (1 h at 90 volts) in 1% agarose gels in Tris-acetate-EDTA buffer, pH 8. Gels were stained with ethidium bromide (0.0015 mg/mL), viewed by a UV transilluminator. Restriction Enzyme Analysis of Purified PCR Products The PCR products from the Andean samples were purified using the QIAquick gel extraction kit (QIAGEN Inc. Valencia, CA). The purified PCR products generated from primer set PAR1c496 and PAL1v1978 (~1100 bp) were digested with BglII, EcoRI, NcoI, and NdeI. The purified PCR products generated from primer set PVL1v2040 and PCRc154 (~700 bp) were digested with EcoRI, HindIII, KpnI, NdeI, and XbaI. Digestion conditions were performed according to manufacturers instructions (New England Biolab, Inc., Beverly, MA). Cloning and DNA Sequence Determination Ligation of purified PCR products amplified from Andean samples was performed using the pGEM-T Easy vector system I (Promega Corporation, Madison, WI) according to the manufacturers instructions. Transformation was performed using XL1-Blue MRF Supercompetent Escherichia coli cells (Stratagene, La Jolla, CA). The

PAGE 35

23 clones were screened with the same restriction enzymes used above. Nucleotide sequences of plasmid DNA from clones were determined by automated sequence analysis at the DNA Sequencing Core Laboratory, Interdisciplinary Center for Biotechnology Research (ICBR) University of Florida, Gainesville. Comparison of DNA Sequence Nucleic acid sequences were analyzed using a Wisconsin Package Version 10.3, [Accelrys (GCG), San Diego, CA]. Basic Local Alignment Search Tool (BLAST), was used to search for similarities between a query sequence and all the sequences in the database. The six begomovirus sequences from the database that had the highest similarity to each Andean sequences were selected to perform GAP analysis. GAP was used to obtain the values of NSI. GAP uses the algorithm of Needleman and Wunsch, which considers all possible alignments and gap positions between two sequences and creates a global alignment that maximizes the number of matched residues and minimizes the number and size of gaps. Multiple alignments of Andean sequences and the begomovirus sequences (selected by the similarity values) were performed using PILEUP, which is the first step for phylogenetic analysis. Phylogenetic Analysis Phylogenetic analysis was carried out using PAUP* (Phylogenetic Analysis Using Parsimony). A maximum parsimony and heuristic tree search were used to generate the best tree. The initial tree was created by stepwise addition with tree-bisection-reconnection as branch swapping method. The reliability of the tree was estimated by performing 500 bootstrap repetitions. The begomoviruses selected for the phylogenetic analysis are listed in table 2-1.

PAGE 36

24 Table 2-1. Acronyms and accession numbers of known begomoviruses used in the phylogenetic analysis of partial sequences of samples from Andean states Species Acronym Accession number Bean dwarf mosaic virus BDMV M88179 Bean golden yellow mosaic virus-[Puerto Rico] BGMV-[PR] M10080 Chino del tomate virus CdTV AF226664 Chino del tomate virus-[IC] CdTV-[IC] AF101476 Dicliptera yellow mottle virus DiYMoV AF170101 Jatropha mosaic virus JMV AF324410 Macroptilium golden mosaic virus-[Jamaica] MGMV-[JM] AF098940 Macroptilium yellow mosaic Florida virus MaYMFV AY044136 Potato yellow mosaic Panama virus PYMPV Y15033 Potato yellow mosaic virus [Guadeloupe] PYMV-[GP] AY120882, Y120883 Potato yellow mosaic Trinidad virus-Trinidad & Tobago PYMTV-TT AF039031, AF039032 Potato yellow mosaic virus[Dominican Republic] PYMV-[DR] AY126611, AY126614 Potato yellow mosaic virusMartinique PYMV-Mart AY126610, AY126612 Potato yellow mosaic virus-Venezuela PYMV-VE D00940, D00941 Potato yellow mosaic virus-Venezuela strain tomato PYMV-VE-[tom] AF026553 Rhynchosia golden mosaic virus RhGMV AF239671 Sida golden mosaic Costa Rica virus SiGMCRV X99550, X99551 Sida golden mosaic Honduras virus SiGMHV Y11097 Tomato dwarf leaf curl virus ToDLCuV AF035225 Tomato golden mottle virus ToGMoV AF138298 Tomato leaf curl Malaysia virus ToLCMV AF327436 Tomato leaf curl Sinaloa virus ToLCSinV AF131213 Tomato mottle Taino virus ToMoTV AF012300, AF012301 Tomato mottle virus-[Florida] ToMoV-[FL] L14461 Tomato Venezuela virus ToVEV AF026464 Results Comparison of DNA Extraction Protocols The Rath protocol (Rath et al. 1998) produced a whiter pellet, and more stable DNA than the Doyle and Doyle DNA extraction protocol (Doyle and Doyle, 1987). A greater number (44%) of samples using the Rath protocol produced the expected fragment size after amplification, especially for primers that amplified the B component

PAGE 37

25 (Table 2-2). DNA that was extracted using the Rath protocol could be amplified after 6 months of storage at 4 o C. There was no difference in PCR results when fresh tissue was used for either DNA extraction protocol (ToMoV-[FL] and Tomato yellow leaf curl virus-[Florida]), however when DNA was extracted from frozen tissue there was a difference in intensity of the bands after electrophoresis of the PCR product from Potato yellow mosaic virus-[Guadeloupe]. In addition, the stability of the DNA extracted with the Doyle and Doyle protocol was low which resulted in inconsistent amplifications. For several dehydrated tomato samples from the survey, brownish pellets were obtained after using the Doyle and Doyle protocol. The color of the pellet was eliminated with an additional incubation period with RNAse A and proteinase K, followed by phenol: chloroform:isoamyl extraction, and cesium chloride precipitation. PCR Amplification of Begomovirus Sequences from Field Samples Approximately 50% of the samples produced a PCR product of ~1100 bp using the degenerate primers PAR1c496 and PAL1v1978 (Table 2-3). The primer set PCRv181 and PAR1c496 was able to amplify a ~300 bp fragment which is located within the ~1100 bp fragment from some but not all of the samples from which a ~1100 bp fragment was obtained (Tables 2-2 and 2-3). Most likely this was due to sequence differences at the binding site for PCRv181 which prevented annealing. It could also be due to the presence of an inhibitor, which might have affected the performance of PCRv181. The percentage of samples from which the B component primers amplified a product was almost 50% less than the percentage obtained for the primer set PAR1c496/PAL1v1978 used to amplify products from the A component (Table 2-3). The results suggested that the primer sets PAR1c496/PAL1v1978 and PBL1v2040/PCRc142

PAGE 38

Table 2-2. Comparison of two DNA extraction protocols through PCR results of samples from Andean states, Venezuela using 4 primer sets to amplify partial sequence of A and B components of begomoviruses A component B component Primers PCRv181/PAR1c496 Primers PAR1c496/PAL1v1978 Primers PBL1v2040/PCRc154 Primers JAP58/JAP59 State Sample ID No. D&D* R** D&D R D&D R D&D R Trujillo 2.2-v + + + + + + + + 2.4-v + + + + + + + 2.7-v + + + + 2.8-v + + + + 2.9-v + + + + 6-v + + + + + + + + 8-v + + + + + + + + 10-v + + + + + + + + Tchira 12t-v + + + 28-v + + + + + + + + 37-v + Mrida 56-v + + + + + + 57-v + + + + + + + 58-v + + 82-v + + + ToMoV-[FL] + + + + + + TYLCV-FL] + + + + PYMV-[GP] + + + + + + + + Healthytom. 26 D&D = Doyle and Doyle DNA extraction; **R = Rath DNA extraction; + = Right PCR product was amplified; = No PCR product was produced ToMoV [FL] = Tomato mottle virus [Florida]; TYLCV [FL] = Tomato yellow leaf curl virus [Florida]; PYMV[GP] = Potato yellow mosaic virus [Guadeloupe]; Healthy tom.= Healthy tomato

PAGE 39

Table 2-3. Polymerase chain reaction results using degenerate primers to amplify A and B components of begomoviruses using Doyle and Doyle DNA extraction of Venezuelan samples from tomato-production areas No. and % of positive PCR samples A component B component State No. of fields/ No. of locations No. of samples Primers PCRv181/PAR1c496 Primers PAR1c496/PAL1v1978 Primers PBL1v2040/PCRc154 Year of collection Aragua 15/ 4 178 64 ( 34%) 81 ( 46%) 44 (25%) 1995-1996 Barinas 2 / 2 8 6 ( 75%) 6 ( 75%) 0 ( 0%) 1995 Cojedes 1/ 1 3 0 ( 0%) 0 ( 0%) 0 ( 0%) 1995 Gurico 3/ 2 13 13 (100%) 13 (100%) 12 (92%) 1998 Lara 10/ 8 27 21 ( 78%) 21 ( 78%) 8 (30%) 1994 Mrida 11/ 6 23 2 ( 9%) 4 ( 17%) 1 ( 4%) 1994 Portuguesa 2/ 2 5 1 ( 20%) 2 ( 40%) 1 (20%) 1995 Tchira 10/ 7 26 2 ( 8%) 2 ( 8%) 2 ( 8%) 1993 Trujillo 6/ 6 19 6 ( 32%) 9 ( 48%) 6 (32%) 1993 Zulia 20/12 32 20 ( 63%) 21 ( 66%) 6 (19%) 1993-1995 Total 80/50 334 135 ( 40%) 159 ( 48%) 80 (24%) 27

PAGE 40

28 are useful for studies of variability of begomoviruses because they were able to amplified larger number of samples, the size of amplified a fragment was optimum for sequencing determination, and the presence of part of the CR in the amplified fragment from the B component enabled the confirmation that the sequences for the A and B components belonged to the same virus. Table 2-4. Expected restriction-fragment sizes (bp) in the fragment amplified by primer set PAR1c496/PAL1v1978 of the A component of known begomoviruses Restriction Enzyme fragment sizes (bp) Sample NdeI NcoI EcoRI BglII TYLCV-[ FL] 1000+291 1234+ 57 760+531 TYLCV IL sev 1015+288 1246+ 57 ToVEV 829+293 762+253+107 781+341 566+329+227 ToMoV-[FL] 844+294 1031+107 808+330 TLCV 944+345 SiGMV 846+294 667+317+156 810+330 PYMV-VE-[tom] 753+375 1021+107 PYMV-VE 1023+107 PYMTV-TT PYMV-[GP] 1025+134 602+556 PHYVV 873+291 819+345 837+327 = No restriction site TYLCV-[FL] = Tomato yellow leaf curl virus[Florida] TYLCV IL sev = Tomato yellow leaf curl virus [Israel strain severe] ToVEV = Tomato Venezuela virus (ToVEV) ToMoV-[FL] = Tomato mottle virus-[Florida] TLCV = Tobacco leaf curl virus SiGMV = Sida golden mosaic Virus PYMV-VE-[tom] = Potato yellow mosaic virus-Venezuela [strain tomato] PYMV-VE = Potato yellow mosaic virus-Venezuela PYMTV-TT = Potato yellow mosaic Trinidad virus-Trinidad & Tobago PYMV-[GP] = Potato yellow mosaic virus [Guadeloupe] PHYVV = Pepper huasteco yellow vein virus Restriction Enzyme Analysis of Purified PCR Product Purified PCR products amplified by primer set PAR1c496/PAL1v1978 for the A component of the Andean samples were digested. Restriction patterns were compared with those expected of the same region of selected begomoviruses (Table 2-4). None of

PAGE 41

29 the Andean samples produced restriction-fragment sizes identical to those selected begomoviruses (Table 2-5). However, the restriction-fragment sizes of each sample were useful to distinguish the presence of single or mixed infections en each sample. There were five samples that appeared to be infected with multiple begomoviruses, 2.2-v, 2.4-v, 8-v, 10-v, and 82-v (Table 2-5). Three of the mixed infections were confirmed by sequence determination of different clones from the same sample (multiple sequences from samples 2.2-v and 8-v could not be obtained). Table 2-5. Restriction-fragment sizes in the fragment amplified by primer set PAR1c496/PAL1v1978 in the A component of begomoviruses from Andean states, in Venezuela Restriction Enzyme fragment sizes (bp) State Sample BglII EcoRI NcoI NdeI Trujillo 2.2-v 600+600 1150 1000+160+116 1150+950+850+175+134 2.3-v 600 +600 1200+850+350 1100+1000+160+116 1100+950+900+500+300 134+260+220+175 6-v 600 +600 1150 900+160+116 1150 2.4-v 600+350 1150+750+350 1000+900+160+116 1150+1000+850 750+300+134 2.7-v 800+350 1150 900+160+108 1150 2.8-v 800+350 1150 900+160+108 1150 2.9-v 800+350 1200 900+160+108 1150 8-v 600+350+250 1200 1100+116 1150+900+800+350 10-v 600+350+250 750+350 1100+900+200+161+130 1150+900+300 Tachira 28-v 600+600 1200 900+185+130 1150+182 Mrida 56-v 800+350 750+350 1000+185 1150+182 57-v 800+350 750+350 1000+185 1150 82-v 800+350 800+370 1000+900+185+130 900+300+182 There were fewer differences in restriction patterns among amplified products using the primer set, PBL1v2040/PCRc154, which amplifies a region of the B component. This could be due to the short length of the PCR product (~700 bp) used for this analysis, and/or the fact that this region has fewer restriction sites than other regions of the genome (Table 2-6). The PCR products of three other primer sets for the B component were evaluated for restriction analysis (data not shown). These generated

PAGE 42

30 larger PCR products, included other regions of the genome and all were more suitable for restriction analysis than the product generated by primer set PBL1v2040/PCRc154. However, the primer set PBL1v2040/PCRc154 was selected for restriction analysis because it amplified a fragment from almost all of the Andean samples that were positive for the A component based on primer set PAR1c496/PAL1v1978 (except for sample 82-v). In addition, this amplified fragment included part of the CR, which is necessary to identify the A and B components of the same virus. The following enzymes were used in the analysis: XbaI which had no restriction sites in any of the samples, HindIII which had a restriction site in one sample, and NdeI and EcoRI which had one restriction site in three and six of 12 samples, respectively (Table 2-7). The restriction analysis suggested that there were several A component sequences but just one B component sequence in three of the samples with mixed infections (samples 2.4-v, 10-v and 82-v). Table 2-6. Expected restriction-fragment sizes in the fragment amplified by primer set PBL1v2040/PCRc154 of the B component of known begomoviruses Restriction Enzyme fragment sizes (bp) X ba I Nde I Hind III EcoR I Kpn I PYMV-GP 458+160 PYMV-VE 444+160 239+365 PYMTV TT 436+148 424+160 ToMoTV 396+231 484+143 331+296 = No restriction site PYMV-[GP] = Potato yellow mosaic virus [Guadeloupe] PYMV-VE = Potato yellow mosaic virus-Venezuela PYMTV-TT = Potato yellow mosaic Trinidad virus-Trinidad & Tobago ToMoTV = Tomato mottle Taino virus DNA Sequence Comparison The sequence of 15 clones obtained from purified PCR products amplified by primer set PAR1c496/PAL1v1978 from 12 samples of Andean states were submitted for

PAGE 43

31 Table 2-7. Restriction-fragment sizes in the fragment amplified by PBL1v2040/PCRc154 of the B component of begomoviruses from Andean states, in Venezuela Restriction Enzyme fragment sizes (bp) State Sample X baI NdeI HindIII EcoRI KpnI Trujillo 2.2-v 700 700 700 550+200 700 2.3-v 700 700 700 700 700+400 2.4-v 700 700 700 500+200 700+500+200 2.7-v 700 550+180 700 700 2.8-v 700 700 700+400+289 700 700 2.9-v 700 700 700 700 6-v 700 700 700 700*+550+200 700+500+200 8-v 700 700 700 700*+550+200 700+500+200 10-v 700 700 700 700*+550+200 700+500+200 Tachira 28-v 700 700 700 700*+550+200 700+500+200 Mrida 56-v 700 700+400+250 700 700 700 57-v 700 700+400+250 700 700 700 a BLAST search. These ~1100 bp sequences were clustered into four groups according to the percentage of NSI by GAP analysis, each group having high NSI with a known begomovirus (Table 2-8). There was a high value of NSI (from 94 to 99%) within the sequences in groups 1 and 2. There was a high NSI among the cloned sequences of group 3 (99%) and 4 (98%). Group 1 sequences (2.2-v37, 6-v, 10-v4, 28-v5, 82-v1, and 82-v5) were characterized by high NSI with PYMV-VE-[tom] (from 92 to 93 %). Group 2 sequences (2.4-v6, 2.4-v7, 10-v1, and 82-v7) were characterized by high NSI with ToVEV (from 93 to 96%). Group 3 sequences (2.7-v1, 2.8-v and 2.9-v1) shared its highest NSI with ToLCSinV (86%), group 4 sequences (56-v4 and 57-v) shared its highest NSI with PYMTV-TT (82% and 83%, respectively). Sequences in groups 3 and 4 had the lowest percentage of NSI with known begomoviruses of the 4 groups. They were therefore considered potential new species of begomoviruses (Table 2-8). Sequences from clones 2.9-v1 and 57-v were selected for complete sequencing of the genome. Sequences belonging to group 1, related to PYMV-VE-[tom], were detected in samples from all three Andean states. Sequences belonging to group 2, related to ToVEV,

PAGE 44

32 were detected in samples from Mrida and Trujillo. All the sequences belonging to group 3 were detected in samples from Trujillo. All the sequences belonging to group 4 were detected in samples from Mrida The sequences of the 15 clones were aligned using PILEUP. The sequences had a conserved nonameric motif 5 -TAATATTAC-3 in the intergenic region, located in the loop of the conserved hairpin element, where Rep introduces a site-specific nick to initiate virus replication via a rolling-circle mechanism (Laufs et al. 1995b). Two inverted repeats, two forward repeats, and a TATA box were present before the stem loop sequence in all the cloned sequences in group 1, the other group sequences showed differences in the number of inverted and forward repeats, keeping the conserved nonameric motif and the TATA box. These inverted repeats or interactive sequences (iterons) have been reported to be specific binding sites for the Rep (Rep iteron-related domain), to initiate the rolling circle replication process (Argello-Astorga et al. 1994, Argello-Astorga and Ruiz-Medrano, 2001) Sequences in groups 1 share the iteron core sequence, GGGGG, and the GSFSIK Rep iteron-related domain with PYMV (Table 2-9). Sequences in group 2 appear to be related to ToVEV. The core sequences of iteron and Rep iteron-related domain in this group were similar to those of ToVEV, there was a nucleotide change resulting in a different amino acid in the Rep iteron-related domain between sequences of group 2 and ToVEV, (lysine for isoleucine due to substitution of A for T in the nucleic acid sequence). The iteron and Rep iteron-related domain sequences of ToVEV were reported as unique, because they did not fit in any other group of iteron and rep iteron-related domains of begomoviruses from the New World (Argello-Astorga and Ruiz-Medrano,

PAGE 45

33 2001). Sequences in group 3 also share the iteron core sequence, GGGGG, and the GSFSIK Rep iteron-related domain with PYMV. PYMV has two inverted and forward repeats, however the sequences in group 3 have only one inverted repeat (Table 2-9). This is a significant difference and implies that sequences in group 3 are unique. The sequences in group 4 are distinctive, since they have one inverted and one forward repeat and are the most similar to those of RhGMV (Table 2-9). Thus, cloned sequences belonging to groups 3 and 4 are tentative new species of begomoviruses from the New World. Partial sequences of the B component (primer set PVL1v 2040/PCRc154) were clustered into four groups after BLAST search and GAP analysis (Table 2-10). Sequences 2.2-vB6, 2.4-vB34, 2.7-vB3, 6-vB2, 8-vB12, 10-vB15, and 28-vB2 had high percentages of NSI (92-95 %) with PYMVVE. The value of NSI within this group ranged from 87 to 99%. These sequences could be strains of PYMV-VE, but further studies would have to be done to confirm this assumption. Most of the samples of this group were separated into two groups based on the comparison of A component sequences (Table 2-8). Thus, a similar comparison could not be made for the B component sequences since B component sequences were not reported for PYMV-VE-[tom] and ToVEV. Therefore, the B component sequences for these samples clustered with PYMV-VE. The sequences 2.8-vB8 and 2.4-vB24 had lower NSI (86 % and 88 %, respectively) with ToMoTV. The sequence 2.9-vB20 had 86 % NSI with ToGMoV, and the sequences 56-vB7 and 57-vB2 had lower percentages of NSI (86 %) with DiYMoV (Table 2-10).

PAGE 46

Table 2-8. Nucleotide sequence identity (%) between partial sequences from PCR using primer set PAR1c496/PAL1v1978 of the A component of begomoviruses from Andean states and four known begomoviruses Virus group 1 Virus group 2 Virus group 3 Virus group 4 Clone 6-v 28-v5 82-v1 82-v5 10-v4 PYMV-VE [tom] 2.4-v7 82-v7 10-v1 ToVEV 2.7-v1 2.9-v1 ToLCSinV 57-v PYMTV-TT 2.2-v37 98.7 96.98 98.22 98.62 98.11 91.93 6-v 97.88 98.71 98.62 98.43 93.14 28-v5 98.15 94.33 97.75 91.67 82-v1 95.87 99.01 91.76 82-v5 98.91 92.53 10-v4 93.00 2.4-v6 96.04 99.52 97.58 93.78 2.4-v7 95.68 93.76 93.35 82-v7 97.63 93.62 10-v1 96.00 2.8-v 99.61 99.32 86.00 2.7-v1 99.06 86.00 2.9-v1 86.00 56-v4 98.77 82.00 57-v 83.00 34 Virus group 1 related to Potato yellow mosaic virus-Venezuela strain tomato (PYMV-VE-[tom] Virus group 2 related to Tomato Venezuela virus (ToVEV) Virus group 3 related to Tomato leaf curl Sinaloa virus (ToLCSinV) Virus group 4 related to Potato yellow mosaic Trinidad virus (PYMTV-TT)

PAGE 47

35 Phylogenetic Analysis The phylogenetic tree of nucleic acid sequences of clones from partial sequences of begomovirus A components is shown in Figure 2-2. The Andean sequences followed the same tendency of clustering that was observed in the GAP analysis. Sequences 2.2-v37, 6-v; 28-v5, 10-v4, 82-v1, and 82-v5 are closely related to the PYMV virus group. Even though this analysis was done with partial sequences it is possible to infer that PYMV-VE strain tomato is distributed in the three Andean states. The data also confirm that MGMV is a sister taxon of the clone group mentioned above. The sequences 10-v1, 2.4-v 6, 2.4-v7 and 82-v7 are closely related to ToVEV. Sequences 57-v and 56-v4 as well as 2.7-v1, 2.8-v and 2.9-v1 had RhGMV and ToLCMV as sister taxa, respectively. These two groups of sequences clustered by themselves, they are not closely related to any known begomovirus. The sequences from the B component were clustered into 3 groups using phylogenetic analysis (Figure 2-4). The largest group of B component sequences (28-vB2, 2.2-vB6, 10-vB15, 6-vB2, 8-vB12, 2.4-vB34, and 2.7-vB3) was most closely related to strains and isolates of PYMV-VE. The next largest group of sequences (2.8-vB8, 2.9-vB20 and 2.4-vB24) was not closely related to any known begomoviruses. The last group of sequences (56-vB7 and 57-vB2) was distantly related to BGYMV-[PR] and MaYMFV. The latter two groups of sequences appeared to be those of new begomoviruses. The partial A and B component sequences from the same sample did not always cluster with the same sequences in the two phylogenetic trees. For example, sequences of from samples, 2.4-v and 2.7-v, were clustered in different groups for the A and B component trees (Figures 2-2 and 2-3). This is primarily due to the fact that there were four clusters of partial sequences in the A component tree and three in the B component

PAGE 48

Table 2-9. Motifs of the intergenic region and Iteron-related domain of Replication protein (Rep IRD) of cloned sequences from Andean states and eight known begomoviruses Group Sequence* Invert repeat Invert repeat Forward repeat Forward repeat Rep IRD 1 2.2-v37 CCCCCGATA CCCCCAATT AATTGGGGG AACTGGGGG MPRKGSFSIK 1 6-v CCCCCGATA CCCCCAATT AATTGGGGG AACTGGGGG MPRKGSFSIK 1 10-v4 CCCCCGATA CCCCCAATT AATTGGGGG AACTGGGGG MPRKGSFSIK 1 28-v5 CCCCCGATA CCCCCAATT AATTGGGGG AACTGGGGG MPRKGSFSIK 1 82-v1 CCCCCGATA CCCCCAATT AATTGGGGG AACTGGGGG MPRKGSFSIK 1 82-v5 CCCCCGATA CCCCCAATT AATTGGGGG AACTGGGGG MPRKGSFSIK PYMV-Mart CCCCCAATA CCCCCAATA AATTGGGGG AACTGGGGG MPRKGSFSIK PYMV-[GP] CCCCCAATA CCCCCAATA AATTGGGGG AACTGGGGG MPRKGSFSIK PYMV-[DR] CCCCCAATA CCCCCAATA AATTGGGGG AACTGGGGG MPRKGSFSIK PYMV-Tom CCCCCAATA CCCCCAATT AATTGGGGG AACTGGGGG MPRKGSFSIK PYMV-VE CTCCCAATA CCCCCTATT AATTGGGGG AACTGGGGG MPRKGSFSIK 2 10-v1 TGCACCGATT AATTGGGGCA AATTGGGGTC MPPPKHFRLN 2 2.4-v6 TGCACCGATT AATTGGGGCA AATTGGGGTC MPPPKHFRLN 2 2.4-v7 TGCACCGATT AATTGGGGCA AATTGGGGTC MPPPKHFRLN 2 82-v7 TGCACCGATT AATTGGGGCA AATTGGGGTC MPPPKHFRLN ToVeV TGCACCGATT AATTGGGGCA AAATGGGGTC MPPPKHFRIN 3 2.7-v1 CCCCCAATT AATCGGGGG AACTGGGGG MPRKGSFSIK 3 2.8-v CCCCCAATT AATCGGGGG AACTGGGGG MPRKGSFSIK 3 2.9-v1 CCCCCAATT AATCGGGGG AACTGGGGG MPRKGSFSIK 4 56-v4 ACACCAATT AATCGGTGT MPTARAFKIN 4 57-v ACACCAATT AATCGGTGT MPTARAFKIN RhGMV ACCCCGATT TATCGGTGT TATCGGTAT MPQPRRFRIN 36 Virus names and accession number are described in material and methods. Iteron core and Rep IRD motif are shown in italic.

PAGE 49

Table 2-10. Nucleotide sequence identity (%) between partial sequences from PCR using primer set PBL1v2040/PCRc154 of the B component of begomoviruses from Andean states and four known begomoviruses Virus group 1 Virus group 2 Virus group 3 Virus group 4 Clone 2.7-vB3 6-vB2 2.2-vB6 8-vB12 10-vB15 28-vB2 PYMV-VE 2.4-vB24 ToMoTV ToGMoV 57-vB2 DiYMoV 2.4-vB34 99.01 95.37 92.00 91.43 91.91 91.22 94 2.7-vB3 95.04 91.90 90.92 91.58 90.89 93 6-vB2 94.00 91.90 93.21 92.88 92 2.2-vB6 88.00 95.10 92.10 92 8-vB12 87.56 87.33 93 10-vB15 93.20 95 28-vB2 97 2.8-vB8 94.86 86 2.4-vB24 86 2.9-vB20 86 56-vB7 96.55 86 57vB2 86 Virus group 1 related to Potato yellow mosaic virus-Venezuela (PYMV-VE) 37 Virus group 2 related to Tomato Mottle Taino virus (ToMoTV) Virus group 3 related to Tomato golden mottle virus (ToGMoV) Virus group 4 related to Dicleptera yellow mottle virus (DiYMoV)

PAGE 50

38 tree. This is probably a result of the presence of fewer B component sequences in the GenBank database. Greater number of sequences allow for greater precision in phylogenetic analysis. Discussion Begomoviruses were detected in almost of samples collected from tomato plants in commercial fields in Venezuela. Begomoviruses were detected in nine out of the ten Venezuelan states included in this study. These are Aragua, Barinas, Gurico, Lara, Mrida, Portuguesa, Tchira, Trujillo and Zulia states. This study reports the presence of begomoviruses in tomato plants for the first time in the states of Barinas, Mrida, Tchira, and Trujillo state. Until this study, tomato-infecting begomoviruses had only been reported from Lara, Aragua, Gurico, Monagas, Portuguesa, and Zulia (Debrot et al. 1963; Guzman et al. 1997; Lastra and Uzctegui, 1975; Nava et al. 1996). The greatest number of samples positive for at least one begomovirus came from the states of Barinas, Gurico, Lara and Zulia. There appeared to be at least four different begomoviruses in the Andean states. Three types of analyses: GAP, iteron and Rep iteron-related domain comparison, and phylogenetic analysis of the partial sequences of the A component supported the presence of four groups of sequences (Tables 2-8, 2-9, 2-10 and Figure 2-2). The individual sequences remained in the same groups regardless of the analysis. All four begomovirus sequence groups appear to be New World in origin. This is the first report of begomovirus sequences in tomato plants from the Andean states of Venezuela. The variety of begomovirus sequences found in this study is significant, as these were found

PAGE 51

39 Figure 2-2. Phylogenetic tree constructed based on the nucleotide sequence of the entire fragment sequenced from amplification using primer set PAR1c496/PAL1v1978 for the A component. Tree was generated using PAUP program. A single most parsimonious tree was predicted by a heuristic search with stepwise addition, random branch-swapping, tree-bisection-reconnection options (500 replication for bootstrapping). Bootstrap indices are shown at each node. Scale bar references branch length as frequency of changes per site

PAGE 52

40 Figure 2-3. Phylogenetic tree constructed based on the nucleotide sequence of the entire fragment sequenced from amplification using primer set PBL1v2040/PCRc154 for the B component. Tree was generated using PAUP program. A single most parsimonious tree was predicted by a heuristic search with stepwise addition, random branch-swapping, tree-bisection-reconnection options (500 replication for bootstrapping). Bootstrap indices are shown at each node. Scale bar references branch length as frequency of changes per site.

PAGE 53

41 in just 15 samples which were positive for a begomovirus from a total of 68 plants sampled. Similarly, numerous variations in begomovirus sequences have been reported from tomato plants in Brazil (seven species) and Trinidad & Tobago (three species) (Ribeiro et al. 2003; Umaharan et al. 1998). Analysis of the partial sequence of the A and B components from group 1 and 2 suggests that these begomovirus components are closely related to ToVEV, PYMV-VE-[tom], and PYMV-VE (Tables 2-8 and 2-10, Figures 2-2 and 2-3), which were previously reported in Venezuela (Debrot et al. 1963; Guzman et al. 1997; Roberts et al. 1986). PYMV-VE and ToVEV were first reported in Venezuela in tomato samples from Aragua state, and PYMV-VE-[tom] was also first reported in Venezuela in tomato growing areas in Monagas, Gurico and Portuguesa states (Debrot et al. 1963; Guzman et al. 1997; Roberts et al. 1986). This suggests that these viruses represented by these partial sequences are established in Venezuela. This study has found PYMV-VE, and PYMV-VE-[tom], are widely distributed within the Andean states. ToVEV was found in two Andean states (Trujillo and Mrida). The Andean states are separated from the states of Aragua, Monagas, Gurico and Portuguesa by more than 370 Km. It is unexpected that these distant tomato production regions share some of the same begomoviruses. The dissemination of these begomoviruses among states might be possible by the movement of Bemisia tabaci biotype B or more likely by the movement of infected plants. Analysis of partial sequences also suggests that there are two unique and uncharacterized begomoviruses present in samples from Trujillo and Mrida states (groups 3 and 4). Only the full-length genome sequences of the A component are considered for comparative analysis and 89% NSI has been proposed to demarcate

PAGE 54

42 species. Since recombination events have been shown in this genus, partial sequences are not sufficient to distinguish new species (Fauquet et al. 2003). Studies are in progress to obtain complete sequences of the genomes of begomoviruses from appropriate samples to determine if they are indeed new species of begomoviruses. This study has shown the presence of four begomoviruses in tomato in Andean states in Venezuela. Two of these viruses are tentatively considered new species of begomoviruses. In addition, a high percent of samples collected from tomato fields were positive for the presence of at least one begomovirus. These data suggest that begomoviruses may be a bigger concern for the Venezuelan tomato industry than was previously believed.

PAGE 55

CHAPTER 3 CHARACTERIZATION OF A NEW BEGOMOVIRUS FROM THE STATE OF TRUJILLO, VENEZUELA Introduction Begomoviruses are members of the Geminiviridae family, and are characterized by their circular single-strand DNA genomes enclosed within a distinctive geminate capsid. Begomoviruses are transmitted to dicotyledonous plants by a whitefly vector, Bemisia tabaci (Fauquet et al. 2003). Begomoviruses have been reported in the Americas since the middle 1960s. Several epidemics have been reported which caused serious problems in different crops in the Americas (Engel et al. 1998; Morales and Anderson, 2001; Polston and Anderson, 1997), especially after the introduction of biotype B of B. tabaci and after changes in traditional cropping systems (Morales and Anderson, 2001; Polston and Anderson, 1997). Four begomoviruses have been reported affecting tomatoes in Venezuela. Tomato yellow mosaic virus (ToYMV) was the first begomovirus reported to affect tomatoes in Aragua state, Venezuela (Debrot et al. 1963). Recently a partial sequence of ToYMV was generated and considered closely related to Potato yellow mosaic virus-Venezuela (PYMV-VE) (Morales et al. 2001). PYMV-VE was reported as the causal agent of a yellow mosaic disease affecting potato crops (Coutts et al. 1991; Roberts et al. 1988). A tomato strain of PYMV-VE and Tomato Venezuela virus were also reported to affect tomatoes in Venezuela (Guzman et al. 1997). 43

PAGE 56

44 The presence of two new begomoviruses has also been reported in tomatoes from Andean states of Venezuela (Chapter 2). One of these two begomoviruses was detected in the state of Trujillo. Based on the nucleotide sequence identity (NSI) (86%) of partial sequences, this virus is distantly related to Tomato leaf curl Sinaloa virus (ToLCSinV). The partial sequences were obtained by the polymerase chain reaction (PCR) using primers PAR1c496 and PAL1v1978 (Rojas et al. 1993). Using the newly established criteria to distinguish species of begomoviruses (Fauquet et al. 2003), the full-length sequence of the genomic components must be determined to confirm that the virus detected from Trujillo is a new and unique begomovirus. The objectives of this research were: Characterize a begomovirus that was found in tomato in Trujillo state, Venezuela Generate full-length sequences of the A and B components of this new virus. Materials and Methods Plant Sample and Extraction of Genomic DNA Young leaves of a symptomatic tomato plant were collected as part of a survey of viruses infecting tomato in Trujillo state, Venezuela (Nava et al. 1997). The symptoms on this plant included curly leaves, chlorotic leaf margins, foliar deformation and reduced leaf size (Figure 3-1). This sample, designated 2.9-v, appeared to be an undescribed begomovirus based on comparison of DNA partial sequences (Chapter 2). Genomic DNA was extracted from frozen, desiccated leaf tissue (Rath et al. 1998) for use in genomic characterization studies.

PAGE 57

45 Figure 3-1. Symptoms of sample 2.9-v collected in Trujillo state, Venezuela. Curly leaves, yellow margin of the leaves, mosaic and reduction of the size of the leaves are shown Obtaining Full-Length Sequences Degenerate primers PAR1c496/PAL1v1987 and PBL1v2040/PCRc154 (Rojas et al. 1993), were used to generate partial sequences of A and B components from sample 2.9-v. From these partial sequences, two sets of specific primers (JAP122/JAP123 and JAP128/JAP129) (Table 3-1) were designed to amplify the remaining sequence of the A and B components. A Wisconsin Package Version 10.3, [Accelrys (GCG), San Diego, CA] was used to design the primers. The PCR parameters were: 35 cycles of denaturation for 1 min at 94 o C, primer annealing for 1 min at 57 o C (JAP122/JAP123) or 62 o C (JAP128/JAP129) and primer extension for 1 min at 72 o C, with an initial denaturation at 94 o C for 5 min and a final extension of 7 min at 72 o C. PCR reactions were carried out in a PE Applied Biosystems GeneAmp PCR System 9700 thermocycler (PE Applied

PAGE 58

46 Biosystems, Foster City, CA). Genomic DNA (~100 ng) extracted from desiccated tissue was amplified in a volume of 25 L containing 10 mM Tris-HCl (pH 9), 50 mM KCl and 1% Triton X-100, 1.5 mM MgCl 2 (JAP122/JAP123) or 1.25 mM MgCl 2 (JAP128/JAP129), 250 M dNTPs, 1.0 mM spermidine, 1 M of each primer, and 1.25 U of Taq polymerase. PCR products were electrophoresed (1 h at 90 volts) in 1% agarose gels in Tris-acetate-EDTA buffer, pH 8. Gels were stained with ethidium bromide (0.0015 mg/mL), viewed with a UV transilluminator. The PCR product was purified using a QIAquick gel extraction kit (QIAGEN Inc. Valencia, CA), cloned using the pGEMT Easy system (Promega Corporation, Madison, WI), and transformed into XL1-Blue supercompetent Escherichia coli cells (Stratagene, La Jolla, CA) according to the manufacturers instructions. Nucleotide sequences were determined (Ana-Gen Technologies, INC. Atlanta, GA). Internal unique primers (JAP134/JAP135) for the A and (JAP138/JAP139) the B component (Table 3-1) were designed to complete the viral genome. The PCR conditions for the internal unique primers were: 35 cycles of denaturation for 1 min at 94 o C, primer annealing for 1 min at 57 o C, and primer extension for 1 min at 72 o C, with an initial denaturation at 94 o C for 5 min and a final extension for 7 min at 72 o C. All amplifications were performed as previously described above, using 2 mM MgCl 2 for primer set JAP134/JAP135 and 1.5 mM MgCl 2 for primer set JAP138/JAP139, and plasmid DNA diluted 1:100 (v/v) as a template. The plasmid DNA was derived from clones of the PCR products generated by primer sets JAP122/JAP123 and JAP128/JAP129.

PAGE 59

47 Table 3-1. Specific primers designed to determine the full-length sequence of a new begomovirus from Trujillo, Venezuela. Primer Sequence Amplified fragment size(bp) /DNA component JAP122 5'-GTCGAACCGGAAAGACAATG-3' 1479 bp / A component JAP123 5'-ACAGGCCCATGTACAGGAAG-3' JAP128 5'-ATATGAATCGGGGGAACTGGG-3' 2005 bp / B component JAP129 5'-GCAGCACAATTAACGGCAAG-3' JAP134 5'-TATGCCAGTAACGAGCAGTC-3' 528 bp / A component JAP135 5'-ACCTCCCAAATAAAAACGCC-3' JAP138 5'-CCCAATTAAATGACCTGGTTCG-3' 654 bp / B component JAP139 5'-ACGTTCATCAATTACCCTGTTC-3' JAP150 5'-TACACTGCA GGGCCC TTTGAG-3' 2525 bp / Full B component JAP151 5'-GAGATTCATA GGGCCC AGTCCA-3' JAP166 5'-GCAA TGGCCA CTTAGATAG-3' 2560 bp / Full A component JAP167 5'-CGTAG TGGCCA TCTTGA-3' Restriction-enzyme sites are underlined Obtaining Infectious Clones Two sets of overlapping primers, JAP166/JAP167 and JAP150/JAP151 (Table 3-1) were designed to amplify a PCR product of the full-length sequence of each component, using genomic DNA extracted from sample 2.9-v. The primers JAP166/JAP167 overlap at an MscI site, which is located in the replication associated protein (Rep) gene in the A component sequence of 2.9-v (Figure 3-2 A). The primers JAP150/JAP151 overlap at an ApaI site in the movement protein (MP) gene of the B component sequence of 2.9-v (Figure 3-2 B). The PCR conditions for the overlapping primers were: 10 cycles of denaturation for 30 s at 94 o C, primer annealing for 30 s at 52 o C for primer set JAP166/JAP167 and 60 o C for primer set JAP150/JAP151, and primer extension for 2.5 min at 72 o C, following by 25 cycles in which gradually the extension time was increased in 10 s/cycle, an initial denaturation at 94 o C for 2 min and a final extension for 7 min at 72 o C. PCR reactions were carried out in a MJ Research PTC200 DNA Engine thermocycler (MJ Research, Inc. Waltham, Massachusetts). Genomic DNA

PAGE 60

48 (~100 ng) extracted from desiccated tissue was amplified in a volume of 25 L containing 10 mM Tris-HCl (pH 9), 50 mM KCl and 1% Triton X-100, 2.5 mM MgCl 2 100 M dNTPs, 0.4 M of each primer, and 1.25 U of Expand High Fidelity Plus Taq DNA polymerase (Roche Diagnostics Corporation, Indianapolis, IN). The PCR products were purified using a QIAGEN kit, then ligated using the pGEM-T Easy system, and transformed into XL-1 Blue supercompetent E. coli cells. Comparison of DNA and Amino Acid Sequences The full-length sequences of the A and B components were analyzed using the Wisconsin Package Version 10.3, Accelrys (GCG). Similarities between a query sequence and all the sequences in the database were established using Basic Local Alignment Search Tool (BLAST). Full-length sequences of 15 begomoviruses were selected based on their high percentage of identity with the A and B component sequences obtained from 2.9-v sample as established by BLAST. The values of NSI and amino acid sequence similarity were generated using GAP analysis. Multiple sequence comparisons between the 15 known begomoviruses and the sequence from 2.9-v sample were performed using PILEUP. Phylogenetic Analysis Phylogenetic analysis was performed using PAUP* (Phylogenetic Analysis Using Parsimony) [Wisconsin Package Version 10.3, Accelrys (GCG)]. A maximum parsimony and heuristic tree search was used to create the best tree with stepwise addition, tree-bisection-reconnection, and random branch-swapping options. The reliability of the tree was estimated by performing 500 bootstrap repetitions. The selected 15 sequences of begomoviruses (Table 3-2) were used for DNA and amino acid sequence comparison, and also for phylogenetic analysis.

PAGE 61

49 Table 3-2. Acronyms and accession numbers of known begomoviruses used for nucleic acid and amino acid sequence comparison and phylogenetic analysis of 2.9-v virus. Species Acronym Accession number Bean calico mosaic virus BCaMV AF110189, AF110190 Bean dwarf mosaic virus BDMV M88179, M88180 Bean golden mosaic virusBrazil BGMV-[BZ] M88686, M88687 Cabbage leaf curl virus CaLCuV U65529, U65530 Cucurbit leaf curl virus-[Arizona] CuLCuV-[AZ] AF256200, AF327559 Chino del tomate virus[H8] CdTV-[H8] AF101476, AF101478 Dicliptera yellow mottle virus DiYMoV AF139168, AF170101 Pepper golden mosaic virus PepGMV U57457, AF499442 Potato yellow mosaic Trinidad virus-Trinidad & Tobago PYMTV-TT AF039031, AF039032 Sida golden mosaic Costa Rica virus SiGMCRV X99550, X99551 Sida golden mosaic Honduras virus SiGMHV Y11097, Y11098 Sida golden mosaic virus SiGMV U77963, AF039841 Tomato chlorotic mottle virus-[Brazil] ToCMoV-[BZ] AF90004, AF491306 Tomato mottle virus[Florida] ToMoV-[FL] L14460, L14461 Tomato mottle Taino virus ToMoTV AF012300, AF012301

PAGE 62

50 B PBL1v2040 ( 1893 nt ) JAP129 ( 1926 nt ) JAP128 ( 2446 nt ) MP ( 1220 2059 nt ) JAP151 ( 1228 nt ) JAP150 ( 1279 nt ) JAP138 ( 1242 nt ) N SP ( 285 1016 nt ) JAP139 ( 588 nt ) PCRc154 ( 2517 nt ) A C4 ( 2007 2261 nt ) R e p ( 1345 2418 nt ) JAP122 ( 1729 nt ) PAL1v1978 ( 1724 nt ) JAP 166 ( 1684 nt ) JAP 167 ( 1682 nt ) TrA P ( 1029 1421 nt ) JAP135 ( 1212 nt ) R En ( 884 1279 nt ) JAP134 ( 684 nt ) CP ( 138 881 nt ) PAR1c496 ( 255 nt ) JAP123 ( 250 nt ) A Figure 3-2. Genome organization of 2.9-v sequence a bipartite begomovirus from Trujillo state, Venezuela. A) Open reading frames of the A component; B) Open reading frames of the B component. Start and end position of each ORF, primers and binding sites are shown. CP = Coat protein, TrAP = Transcriptionalactivator protein, REn = Replication-enhancer protein, AC4 = AC4 protein, Rep = Replicationassociated protein, MP = Movement protein, NSP = Nuclearshuttle protein.

PAGE 63

51 Results Full-Length Sequencing The full-length sequences of the A and B components were obtained from sample 2.9-v. The common region shared by both components contains 159 nt. The NSI value of this common region was 97.16%, and therefore both genomes belong to the same bipartite begomovirus, designated 2.9-v virus. The genome organization of 2.9-v virus was similar to those of bipartite begomoviruses from the New World. The A component has five putative genes (Figure 3-2A) and the B component has two putative genes (Figure 3-2B). Conserved sequences for begomoviruses were present in the intergenic region of the A component, such as a nonameric motif 5'-TAATATTAC-3', one inverted repeat (5'-CCCCCAATT-3'), two forward repeats (5'-AAYYGGGGG-3'), and a TATA box, which is located before the hairpin element sequence. The core sequences of the iteron or forward repeat (5'-GGGGG-3'), and the iteron-related domain of the replication protein (GSFSIK) were most similar to those of PVMV-VE, ToMoTV, and DiYMoV (Argello-Astorga and Ruiz-Medrano, 2001). Full-length sequences of each component of 2.9-v virus were generated using overlapping primers JAP150/JAP151 and JAP166/JAP167. Full-length clones were constructed. Comparison of DNA and Amino Acid Sequences The values of NSI between 2.9-v virus and known begomoviruses for the full-length sequence of the A component were low (68.5-78.39%) (Table 3-3). The NSI values are lower than the cut-off value to distinguish different species of begomoviruses (Fauquet et al. 2003), which confirms that 2.9-v virus is an uncharacterized begomovirus. The highest NSI value for the full 2.9-v A component (78.39%) and Nuclea-shuttle

PAGE 64

52 protein (NSP) gene (70.35%) was with the NSP of SGMCRV (Table 3-3). The highest NSI value (82.17%) for the 2.9-v coat protein (CP) gene was with the CP of PepGMV. The highest NSI value for 2.9-v transcriptional activator protein (TrAP), and replication enhancer protein (REn) genes were with those of BCaMV. The highest NSI value for 2.9 AC4 gene were with those of PYMTV-TT and CdTV-[H8]. The sequence of the replication associated protein (Rep) gene and the common region had 86.53 % and 78.52% of NSI with ToMoV-[FL], respectively. The highest value of NSI for the movement protein (MP) gene was obtained between 2.9-v virus and DiYMoV (Table 3-3). Table 3-3. Percent of nucleic acid identity between 2.9-v virus sequence and 15 known begomoviruses Virus A Comp B Comp CR CP TrAP REn AC4 Rep MP NSP SiGMCRV 78.39 67.31 61.81 79.57 74.41 80.51 85.66 78.02 78.21 70.35 PYMTV-TT 77.95 64.91 67.85 78.76 76.54 81.36 89.92 79.11 76.31 64.82 SiGMHV 77.55 64.60 71.97 78.76 73.97 78.09 88.37 78.49 77.02 70.35 BDMV 77.50 64.54 68.84 81.45 74.22 78.09 87.98 77.74 76.67 68.51 ToMoTV 76.17 63.90 78.52 79.44 70.61 77.33 88.76 78.73 77.86 68.19 DiYMoV 76.08 67.55 70.80 81.37 76.61 76.13 83.33 77.16 78.65 69.27 CdTV-[H8] 76.02 65.30 69.53 80.16 73.27 77.89 89.88 77.75 77.22 69.88 ToMoV-[FL] 75.26 65.93 65.65 78.68 71.72 77.38 86.82 86.53 77.22 67.58 BGMV-[BZ] 74.62 64.41 68.84 79.38 78.97 81.20 72.94 72.42 73.55 67.75 ToCMoV-[BZ] 74.21 61.81 68.64 78.98 75.13 78.69 82.17 73.53 70.82 65.14 SiGMFV-A1 74.07 64.25 72.00 78.87 73.70 75.44 85.99 74.86 77.58 68.64 BcaMV 72.52 61.55 62.59 79.65 81.03 82.46 56.91 69.16 71.31 66.21 CaLCuV 72.10 66.39 51.10 81.96 74.67 75.13 56.08 68.05 76.51 69.57 CuLCuV-AZ 70.88 61.71 45.93 80.29 76.23 77.89 53.72 66.35 72.12 64.98 PepGMV 68.50 62.54 43.18 82.17 74.81 75.63 NA 63.65 74.49 69.11 Virus names and accession numbers are described in material and methods. NA = Not available A Comp = A component, B Comp = B component, CR = Common region, CP = Coat protein, TrAP = Transcriptional-activator protein, REn = Replication-enhancer protein, AC4 protein = AC4, Rep = Replication-associated protein, MP = Movement protein, NSP = Nuclear-shuttle protein.

PAGE 65

53 Amino acid sequence similarities between 2.9-v virus and known begomoviruses were higher (ranging from 32.24% to 94.38%) than the NSI values for almost all the open reading frames, except for AC4 (Table 3-4). Higher values of similarity were obtained for the CP, followed by the similarity values of the Rep, REn, TrAP, and AC4. The values of similarity were higher for the MP than those for the NSP (Table 3-4). Phylogenetic analysis The relationships between the 2.9-v virus sequence and 15 known begomoviruses were analyzed using the full-length A and B component sequences as well as the sequences of the CP and Rep genes. The phylogenetic trees of the full-length sequences of the A and B components revealed that 2.9-v sequence is in the DiYMoV clade and Table 3-4. Percent of similarity of amino acid sequences between proteins of 2.9-v virus and proteins of 15 known begomoviruses Virus* CP TrAP Ren AC4 Rep MP NSP SiGMCRV 92.77 69.77 81.82 73.26 85.03 86.83 75.68 PYMTV-TT 93.57 71.52 82.71 82.56 87.18 87.54 69.73 SGMHV 91.57 67.69 79.70 76.74 84.60 87.54 79.26 BDMV 94.38 71.54 81.95 77.90 84.55 88.97 76.03 ToMoTV 91.97 66.92 80.45 79.10 87.90 72.72 DiYMoV 93.57 67.69 80.45 82.56 83.01 86.48 74.31 CdTV-[H8] 93.57 67.69 80.45 82.56 83.08 86.47 74.31 ToMoV-[FL] 91.57 66.92 81.20 74.12 81.06 89.68 71.56 BGMV-[BZ] 94.78 78.46 81.20 51.77 79.11 85.41 74.88 ToCMoV-[BZ] 91.97 72.31 82.71 66.28 81.25 83.63 72.48 SiGMFV-A1 91.57 67.18 76.52 73.26 83.38 86.61 75.34 BcaMV 93.98 78.46 84.96 34.15 73.28 86.43 72.73 CaLCuV 94.38 71.54 75.94 32.94 71.26 86.12 73.85 CuLCuV-AZ 93.17 81.20 50.00 NA 71.71 81.45 65.14 PepGMV 93.57 72.31 79.70 NA 69.65 84.84 74.77 82.40 = Virus names and accession numbers are described in material and methods. NA = Not available. CP = Coat protein, TrAP = Transcriptional-activator protein, REn = Replication-enhancer protein, Rep = Replication-associated protein, AC4 = AC4 protein, MP = Movement protein, NSP = Nuclear-shuttle protein.

PAGE 66

54 clustered with CaLCuV and PepGMV (Figures 3-3 and 3-4). The phylogenetic trees for the Rep and the CP confirm that 2.9-v virus is closely related to DiYMoV (Figures 3-5 and 3-6). Discussion The full-length sequences of the DNA-A and DNA-B of 2.9-v virus were generated. The A and the B components had 2,560 nt and the 2,525 nt, respectively (Figure 3-2). Although, the B component DNA of 2.9-v virus appears to be smaller than other bipartite-begomovirus genomes, smaller B DNA components have been reported for several begomoviruses (Abouzid et al. 1992a; Abouzid, et al. 1992b; Coutts et al. 1991; Umaharan et al. 1998). The inferred-genome organization of 2.9-v virus was similar to those of bipartite begomoviruses from the New World (Figure 3-2). The A component contains four genes in complementary sense (Rep, TrAP, REn, AC4) and one gene (CP) in viral sense. The B component contains two genes MP and NSP (Gutierrez, 2000). The conserved sequences in the IR are similar to those of other New World bipartite begomoviruses. The core of the forward repeat or iteron core (GGGGG) and the Rep iteron-related domain sequences (GSFSIK) are similar to those of PYMV-VE. PYMV-VE has two inverted and forward repeats (Arguello-Astorga et al. 2001). However, one inverted and two forward repeats were observed in the IR of 2.9v virus. These results confirm that 2.9-v virus is a unique begomovirus from the New World. The NSI values for the full-length of the A component between of 2.9-v virus sequence and 15 known begomoviruses were always less than 89% (Table 3-3), which confirmed that 2.9-v virus sequence is a distinct begomovirus (Fauquet et al. 2003). The

PAGE 67

55 Figure 3-3. Phylogenetic tree based on the complete nucleotide sequence of the A component of the 2.9-v sample from Trujillo state and fifteen bipartite begomoviruses.The tree was generated using the PAUP program. A single most parsimonious tree was predicted by a heuristic search with stepwise addition, random branch-swapping, tree-bisection-reconnection options (500 replication for bootstrapping). Bootstrap indices are shown at each node. Scale bar references branch length as frequency of changes per site

PAGE 68

56 Figure 3-4. Phylogenetic tree based on the complete nucleotide sequence of the B component of the 2.9-v sample from Trujillo state and fifteen bipartite begomoviruses. The tree was generated using the PAUP program. A single most parsimonious tree was predicted by a heuristic search with stepwise addition, random branch-swapping, tree-bisection-reconnection options (500 replication for bootstrapping). Bootstrap indices are shown at each node. Scale bar references branch length as frequency of changes per site

PAGE 69

57 Figure 3-5. Phylogenetic tree based on the complete coat protein nucleotide sequence of the 2.9-v sample from Trujillo state and fifteen bipartite begomoviruses.The tree was generated using the PAUP program. A single most parsimonious tree was predicted by a heuristic search with stepwise addition, random branch-swapping, tree-bisection-reconnection options (500 replication for bootstrapping). Bootstrap indices are shown at each node. Scale bar references branch length as frequency of changes per site

PAGE 70

58 Figure 3-6. Phylogenetic tree based on the complete replication-associated protein nucleotide sequence of the 2.9-v sample from Trujillo state and fifteen bipartite begomoviruses.The tree was generated using the PAUP program. A single most parsimonious tree was predicted by a heuristic search with stepwise addition, random branch-swapping, tree-bisection-reconnection options (500 replication for bootstrapping). Bootstrap indices are shown at each node. Scale bar references branch length as frequency of changes per site

PAGE 71

59 values of NSI were noticeably higher for the CP. This is not surprising since the CP sequence is very conserved within the begomovirus genus (Brown et al. 2001; Padidam et al. 1995). The CP determines vector specificity in begomoviruses. This fact explains why almost all the begomoviruses are transmitted by whiteflies (Timmermans et al. 1994; van den Heuvel et al. 1999). Full-length clones for the A and B components have been constructed from purified PCR product using overlapping primer methodology (Patel et al. 1993). Determination of the host range using biolistic inoculation is in progress. Symptoms in inoculated plants will be recorded and correlated with those observed in the 2.9-v sample. Mixed infection can occur in the field, thus this correlation may be of little practical use. Therefore, the viral characterization of 2.9-v virus was based on molecular data. The 2.9-v virus is a begomovirus from the New World closely related to DiYMoV based on phylogenetic trees of the full-length sequence of the A and the B components (Figures 3-3 and 3-4). DiYMoV and 2.9-v virus appeared to have a common ancestor, based on phylogenetic trees of the full-length of the A and B components, and of the Rep gene (Figures 3-3, 3-4, and 3-6). The begomoviruses clustered with the 2.9-v virus do not have tomato as a host with the exception of PepGMV (Torres-Pacheco et al. 1996) (Figures 3-3, 3-4, and 3-6). The 2.9-v virus is probably a distinct species, albeit related to DiYMoV. The phylogenetic tree of the sequence of the CP was not congruent with the other trees generated in this research (Figure 3-5). However, the relationship between the 2.9-v virus and DiYMoV was maintained in this tree (Figure 3-5). Remarkably, 2.9-v virus has the longest branch length in the CP tree, which implies that 2.9-virus is more divergent from the other begomoviruses in that clade (Baldauf, 2003) (Figure 3-5).

PAGE 72

60 Possibly, the 2.9-v virus evolved and acquired the ability to infect tomato crops. These results confirm that 2.9-v is a distinct begomovirus. Since the 2.9-v virus is an undescribed bipartite begomovirus from Trujillo state, Venezuela, it is necessary to determine its distribution in other states where tomato crops are grown in Venezuela. Thus, it is possible to infer the actual importance of this virus in Venezuela. It could be the first step to initiate a management program of begomoviruses in tomato crops.

PAGE 73

CHAPTER 4 CHARACTERIZATION OF A NEW BEGOMOVIRUS FROM THE STATE OF MERIDA, VENEZUELA Introduction Begomoviruses are members of the family Geminiviridae, they have circular single-strand DNA genomes packaged within twinned particles, and are transmitted by the whitefly, Bemisia tabaci, to dicotyledonous plants (Fauquet et al. 2003). Begomoviruses are a limiting factor of many crops in the Americas, particularly with the introduction of the polyphagous biotype B of Bemisia tabaci and changes in traditional cropping systems (Morales and Anderson, 2001; Polston and Anderson, 1997). In addition, the misuse of insecticides has increased the development of pesticide-resistant populations of this whitefly and the elimination of natural enemies of the whitefly resulting in epidemics of geminivirus-like symptoms in the Americas (Engel et al. 1998; Morales and Anderson, 2001; Polston and Anderson, 1997). A begomovirus has been reported affecting tomatoes in Venezuela. Tomato yellow mosaic virus (ToYMV) was the first begomovirus reported to affect tomatoes in Venezuela (Debrot et al. 1963). The symptoms produced by this virus were described as yellow mosaic, stunting, and upward cupping and deformation of the leaves (Debrot et al. 1963). Whitefly transmission and physical properties of ToYMV were published in 1978 (Uzctegui and Lastra, 1978). A yellow mosaic disease affecting potato crops was first reported in Venezuela (Debrot and Centeno, 1985a). The causal agent of this disease was determined to be a 61

PAGE 74

62 bipartite begomovirus and named Potato yellow mosaic virus-Venezuela (PYMV-VE) (Roberts et al. 1988). The nucleotide sequence of PYMV-VE indicated that it is a begomovirus of the New World (Coutts et al. 1991). Several strains of PYMV have been reported from Central and South America, and the Caribbean. A bipartite begomovirus closely related to PYMV-VE was reported to infect tomato in Panama (Engel et al. 1998). The virus was molecularly characterized and called Potato yellow Panama mosaic virus (PYMPV) (formerly named Tomato leaf curl virus-Panama). High amino acid sequence similarity between PYMPV and PYVM-VE was determined for the open reading frames with the exception of the AC4 gene product. Venezuela was proposed as the possible origin of PYMPV because of the close relationship between these two viruses (Engel et al. 1998). Another bipartite begomovirus was reported to infect pepper, sweet pepper, okra, beans, and several weeds in different locations in Trinidad (Umaharan et al. 1998). The virus was named Potato yellow mosaic Trinidad virusTrinidad & Tobago (PYMTV-TT). The A component of PYMTV-TT has 85% nucleotide sequence identity (NSI) with PYMV-VE and, and may be a recombinant between either PYMV-VE or PYMPV and Sida golden mosaic Honduras virus (SiGMHV) (Umaharan et al. 1998). Potato yellow mosaic virus-[Guadeloupe] (PYMV-[GP]) was first reported in 1998 as a strain of PYMV-VE (Polston and Bois, 1998). It was responsible for epidemics of virus-like symptoms in tomato in Guadeloupe, Martinique, and Puerto Rico. It was suggested that PYMV-[GP] could be a recent introduction based on high value of NSI between begomoviruses sequences at distant locations (Polston and Bois, 1998). Phylogenetic analysis of the A component of PYMTV-[GP] revealed that this virus is closely related to PYMV-VE, and PYMTV-TT (Urbino et al. in press).

PAGE 75

63 Although partial sequences of the A component have been used to identify PYMV-VE strains such as PYMV-Martinique, PYMV-Puerto Rico, PYMV-Dominican Republic, Tomato yellow mosaic virus and PYMV-VE strain tomato (Guzman et al. 1997; Morales et al. 2001; Polston and Bois, 1998), only full-length sequences of the A component are considered for determining taxonomic status in begomoviruses, and the species demarcation would be < 89% of NSI. Thus, virus isolates with >89% of NSI may be considered to be the species and should have the same name, irrespective of the host from which they were derived. Further designation into strains may be justified by demonstration of biological differences (Fauquet et al. 2003). The objectives of this research were: Characterize a new begomovirus that was found in tomato in Mrida state Venezuela Generate full-length sequences of the A and B components of this new virus. Materials and Methods Plant Sample and Extraction of Genomic DNA A leaf sample from a plant with mottled and severely deformed leaves (Figure 4-1) was collected as part of a survey of viruses in tomato in Mrida state, Venezuela (Nava et al. 1997). This sample, designated 57-v, appeared to be a unique begomovirus based on results from a study which compared partial DNA sequences amplified from tomato samples collected from 10 states in Venezuela (Chapter 2). Genomic DNA was extracted from frozen, desiccated leaf tissue (Rath et al. 1998). Obtaining Full-Length Sequences Partial sequences of A and B components from sample 57-v were generated using degenerate primers PAR1c496/PAL1v1987 and PBL1v2040/PCRc154 (Rojas et al.

PAGE 76

64 Figure 4-1. Symptoms of sample 57-v collected in Mrida state, Venezuela. Mottling and severe deformation of the leaves were observed. 1993). From these partial sequences two sets of specific primers (JAP124/JAP125 and JAP130/JAP131) (Table 4-1) were designed to amplify the remainder of the A and B component sequences. The specific primers were designed using a Wisconsin Package Version 10.3, [Accelrys (GCG), San Diego, CA]. The PCR parameters were: 35 cycles of denaturation for 1 min at 94 o C, primer annealing for 1 min at 55 o C (JAP124/JAP125) or 62 o C (JAP130/JAP131) and primer extension for 1 min at 72 o C, with an initial denaturation at 94 o for 5 min and a final extension of 7 min at 72 o C. PCR reactions were carried out in a PE Applied Biosystems GeneAmp PCR System 9700 thermocycler (PE Applied Biosystems, Foster City, CA). Genomic DNA (~100 ng) extracted from desiccated tissue was amplified in a volume of 25 L containing 10 mM Tris-HCl (pH 9), 50 mM KCl and 1% Triton X-100, 1.5 mM MgCl 2 (JAP124/JAP125) or 1.75 mM MgCl 2 (JAP130/JAP131), 250 M dNTPs, 1.0 mM spermidine, 1 M of each primer, and 1.25 U of Taq polymerase. PCR products were electrophoresed (1 h at 90 volts) in

PAGE 77

65 1% agarose gels in Tris-acetate-EDTA buffer, pH 8. Gels were stained with ethidium bromide (0.0015 mg/mL)and viewed with a UV transilluminator. The PCR product was purified using QIAquick gel extraction kit (QIAGEN Inc. Valencia, CA), cloned using the pGEMT easy system (Promega Corporation, Madison, WI), and transformed into XL1-Blue supercompetent Escherichia coli cells (Startagene, La Jolla, CA) according to manufacturers instructions. Nucleotide sequences were obtained from the partial sequences and combined (Ana-Gen Technologies, INC. Atlanta, GA). Internal unique primers (JAP136/JAP137) for the A and (JAP140/JAP141) the B component (Table 4-1) were designed to complete the viral genome. The PCR conditions for the internal unique primers were: 35 cycles of denaturation for 1 min at 94 o C, primer annealing for 1 min at 55 o C (JAP136/JAP137) or 57 o C (JAP140/Jap141), and primer extension for 1 min at 72 o C, with an initial denaturation at 94 o C for 5 min and a final extension for 7 min at 72 o C. All amplifications were performed as previously described above, using 1.5 mM MgCl 2 for both sets of primers, and plasmid DNA diluted 1:100 (v/v) as the template. The plasmid DNA was derived from clones of the PCR products generated by primer sets JAP124/JAP125 and JAP130/JAP131. Obtaining Full-Length Clones Two sets of overlapping primers, JAP154/JAP155 and JAP168/JAP169 (Table 4-1), were designed to amplify a PCR product of the full-length sequence of each component, using genomic DNA from sample 57-v. The primers JAP154/JAP155 overlap in a KpnI site, site in the movement protein (MP) gene of the B component sequence of 57-v (Figure 4-2 B). The PCR conditions for the overlapping primers were: 35 cycles of denaturation for 1 min at 94 o C, primer annealing for 2 min at 53 o C

PAGE 78

66 (JAP154/JAP155) or 57 o C (JAP168/JAP169), and primer extension for 3 min at 72 o C, with an initial denaturation at Table 4-1. Specific primers designed to determine the full-length sequence of 57-v virus from Mrida, Venezuela. Primer Sequence Amplified-fragment size (bp) /DNA component JAP124 5'-CAGTCGTTCCTCCAATTATTCC-3' 1621 bp / A component JAP125 5'-TGATATGCAAGAATGGGAG-3' JAP130 5'-CAGTTTCCTTCCACTGCTGC-3' 2079 bp / B component JAP131 5'-ATTGGAGTCTCTCAACTCTCTC-3' JAP136 5'-ATGCCAGCAATGAGCAAG-3' 446 bp / A component JAP137 5'-GAGTGTACCACATCCAAATCAG-3' JAP140 5'-CAGTGATAGGGGGAACAAAGGG-3' 857 bp / B component JAP141 5'-GTGTTAAACGTGCAGATGGG-3' JAP154 5'-TCACGCGG GTAC CTTTTAT-3' 2575 bp / Full A component JAP155 5'-TTCGCACTG GTAC CTAGACA-3' JAP168 5'-GATAGCAG GAC CCCAGTCT-3' 2543 bp / Full B component JAP169 5'-TTGGG GTC CCTGCACATTAG-3' Restriction-enzyme sites are underlined 94 o C for 5 min and a final extension for 10 min at 72 o C. PCR reactions were carried out in a PE Applied Biosystems GeneAmp PCR System 9700 (PE Applied Biosystems, Foster City, CA) thermocycler. Genomic DNA (~100 ng) extracted from desiccated tissue was amplified in a volume of 25 L containing 10 mM Tris-HCl (pH 9), 50 mM KCl and 1% Triton X-100, 3.0 mM MgCl 2 (JAP154/JAP155) or 2.0 mM MgCl 2 (JAP168/JAP169), 250 M dNTPs, 1.0 mM spermidine, 1 M of each primer, and 1.25 U of Expand High Fidelity Plus Taq DNA polymerase (Roche Diagnostics Corporation, Indianapolis, IN). PCR product of the full-length sequence was ligated into pGEM-T Easy system and transformed in XL-1 Blue supercompetent E. coli cells. The ligation reaction was most efficient if done the same day that the PCR was performed. It was assumed that the A overhangs (necessary for cloning with pGEM T Easy) on the ends of the PCR products were lost during storage. Thus, the formation of blunt-ended fragments

PAGE 79

67 is reduced. In addition, exposure to UV light must be limited as much as possible to reduce the production of pyrimidine dimers. High efficiency in the cloning was obtained, when the PCR products of the full-length sequences were not exposed to UV and not frozen. Comparison of DNA and Amino Acid Sequences The full-length sequences of the A and B components were analyzed using the Wisconsin Package Version 10.3, Accelrys (GCG). Similarities between a query sequence and all the sequences in the database were established using Basic Local Alignment Search Tool (BLAST). Full-length sequences of 14 begomoviruses were selected based on their high percentage of identity with the A and B component sequences obtained from sample 57-v as established by BLAST. The values of NSI and amino acid sequence similarity were generated using GAP analysis. Multiple sequence comparisons between the fourteen known begomoviruses and the sequence from sample 57-v were performed using PILEUP. Phylogenetic Analysis Phylogenetic analysis was carried out using PAUP* (Phylogenetic Analysis Using Parsimony) [Wisconsin Package Version 10.3, Accelrys (GCG)]. A maximum parsimony and heuristic tree search was used to generate the best tree with stepwise addition, tree-bisection-reconnection, and random branch-swapping options. The reliability of the tree was estimated by performing 500 bootstrap repetitions. The selected 14 sequences of begomoviruses (Table 4-2) were used for DNA and amino acid sequence comparison, and also for phylogenetic analysis.

PAGE 80

68 Table 4 -2. Acronyms and accession numbers of known begomoviruses used for nucleic acid and amino acid sequence comparison and phylogenetic analysis of 57-v virus. Species Acronym Accession number Bean golden mosaic virus Brazil BGMV-[BZ] M88686, M88687 Cabbage leaf curl virus CaLCuV U65529, U65530 Chino del tomate virus[IC] CdTV-[IC] AF101476, AF101478 Dicliptera yellow mottle virus DiYMoV AF139168, AF170101 Pepper golden mosaic virus PepGMV U57457, AF499442 Potato yellow mosaic Panama virus PYMPV Y15033, Y15034 Potato yellow mosaic virus [Guadeloupe] PYMV-[GP] AY120882, AY120883 Potato yellow mosaic Trinidad virus-Trinidad & Tobago PYMTV-TT AF039031, AF039032 Potato yellow mosaic virus-Venezuela PYMV-VE D00940, D00941 Sida golden mosaic Costa Rica virus SiGMCRV X99550, X99551 Tomato chlorotic mottle virus-[Brazil] ToCMoV-[BZ] AF90004, AF491306 Tomato golden mosaic virus-yellow vein TGMV-YV K02029, K02030 Tomato mottle Taino virus ToMoTV AF012300, AF012301 Tomato rugose mosaic virus [Uberlandia] ToRMV-[Ube AF291705, AF291706 Results Full-Length Sequencing and Cloning The full-length sequences of an A and the B component from tomato sample 57-v were generated. A region of 134 nt was common between the A and B components and the NSI value of this common region was 99.25%, which confirms that both components conform to the bipartite begomovirus of 57-v sample. The genome structure deduced from nucleotide sequence resembled that of other bipartite begomoviruses from the New World, with five and two putative genes for the A (Figure 4-2A) and the B components (Figure 4-2B), respectively. Conserved begomovirus sequences were present in the A component, such as a nonameric motif 5 -TAATATTAC-3 in the intergenic region (IR),

PAGE 81

69 A JAP124 ( 231 nt ) PAR1c496 ( 298 nt ) A C4 ( 2041 2301 nt ) CP ( 181 924 nt ) R e p ( 1421 2452 nt ) JAP136 ( 728 nt ) JAP125 ( 1852 nt ) PAL1v1978 ( 1758 nt ) R En ( 927 nt ) JAP155 ( 1495 nt ) JAP154 ( 1494 nt ) JAP137 ( 1174 nt ) TrAP ( 1072 1455 nt ) B JAP131 ( 2439 nt ) PCRc154 ( 2534 n t ) JAP141 ( 468 nt ) JAP130 ( 1975 nt ) 57 v B 2543 nt PBL1v2040 ( 1927 nt ) N SP ( 383 1153 nt ) M P ( 1215 2093 nt ) JAP168 ( 1367 nt ) JAP169 ( 1364 nt ) JAP140 ( 1325 nt ) Figure 4-2. Genome organization of 57-v sequence a bipartite begomovirus from Mrida state, Venezuela. A) Open reading frames of the A component; B) Open reading frames of the B component. Start and end position of each ORF, primers and binding sites are shown. CP = Coat protein gene, TrAP = Transcriptional-activator protein, REn = Replication-enhancer protein, AC4 = AC4 protein, Rep = Replication-associated protein, MP = Movement protein, NSP = Nuclear-shuttle protein

PAGE 82

70 this motif is located in the loop of a hairpin element, where the Rep protein introduces a site-specific nick to initiate virus replication via the rolling-circle mechanism (Laufs et al. 1995). One inverted repeat (5'-ACACCAATT-3'), one forward repeat (5'-AATCGGTGT-3'), and a TATA box were present before the stem loop sequence. The core sequences of the iteron or forward repeat (5'-GGKGT-3'), and the iteron-related domain of the replication protein (FRIN) were most similar to those of BGMV-[BZ], and Rhynchosia golden mosaic virus (Argello-Astorga and Ruiz-Medrano, 2001). Full-length genomic PCR product was generated using primer sets JAP154/JAP155 and JAP168/JAP169 for the A and B components, respectively. Full-length clones were constructed for both genomic A and B components. Comparison of DNA and Amino Acid Sequences The percentages of NSI between 57-v sample and known begomoviruses for the full-length sequence of the A component were low (70-77%) (Table 4-3), which indicated that 57-v sequence was a distinct begomovirus species (Fauquet et al. 2003). The highest NSI value for the full 57-v A component (77%), replication-enhancer protein (REn) gene (81.82 %), Rep (87.84 %), and the common region (79.85 %) were with PYMTV-TT (Table 4-3). The highest NSI value (75.78 %) for the 57-v transcriptionalactivator protein gene (TrAP) was with TGMV-YV. PYMTV-TT and CdTV-[CI] had the highest values of NSI with the AC4 sequence of the begomovirus from the 57-v sample. The sequence of the coat protein (CP) had 84.08 % of NSI with ToCMoV-[BZ] (Table 4-3). Regarding the B component sequence and its genes, the highest values of NSI were always obtained between 57-v sequences and DiYMoV.

PAGE 83

71 Table 4-3. Percent of nucleotide sequence identity between 57-v virus sequence and 14-known begomoviruses Virus A Comp B Comp CR CP TrAP REn AC4 Rep MP NSP PYMTV-TT 77.07 65.94 79.85 80.11 73.70 81.82 87.84 79.69 76.56 68.75 SiGMCRV 76.32 65.15 63.3. 80.78 73.96 78.79 NA 79.36 76.68 73.57 TGMV-YV 76.27 64.06 65.87 82.86 76.82 78.79 83.46 74.65 79.28 69.79 ToRMV-[Ube] 75.83 64.82 62.69 81.86 75.78 79.80 85.71 76.65 76.11 70.44 ToCMoV-[BZ] 75.40 63.46 74.44 84.08 75.78 81.82 77.31 75.97 72.24 70.83 CdTV-[IC] 75.28 63.84 68.18 79.12 73.64 79.55 87.88 77.97 76.07 72.40 DiYMoV 74.53 66.98 59.70 82.93 72.92 77.78 82.35 75.87 79.07 74.09 PYMV-VE 74.49 64.75 62.12 81.26 73.39 79.45 NA 75.09 76.76 68.87 BGMV-[BZ] 74.26 65.01 69.92 80.65 75.78 80.80 72.94 73.74 74.52 73.39 TTMV 74.10 63.43 71.43 79.92 71.32 78.03 86.43 77.10 76.64 70.82 PYMV-[GP] 74.04 64.73 68.42 79.65 73.13 79.70 81.40 74.69 76.30 68.75 PYMPV 73.85 64.25 64.44 80.32 73.90 81.06 77.52 74.11 74.94 70.04 CaLCuV 72.17 64.76 58.12 82.66 75.52 78.53 54.65 65.09 76.56 72.91 PepGMV 70.14 64.71 55.17 82.79 74.48 79.04 NA 63.71 74.74 72.01 Virus names and accession numbers are described in material and methods. NA = Not available. A Comp = A component, B Comp = B component, CR = Common region, CP = Coat protein gene, TrAP = Transcriptional-activator protein, REn = Replication-enhancer protein, AC4 = AC4 protein, Rep = Replication-associated protein, MP = Movement protein, NSP = Nuclear-shuttle protein. The values of similarity of the amino acid sequences between 57-v virus and known begomoviruses were higher (ranged from 70% to a 100%) for almost all the open reading frames, except for AC4 (Table 4-4). Higher values of similarity were obtained for the CP, followed by the similarity values of the Rep, REn, TrAP, and AC4. The values of similarity were higher for the MP than those for the nuclearshuttle protein (NSP) (Table 4-4). Phylogenetic Analysis The relationships between 57-v sequence and known begomoviruses were analyzed using the genomic A and B components sequences as well as the sequences of the coat protein (CP) and Rep genes. The phylogenetic trees of the sequences of the A and B components revealed that 57-v virus is a New World begomovirus (Figures 4-3 and 4-4). The B component sequence of 57-v is clustered with DiYMoV, CaLCuV and

PAGE 84

72 PepGMV (Figure 4-4). The phylogenetic trees for the Rep and the CP confirm that 57-v is closely related to bipartite begomoviruses from the New World (Figures 4-5 and 4-6). Table 4-4. Percent of similarity of amino acid sequences between proteins of 57-v virus and proteins of 14-known begomoviruses Virus* CP TrAP REn AC4 Rep MP NSP PYMTV-TT 93.95 67.19 85.61 71.77 89.80 87.42 70.82 SGMCRV 94.35 68.75 84.09 NA 88.08 88.09 73.93 TGMV-YV 95.55 71.88 85.61 72.73 83.43 89.73 71.60 ToRMV-[Ube] 93.15 73.44 83.33 72.73 84.30 87.08 75.10 ToCMoV-[BZ] 93.15 71.09 85.61 60.00 85.18 83.33 73.15 CdTV-[IC] 93.95 67.18 84.09 74.71 86.63 86.74 73.93 DiYMoV 93.15 69.53 78.03 65.88 84.01 87.75 75.30 PYMV-VE 93.95 68.75 82.58 NA 83.14 86.73 70.43 BGMV-[BZ] 95.16 76.56 86.36 52.94 82.27 85.03 72.73 ToMoTV 93.54 66.41 83.33 71.76 84.01 89.46 72.76 PYMV-[GP] 93.95 66.41 100.00 62.35 83.14 87.42 70.04 PYMPV 93.55 68.75 84.85 62.35 84.85 87.42 71.60 CaLCuV 95.16 70.31 81.06 28.57 71.47 85.71 75.10 PepGMV 93.55 71.09 83.33 NA 70.21 85.52 75.49 = Virus names and accession numbers are described in material and methods. NA = Not available. CP = Coat protein, TrAP = Transcriptional-activator protein, REn = Replication-enhancer protein, AC4 = Putative AC4 protein, Rep = Replication-associated protein, MP = Movement protein, NSP = Nuclear-shuttle protein. Discussion The full-length sequences of A and B components of 57-v virus were generated. The A and the B components contained 2,575 nt and the 2543 nt, respectively (Figure 4-2). Although, the genome of 57-v virus seems to be smaller than other bipartite-begomovirus genomes, there are several bipartite begomoviruses with short genomes such as Tomato mottle virus-[Florida] (Abouzid et al. 1992a), CaLCuV (Abouzid, et al. 1992b), PYMV-VE (Coutts et al. 1991), PYMTV-TT (Umaharan et al. 1998), among others. The inferred-genome organization of 57-v virus was similar to those of bipartite begomoviruses from the New World (Figure 4-2). The A component contains four genes in the complementary sense (Rep, TrAP, REn, AC4) and one gene (CP) in the viral sense.

PAGE 85

73 Figure 4-3. Phylogenetic tree based on the complete nucleotide sequence of the A component of the 57-v sample from Mrida state and fourteen bipartite begomoviruses. The tree was generated using the PAUP program. A single most parsimonious tree was predicted by a heuristic search with stepwise addition, random branch-swapping, tree-bisection-reconnection options (500 replication for bootstrapping). Bootstrap indices are shown at each node. Scale bar references branch length as frequency of changes per site

PAGE 86

74 Figure 4-4. Phylogenetic tree based on the complete nucleotide sequence of the B component of the 57-v sample from Mrida state and fourteen bipartite begomoviruses. The tree was generated using the PAUP program. A single most parsimonious tree was predicted by a heuristic search with stepwise addition, random branch-swapping, tree-bisection-reconnection options (500 replication for bootstrapping). Bootstrap indices are shown at each node. Scale bar references branch length as frequency of changes per site

PAGE 87

75 Figure 4-5. Phylogenetic tree based on the complete coat protein nucleotide sequence of the 57-v sample from Mrida state and fourteen bipartite begomoviruses. The tree was generated using the PAUP program. A single most parsimonious tree was predicted by a heuristic search with stepwise addition, random branch-swapping, tree-bisection-reconnection options (500 replication for bootstrapping). Bootstrap indices are shown at each node. Scale bar references branch length as frequency of changes per site

PAGE 88

76 Figure 4-6. Phylogenetic tree constructed based on the complete replication-associated protein nucleotide sequence of the 57-v sample from Mrida state and fourteen bipartite begomoviruses. The tree was generated using the PAUP program. A single most parsimonious tree was predicted by a heuristic search with stepwise addition, random branch-swapping, tree-bisection-reconnection options (500 replication for bootstrapping). Bootstrap indices are shown at each node. Scale bar references branch length as frequency of changes per site

PAGE 89

77 The B component contains two genes MP and NSP (Gutierrez, 2000). The conserved sequences in the IR are similar to those of bipartite begomoviruses. However, one inverted and one forward repeat was observed in the IR. Usually begomoviruses have two inverted and two forward repeats. The sequence of the iteron-related domain of replication protein is similar to those of BGMV-[BZ], PYMTV-TT, Sida golden mosaic Honduras virus, Rhynchosia golden mosaic virus, Tomato mosaic Havana virus [Quivican], and Cowpea golden mosaic virus-[Brazil] (Argello-Astorga et al. 2001). The NSI values for the A component between of 57-v virus sequence and 14-known begomoviruses were always less than 89% (Table 4-3), which confirmed that 57-v virus is a distinct begomovirus (Fauquet et al. 2003). The highest values of NSI were observed for the CP, this corroborates that the CP sequence is more conserved than the remainder of the genome of begomoviruses (Brown, et al. 2001; Padidam et al. 1995). Full-length clones for the A and B components were generated from purified PCR product using overlapping primer methodology (Patel et al. 1993). Biolistic inoculation will be conducted to determine host range of this new begomovirus from Mrida state, Venezuela. This will corroborate that the full-length clones are in fact infectious and representative of the virus. The phylogenetic analysis revealed that the 57-v virus is not closely related to PYMV-VE or to other PYMV species. The 57-v virus is an undescribed begomovirus related to DiYMoV based on phylogenetic trees of the full-length sequence of the A and the B components (Figures 4-3 and 4-4). A similar relationship was observed in the phylogenetic trees of the sequence of the CP and Rep genes (Figures 4-5 and 4-6).

PAGE 90

78 DiYMoV was first reported to be closely related to BGMV isolates. In addition, tomato was reported as a no host of DiYMoV (Lotrakul et al. 2000). In some way, the DiYMoV lost the ability to recognize tomato plants as a host in its evolution process. Since the 57-v virus is an undescribed bipartite begomovirus, it is necessary to determine its distribution in other states where tomato crops are grown in Venezuela. The knowledge of 57-v virus sequence and its distribution in Venezuela could be the first step to initiate a management program of begomoviruses in tomato crops.

PAGE 91

CHAPTER 5 DISTRIBUTION OF TWO BEGOMOVIRUSES IN TOMATO-PRODUCTION AREAS OF VENEZUELA Introduction Begomoviruses are circular single strand DNA virus. Most begomoviruses, particularly those found in the New World, have a bipartite genome. The A component gives rise to a single virion-sense mRNA that codes for the coat protein (CP) while complementary transcription results in four mRNAs that code for proteins related to replication of the virus, replication-associated protein (Rep), replication enhancer protein (REn, former AC3), transcriptional activator protein (TrAP, former AC2) and AC4, which function is still unknown (Gutierrez, 2000; Lazarowitz, 1992). There is a non-coding region between the Rep and the CP, called intergenic region. This region has conserved sequences shared by both the A and B components of the bipartite begomoviruses. These conserved sequences are identified as a stem-loop motif, TATA box, forward, and inverted repeats, which are involved with replication and bi-directional transcription (Argello-Astorga and Ruiz-Medrano, 2001). The B component has the genes related to the movement of the virus. The A components of begomoviruses share a higher degree of sequence homology than the B components. This is due in large part to the presence of the coat protein (CP) and conserved sequences in the Rep. The amino acid sequence of the CP is more conserved than the remainder of the genome, this is probably because of the many constraints placed on the CP for viral structure, vector transmission, host specificity and possibly 79

PAGE 92

80 other unknown functions (Padidam et al. 1995). In contrast, the B component DNA sequences exhibit a high degree of sequence divergence, especially the hypervariable that is located between the beginning of the movement protein and the common region of the A and B components (Rani et al. 1996). Begomoviruses cause serious diseases in vegetables and fiber crops and have emerged as important viral pathogens in tropical and subtropical regions (Brown et al. 2001; Polston and Anderson, 1997). There is a need for accurate and simple methodologies for rapid and accurate detection of begomoviruses in order to develop disease management strategies. Biological assays to identify begomoviruses have been very difficult, since many begomovirus are not mechanically transmissible. Serology has been of limited use for begomovirus characterization because of the low titer of antigens, the coat protein amino acid sequence the target for serological probes is highly conserved, cross-reaction of antibodies with heterologous antigens, and developmental or environmental regulation of antigen production (Brown et al. 2001; Czosnek and Laterrot, 1997; Rojas et al. 1993; Wyatt and Brown, 1996). DNA-DNA hybridization assays, polymerase chain reaction (PCR), molecular cloning and DNA sequencing of viral genomes have become the tool of choice, allowing one to accurately identify the virus and to evaluate its relationship to other virus isolates (Padidam et al. 1995; Polston et al. 1989; Rybicki, 1994). Even though hybridization is less sensitive than PCR, this technique offers the possibility to evaluate many samples in less time and with less cost than PCR. Probes that include most of the intergenic region have been used to detect only isolates closely related to TYLCV-Is (Czosnek and Laterrot, 1997). Full-length sequences

PAGE 93

81 have been used to detect a wide range of begomoviruses, especially when the probes include the conserved regions (Czosnek and Laterrot, 1997). The A components of begomoviruses have been used as a general probe, and the B components have been used as specific probes because of the high degree of sequence divergence (Rani et al. 1996). Begomoviruses have been reported in Venezuela since 1963 (Debrot et al 1963). Four begomoviruses have been described affecting tomato crops in different states in Venezuela, Potato yellow mosaic virus-Venezuela (PYMV-VE) (Coutts et al. 1991; Roberts et al. 1986; Roberts et al. 1988), Tomato yellow mosaic virus (ToYMV) (Debrot et al. 1963; Lastra and Uzctegui, 1975; Morales et al. 2001), Potato yellow mosaic virus-VE strain tomato, and Tomato Venezuela virus (ToVEV) (Guzman et al. 1997). Recently, two new begomoviruses have been reported in a survey conducted in 1994 in the states of Trujillo and Mrida (Chapters 3 and 4). The begomoviruses were designated as 2.9-v and 57-v virus. To determine the distribution of these two new begomoviruses in tomato-growing areas of Venezuela, the following objectives were formulated: Determine the distribution of 57-v virus and 2.9-v virus in tomato samples from 10 states in Venezuela by hybridization with specific probes Confirm the results of hybridization using virus-specific probes by PCR followed by restriction analysis Confirm the results of hybridization using virus-specific probes by sequence analysis of selected plant samples. Materials and Methods Survey Tomato leaves with begomovirus-like symptoms were collected in 10 states in Venezuela. The sampling as well as the preservation of the samples has been previously described (Chapter 2).

PAGE 94

82 DNA Extraction, Begomovirus Detection by PCR, and Hybridization Genomic DNA was extracted from freeze-dried tissue (Rath et al. 1998). Previously, begomoviruses were detected by PCR using three sets of degenerate primers and one set of primers specific for Potato yellow mosaic virus (PYMV) (Chapter 2). Based on DNA sequence comparison and phylogenetic analysis, two new begomoviruses designated 2.9-v (from Trujillo) and 57-v (from Mrida) were reported (Chapter 2). For the hybridization, genomic DNA was spotted onto nylon membranes. Only those samples that were positive for at least one out of four PCR reactions, which amplify partial sequences of the A or B components (Chapter 2), were used for the hybridization assay. TE-8 (120 L) was placed in a chilled 1.5 mL tube and 40 L of genomic DNA were added to the tube. To denature the DNA, 20 L of 1 M NaOH (final concentration 0.1 M) were added and mixed well with a pipette, and incubated for 10 min at room temperature. Then, the DNA was neutralized by adding 20 L of 3 M sodium acetate (pH 4.5), inverting the tube several times and incubating for 10 min at room temperature. Nylon membrane, Hybond N + (0.45 m) (Amersham Pharmacia Biotech Inc, Pisscataway, NJ), was presoaked first in H 2 O for 5 min and then in TAE for 10 min. After that, the membrane was placed in a blotting manifold with a layer of water-soaked filter paper (Whatmann 3MM) beneath the nylon membrane. Then, 20 L of the DNA sample were loaded per well. Genomic DNA extracted from healthy tomato leaves was used as a negative control, and DNA extracts from samples 57-v and 2.9-v were used as positive controls. The membranes were placed on dry filter paper and crosslinked twice for 1200 s at 120 J/cm 2 in a UV Stratalinker 1800 (Stratagene, La Jolla, CA). The membranes were kept in a plastic bag at 4 o C until hybridization.

PAGE 95

83 PCR Probe The hypervariable region (~300 nt) of the B component was used as virus specific probes, this region was selected based on nucleotide sequence identity (NSI) value among partial sequences of 12 different DNA samples from Andean states (Chapter 2) as well as previous reports, in which the intergenic region, partial or full-length sequences of the B component have been used as specific probe to distinguish virus strains (Czosnek et al. 1990; Czosnek and Laterrot, 1997; Mansoor et al. 2003; Polston et al. 1993; Rani et al. 1996). Primer sets were designed to amplify this region of virus 2.9-v and 57-v using a Wisconsin Package Version 10.3, Accelrys (GCG), San Diego, CA. Primer set JAP176/JAP177 (5 -AGTAAAATTAGCCCGCCAG-3 and 5 -TGCCACTCCAAGCTCCTATC-3 ) amplifies the hypervariable region of the 2.9-v virus. Primer set JAP180/JAP181 (5 -AGGTTGCGCAGCTAAATG-3 and 5 -CTGCCATAAATTATCCCTTCTC-3 ) amplifies the hypervariable region of the 57-v virus. DNA amplification parameters for both sets of primers were as follows: 35 cycles of denaturation for 1 min at 94 o C, primer annealing for 1 min at 57 o C, and primer extension for 1 min at 75 o C, with an initial denaturing at 94 o C for 5 min and a final extension for 5 min at 72 o C. PCR reactions were carried out in a PE Applied Biosystems GeneAmp PCR System 9700 thermocycler (PE Applied Biosystems, Foster City, CA). All amplifications were performed in volumes of 25 L containing 10 mM Tris-HCl (pH 9), 50 mM KCl and 1% Triton X-100, 1.5 mM MgCl 2 150 M dNTPs (dCTP was

PAGE 96

84 labeled with, approximately 50 Ci dCTP, [ 32 P] 3000 Ci/mmol, aqueous solution), 0.5 mM spermidine, 1 M of each primer, 1 L of 1:100 dilution of plasmid DNA from clones 57-vB2 and 2.9-vB20 [purified-PCR products, using PBL1v2040/PCRc154 primers (Rojas et al. 1993), from samples 2.9-v and 57-v cloned in pGEM-T easy system], and 1.25 U Taq polymerase. The amplified DNA fragments were filtered through Sephadex columns, which were made with 0.5 mL plastic tubes with a 26.5 gauge-needle hole in the bottom and 1/5th of the tubes filled with glass wool. The columns were placed into a 1.5 mL tubes and 0.5 mL of Sephadex in TAE was added and spun down twice for 30 s, changing the 1.5 mL tubes each time. The probes 2.9-v and 57-v were read in a Bioscan/QC-4000 XER (Bioscan Inc. Washington, DC), boiled for 2 min, and kept on ice until hybridization. Hybridization Conditions The membranes were rolled between two layers of mesh and placed into roller bottles with 25 mL of prehybridization solution (pH 7.4) for 4 h at 65 o C. The prehybridization solution contained 0.25 M NaPO 4 0.001% Ficoll (w/v), 0.001% PVP-40 (w/v), 0.001% BSA (w/v), 1% SDS (w/v), 1 mM EDTA, 100 g/mL of denatured-salmon sperm DNA. The probe was added to 15 mL of the prehybridization solution. The membranes were hybridized for 24 h at 65 o C. The membranes were washed twice in each of the following solutions, 2X SSC for 5 min at room temperature, 0.2X SSC with 1% SDS for 15 min at 65 o C, and 0.1X SSC with 1% SDS for 5 min at room temperature. The membranes were exposed to Kodak BioMax MS film (Sigma-Aldrich St. Louis, MO) with two intensifying screens at o C for 72 h and 120 h for probe 57-v and 2.9v, respectively.

PAGE 97

85 Confirmation of Hybridization Results by PCR and Restriction Analysis All the samples that gave a positive hybridization signal with either probe, were tested by restriction analysis of the PCR products using degenerate primers and the virus specific primers for 2.9-v and 57-v. Specific primer sets: JAP176/JAP177, JAP122/JAP134 (Chapter 3) and JAP138/JAP139 (Chapter 3) for 2.9-v virus, were used to amplify a fragment of ~300 bp (hypervariable region); a fragment of ~1,045 bp that contains the end of the Rep (~384 bp), the TrAP, the REn and the 3 end of CP (~197 bp); and a fragment of ~654 bp that contains the 3 end of the movement protein (MP) (~22 bp), the intergenic region and the 3 end of the nuclear shuttle protein (NSP) (~428 bp), respectively. Specific primer sets: JAP180/JAP181, JAP136/JAP137 (Chapter 4) and JAP140/JAP141 (Chapter 4) for 57-v virus, were used to amplify a fragment of ~300 bp (hypervariable region); a fragment of ~446 bp that contains the 3 end of TrAP (~105 bp), the REn (~247 bp) and part of the CP (~199 bp); and a fragment of ~857 bp that contains the 3 end of the MP (~113 bp), the intergenic region (~56 bp) and the 3 end of the NSP (~688 bp), respectively. Primer sequences and PCR conditions were previously described (Chapter 3 and 4). PCR products of ~700 bp generated from primers PBL1v2040 and PCRc154 (Rojas et al. 1993) of samples that hybridized with 2.9-v probe were digested using four restriction enzymes, NdeI, EcoRI, MfeI and BglII. Same PCR products of ~700 bp of samples that hybridized with 57-v probe were digested using NdeI, EcoRI, MfeI and BglII. Digestion conditions were carried out according to the manufacturers instructions (New England Biolab, Inc. Beverly, MA)

PAGE 98

86 Southern Blot Analysis, Cloning and Sequence Determination Purified PCR products from primers PBL1v2040 and PCRc154 of five samples (103-v, 117-v, 261-v, 297-v and 307v) that hybridized to the 57-v probe were amplified using primers PBL1v2040 and PCRc154 (Rojas et al. 1993), a 700 bp PCR product was obtained and digested with EcoRI. The digestion product was run in agarose gel and viewed with Ethidium bromide. The gel was blotted on charged nylon membrane (0.45 micron) (Osmionics INC. Westborough, MA) using turboblotter TM (Schleicher & Schuell, Keene, New Hampshire). The 57-v probe was labeled with digoxigenin using PCR-Dig probe synthesis kit (Roche Diagnostics Corporation, Indianapolis, IN) and the primers JAP180/JAP181. Visualization of the probe-target hybrids was performed by chemiluminescent assay. The blot was exposed to blue medical X-ray film (Diagnostic Imaging Inc. Jacksonville, FL) for 30 s at room temperature. The amplified 700 bp products of samples 103-v, 117-v, 261-v and 307-v were cloned in pGEM-T Easy vector system I (Promega Corporation, Madison, WI) according to the manufacturers instructions. Transformation was performed using XL1-Blue MRF Supercompetent Escherichia coli cells (Stratagene, La Jolla, CA). The clones were screened by PCR using primers PBL1v2040 and PCRc154 and by digestion with EcoRI. The clones with EcoRI restriction site were sequenced at Ana-Gen Technologies, Inc. Atlanta, GA. Nucleic acid sequences were analyzed using a Wisconsin Package Version 10.3 [Accelrys (GCG), San Diego, CA]. Basic Local Alignment Search Tool (BLAST) was used to search for similarities between a query sequence and all the sequences in the database. GAP analysis was used to obtain the values of NSI.

PAGE 99

87 Results Confirmation of Samples with Hybridization Signal to the 2.9-v Probe by PCR and Restriction Analysis The 2.9-v virus was detected in Trujillo, Lara and Zulia states (Figure 5-1). Strong hybridization signals were detected in three samples from Trujillo and a very weak signal was observed in four samples from Lara and Zulia. The restriction analysis of these samples revealed that they had similar restriction sites as the 2.9-v virus, based on the fragment sizes from MfeI digestion, with the exception of sample 71-v from Zulia (Table 5-1). Primer set JAP176/JAP177 was unable to generate an expected size fragment from two of the seven samples that hybridized to the 2.9-v probe (Table 5-1). Primer set JAP176/JAP177 amplified an approximately 300 bp sequence from the hypervariable region, a region that was present in the virus specific probe. An expected size PCR product was obtained when specific internal primers for 2.9-v virus, JAP122/JAP134, were used for all of the samples, which hybridized to the 2.9-v probe. In some cases extra bands were also amplified (i.e. samples 92-v and 107-v). When specific primers (JAP138/JAP139) for the A component of virus 2.9-v were used, the expected PCR product was generated in all the samples with the exception of 92-v and 107-v samples (Table 5-1). Confirmation of Samples with Hybridization Signal to the 57-v Probe by PCR and Restriction Analysis The virus 57-v was detected by hybridization using specific probe in samples from five out of 10 states covered in this survey. The virus was detected in Mrida, Lara, Zulia, Gurico, and Aragua (Figure 5-1). The restriction analysis of Aragua samples revealed that eight of 11 samples had more than one virus based on the digestion with

PAGE 100

88 EcoR I. The samples 268-v and 270-v had BsrDI and NdeI restriction sites, respectively (Table 5-2). No samples had the same restriction sites as the 57-v sample, from which the 57-v probe was made (Table 5-2). Primers JAP180/JAP181 were able to amplify an expected size fragment from samples from Aragua and Gurico, which hybridized either strongly or weakly to the 57-v probe. However, they did not amplify any fragments from the samples from Lara and Zulia. When specific internal primers, JAP136/JAP137, for the A component were used, PCR products of the expected size were amplified but other bands were also amplified in 11 out of 24 samples with the exception of 2 samples from Mrida, and seven samples from Aragua (Table 5-2). The samples from Lara, Zulia and Gurico states did not show the expected PCR product when specific internal primers of the B component, JAP140/JAP141, were used. The samples from Aragua state presented the expected sized PCR products with those primers but multiple bands were also observed (Table 5-2). Southern Blot Analysis, Sequence comparison An EcoRI restriction site was present in the PCR product of 700 bp from primers PBL1v2040/PCRc154 of samples 103-v, 117-v, 261-v and 307-v samples, this digestion produced two bands, 500 and 200 bp, and also a band of 700 bp was observed on the gel (Figure 5-2-A). Therefore, it was evident that these samples had mixed infection. The 57-v probe labeled with digoxigenin hybridized only with the fragment of 700 bp (Figure 5-2B). The BLAST analysis of the sequences with EcoRI site showed high NSI values with PYMV sequences from different countries. In addition, the sequences had high value of NSI within them (Table 5-3) and with partial sequences from Andean states

PAGE 101

89 that were related with PYMV-VE (Chapter 2). In contrast, low values of NSI were observed between these sequences and the 57-v probe (Table 5-3). Discussion Both 2.9-v and 57-v probes were very specific, based on hybridization with genomic DNA, PCR, restriction and Southern blot analysis (Tables 5-1 and 5-2). High specificity of probes have been also reported for other begomoviruses Tomato leaf curl virus-[India], Tomato mottle virus-[Florida], Cassava mosaic virus-[India], Chino del tomate virus, Tomato leaf curl Barbados virus, Tomato yellow leaf curl virus-[Dominican Republic] (Mansoor et al. 2003; Polston et al.1993; Rani et al. 1996; Roye et al. 2000; Salati et al. 2002; Torres-Pacheco et al. 1996;) in which partial or full length of the A or B component has been used as a specific probe. The 2.9-v virus was limited to two samples from the states of Trujillo and Zulia. The partial sequences of the samples from Trujillo have been reported previously (Chapter 2) and they are considered isolates of the same virus (2.9-v virus) based on the high value (99 %) of NSI (Chapter 2). Weak hybridization signal with 2.9-v probe was observed on 92-v, 107-v, 45-v, 71-v samples, indicating the presence of very low 2.9-v virus titer or begomoviruses distinct from this 2.9-v virus. The expected PCR amplified fragment size was generated with specific primers to the hypervariable region and specific region of the A and B component of 2.9-v virus, with the exception of 92-v and 107-v samples. The 57-v virus was detected in one sample from Mrida, three samples from Gurico and eleven samples from Aragua. The samples, 261-v, 297-v, and 307-v were mixed infected with 57-v virus and PYMV based on restriction, southern blot, PCR

PAGE 102

90 Figure 5-1. Map of Venezuela indicating the states where tomato samples were collected for this research and the localities where 2.9-v and 57-v viruses were detected using specific probe labeled with 32 P. The 2.9-v and 57-v viruses are shown with red star and blue circle, respectively. States numbers: 1= Aragua, 2 = Barinas, 3 = Cojedes, 4 = Gurico, 5 = Lara, 6 = Mrida, 7 = Portuguesa, 8 = Tchira, 9 = Trujillo, 10 = Zulia.

PAGE 103

Table 5-1. PCR products and digestion fragments of samples with hybridization signal with 2.9-v probe PCRproduct fragment size (bp) Digestion-fragment size (bp) of PCR product with PBL1v2040/PCRc154 primers Sample No. State Hybridization Signal () JAP122/ JAP134 JAP138/ JAP139 JAP176/ JAP177 NdeI EcoRI MfeI BglII 2.7-v Trujillo 4 1050 650 300 500 + 200 500 + 200 2.8-v Trujillo 4 1050 650 300 500 + 200 700* + 450 + 250 2.9-v Trujillo 4 1050 650 300 700*+500 + 200 700* + 450 + 250 92-v Lara 1 1050* + 500* 650* 500 + 200 500 + 200 107-v Lara 1 1050*+MB 700+ 500+200 700*+ 500 + 200 45-v Zulia 1 1050 650 300 500 + 200 71-v Zulia 1 1050 650 300 91 = Hybridization-signal scale: 4 = very strong; 3 = strong; 2 = weak; 1 = very weak = Faint band = No PCR product or no restriction site MB = Multiple bands

PAGE 104

Table 5-2. PCR products and digestion fragments of samples with hybridization signal with 57-v probe PCRproduct fragment size (bp) Digestion-fragment size (bp) of PCR product with PBL1v2040/PCRc154 primers Sample No. State Hyb. Signal () JAP136/ JAP137 JAP140/ JAP141 JAP180/ JAP181 NdeI EcoRI FspI BsrDI 56-v Mrida 4 450 850 300 250 + 400 450 + 200 57-v Mrida 4 450 850 300 250 + 400 450 + 200 450 + 250 91-v Lara 1 750 + 250+ MB 350 + MB 500 + 200 92-v Lara 2 750 + 250 + MB 350 + MB 500 + 200 98-v Lara 2 450 + MB 350 + MB 500 + 200 100-v Lara 1 750 + 250 + MB 350 + MB 500 + 200 103-v Lara 2 750 + 450 + 250 + MB 350 + MB 500 + 200 116-v Zulia 1 750 + 450 + 250 + MB 350 + MB 500 + 200 117-v Zulia 1 350 500 + 200 125-v Zulia 2 750 + 450 + 250 + MB 350 + MB 500 + 200 297-v Gurico 1 750 + 450 + 250 350 + MB 300 700* + 500 + 200 298-v Gurico 1 450 1200 + MB 300 700* + 500 + 200 307-v Gurico 3 450 + 250 850 + MB 300 700* + 500 + 200 92 = Hybridization-signal (Hyb. signal) scale: 4 = very strong; 3 = strong; 2 = weak; 1 = very weak = Faint band = No PCR product or no restriction site MB = Multiple bands

PAGE 105

Table 5-2. Continued PCRproduct fragment size (bp) Digestion-fragment size (bp) of PCR product with PBL1v2040/PCRc154 primers Sample No. State Hyb. Signal () JAP136/ JAP137 JAP140/ JAP141 JAP180/ JAP181 NdeI EcoRI FspI BsrDI 259-v Aragua 3 450 + 250 850 + MB 300 + MB 700* + 500+200 260-v Aragua 3 450 + 250 850 + MB 300 + MB 700* + 500+200 261-v Aragua 4 450 850 + MB 300 + MB 700 + 500+200 262-v Aragua 3 450 + 250 850 + MB 300 + MB 700 + 500+200 268-v Aragua 2 450 850 + MB 300 + MB 750 + 500+200 700 + 600 + 100 270-v Aragua 3 450 + 250 850 + MB 300 + MB 700 + 500 + 200 700 + 500+200 271-v Aragua 2 450 850 + MB 300 + MB 500+200 272-v Aragua 2 450 + 250 850 + MB 300 + MB 700* + 500+200 278-v Aragua 1 450 850 + MB 300 + MB 750 + 500+200 281-v Aragua 3 450 850 + MB 300 + MB 700* + 500+200 283-v Aragua 3 450 850 + MB 300 + MB 700* + 500+200 93 = Hybridization-signal (Hyb. Signal) scale: 4 = very strong; 3 = strong; 2 = weak; 1 = very weak = Faint band = No PCR product or no restriction site MB = Multiple bands

PAGE 106

94 A B Figure 5-2. Detection of 57-v virus in digested-PCR product from primers PBL1v2040/PCRc154 of samples from four Venezuelan states: 103-v (Lara), 117-v (Zulia), 261-v (Aragua), 297-v and 307-v (Gurico). A) PCR product digested with EcoRI on agarose gel. B) Hybridization of digested PCR product using 57-v probe labeled with digoxigenin. Lane 1, 9: 57-v sample (positive control), lane 2, 10: H 2 O control, lane 3, 11: 103-v sample, lane 4, 12: 117-v sample, lane 5, 13: 261-v sample, lane 6, 14: 297-v sample, lane 7, 15: 307-v sample, lane 8: 100 bp ladder. The 57-v digested PCR product was diluted 1:10 for hybridization. 10 L of digested PCR product were loaded per lane. was detected is more than 370 Km distant to Aragua and Gurico. It was unexpected that these distant tomato production regions share some of the same begomoviruses. The distribution of the 57-v begomoviruses among these three states might be more likely by the movement of infected plants or by the movement of Bemisia tabaci biotype B. Similar results have been reported in Brazil, where new tomato viruses appear to be emerged simultaneously in different regions (Ribeiro et al. 2003). Rapid distribution of begomoviruses is expected when there is a high population of the insect vector in the Table 5-3. Nucleotide sequence identity (%) of four partial sequences of the B component with 57-v probe. Samples that hybridize with 57-v probe Probe Clones 117-vB3 261-vB6 307-vB3 57-v probe 103-vB3 90 95 92 42 117-vB3 89 93 54 261-vB6 92 54 307-vB3 54 57-vB2 100 In bold nucleotide sequence identity values < 89 %

PAGE 107

95 field. In addition recombination and pseudorecombination events can generate severe hybrids, as in the case of African cassava mosaic virus (ACMV) (Pita et al. 2001). In less than 15 years ACMV isolates and species are no longer geographically distinct, the viruses are spread throughout the continent (Legg and Thresh, 2000; Pita et al. 2001). Based on this experience, it is probable that the current situation of begomoviruses affecting tomato crops in Venezuela has changed, considering that the last survey was preformed in 1998. The probes might be used to monitor the two new begomoviruses in further surveys in tomato-production areas of Venezuela, especially those probes labeled with DIG because of the possibility reusing them during storage for 1 year at o C without affecting the results. Hybridization can be used to facilitate the reduction of sample numbers to be cloned and sequenced in future surveys of begomoviruses affecting tomato-production areas of Venezuela.

PAGE 108

CHAPTER 6 CONCLUSIONS Begomoviruses are considered a limiting factor in the production of tomato and other vegetables in tropical and subtropical regions of the Western and Eastern Hemispheres. The number of new begomoviruses in tomato crops has increased in the last 30 years. The situation is more critical with the introduction of the B biotype of Bemisia tabaci in the Americas. In Venezuela, the production of tomato is affected by begomoviruses, which have been reported since the middle of the 1960s. In some cases a 100% of the plants in commercial tomato fields of Lara and Aragua states had begomovirus-like symptoms. A survey was conducted from 1993 to 1998 to determine the diversity of viruses in tomato-production areas of Venezuela. The samples collected in this survey were used to determine the variability, characterization, and distribution of begomoviruses in tomato crops in Venezuela. The samples were collected from ten states: Aragua, Barinas, Cojedes, Gurico, Lara, Mrida, Portuguesa, Tchira, Trujillo, and Zulia. Since the tissue samples were dry and frozen, it was necessary to test two different DNA extraction protocols, Doyle and Doyle and Rath protocols. The Rath protocol was more reliable than Doyle and Doyle protocol, based on the results with four sets of primers. Three sets of degenerate primers and one set of specific primers for Potato yellow mosaic virus were used. Begomoviruses were detected in approximately fifty percent of tomato samples and in nine out of ten Venezuelan states. They were Aragua, Barinas, Gurico, Lara, 96

PAGE 109

97 Mrida, Portuguesa, Tchira, Trujillo, and Zulia states. These results confirm the prevalence of begomoviruses for the tomato industry in Venezuela. This is the first report of begomoviruses in tomato plants in the states of Barinas, Mrida, Tchira, and Trujillo. The samples from Andean states (Mrida, Tchira, and Trujillo) were selected to study variability of the tomato begomoviruses. Four different begomoviruses were identified in the Andean states, based on three types of analyses: GAP, iteron and replication iteron-related domain comparison, and phylogenetic analysis of the partial sequences of the A component. All four begomovirus sequence groups appear to be New World in origin. The first and the second groups had high values (> 90%) of nucleotide sequence identity with Tomato Venezuela virus and Potato yellow mosaic virus-strain tomato, respectively. Therefore, the sequences of these two groups could be strains of the two viruses mentioned above. The third and the fourth groups had less than 89% of nucleotide sequence identities with Tomato leaf curl Sinaloa virus and Potato yellow mosaic Trinidad virus-Trinidad & Tobago, respectively. The sequences of the third and fourth groups 3 and 4 were considered distinct begomoviruses because they had nucleotide sequence values lower than the cut off value that distinguishes species of begomovirus. A sample from each of these two groups was selected to characterize these potentially new viruses by molecular methods. The selected samples were 2.9-v and 57-v from Trujillo and Mrida states, respectively. The sequence of the A and B component of each virus was generated. This enabled the molecular characterization of two unique and new begomoviruses, designated as 2.9-v and 57-v viruses. The genome of both viruses was similar to those begomovirus from the New World. The values of nucleotide sequence identity and amino acid

PAGE 110

98 similarity confirmed that both 2.9-v and 57-v viruses are undescribed bipartite begomoviruses. The 2.9-v virus was most closely related to Dicliptera yellow mottle virus (DiYMoV) based on phylogenetic trees of the sequences of the A and the B components. DiYMoV and 2.9-v virus appeared to have a common ancestor, based on phylogenetic trees of the A and B components, or of the Rep gene alone. The 57-v virus is also related to DiYMoV based on phylogenetic trees of the sequence of the A and the B components. All the phylogenetic trees generated for 57-v virus sequences were not congruent. These results along with the analysis of forward and inverted repeats corroborate that 57-v virus is a unique bipartite begomovirus from the New world. Full-length clones of the A and B components for each virus were constructed using the overlapping-primers methodology. An experimental host range will be determined using biolistic inoculation. The distribution of each virus was determined using the hypervariable region of the B component as specific probes. The probes were very specific and were designed to detect the 2.9-v and 57-v viruses even when the samples were present in mixed infection with other begomoviruses. The results revealed that 2.9-v virus was detected in Trujillo, Lara and Zulia; and the 57-v virus was detected in Mrida, Gurico, and Aragua states. It is possible that the 2.9-v virus was spread to Trujillo and Zulia through transplant materials from Lara, which is a common source of transplants for the growers from Zulia and Trujillo states. In addition, begomoviruses have been reported in tomato crops since 1963 in Lara. In the case of the 57-v virus, it is hard to predict how the virus was wide spread to fields so distant from each other. The commercial tomato fields between Aragua and Gurico are close enough that whiteflies could spread the virus between these

PAGE 111

99 areas. However, Mrida fields are very distant from those in Aragua and Gurico. Therefore, it is not clear why the 57-v virus is so widely distributed. Infections of multiple begomoviruses in tomato were observed in this study. In some cases two different A components were present with one B component. This fact increases the possibility of recombination (exchange of DNA between different begomoviruses) and pseudorecombination (adoption of herterologous genomic components). Both events play an important role in the evolution of virus strains and species. That could explain part of the high variability of begomoviruses detected in this research. The presence of high populations of whiteflies reported since 1963 could also contribute to the high diversity of begomoviruses. The presence of wild species of Lycopersicon in tomato fields, as well other alternative host could also increase the variability of begomoviruses in tomato-growing areas in Venezuela. The genomic characterization of these two new begomoviruses and the generation of specific probes will allow the monitoring of the current prevalence of these two begomoviruses in tomato commercial areas of Venezuela. This information will also be used to develop a virus resistance program against these new begomoviruses using genetic engineering approaches.

PAGE 112

LIST OF REFERENCES Abouzid, A. M., Hiebert, E., and Strandberg, J. O. 1992a. Cloning, identification and partial sequencing of a new geminivirus infecting Brassicaceae. Phytopathology 82: A70 (Abstract). Abouzid, A. M., Polston, J. E., and Hiebert, E. 1992b. The nucleotide sequence of Tomato mottle virus, a new geminivirus isolated from tomato in Florida. J. Gen. Virol. 73: 3225-3229. Antignus, Y. 2000. Manipulation of wavelength-dependent behaviour of insects: an IPM tool to impede insects and restrict epidemics of insect-borne viruses. Virus Res. 71: 213-220. Anzola, D., and Lastra, R. 1985. Whiteflies population and its impact on the incidence of Tomato yellow mosaic virus in Venezuela. Phytopathology Z. 112: 363-366. Argello-Astorga, G. R., Guevara-Gonzles, R. G., Herrera-Estrella, L. R., and Rivera-Bustamante, R. F. 1994. Geminivirus replication origins have a group-specific organization of iterative elements: a model for replication. Virology 203: 90-100. Argello-Astorga, G. R., and Ruiz-Medrano, R. 2001. An iteron-related domain is associated to Motif 1 in the replication proteins of geminiviruses: identification of potential interacting amino acid-base pairs by a comparative approach. Arch. Virol. 146: 1465-1485. Arnal, E., Debrot, E., Marcano, R., and Montagne, A. 1993a. Fluctuacin poblacional de moscas blancas y su relacin con el mosaico amarillo del tomate en una localidad de Venezuela. Fitopatol. Venez. 6: 21-26. Arnal, E., Debrot, E. Ramos, F., Cermeli, M., Marcano, R., and Montagne, A. 1993b. Lista de moscas blancas (Homoptera: Aleyrodidae) y sus plantas hospederas en Venezuela. Florida Entomologist 76: 365-381. Baldauf, S. L. 2003. Phylogeny for the faint of heart: a tutorial. Trends Genet. 19: 345-351. Bendahmane, M., and Gronenborn, B. 1997. Engineering resistance against Tomato yellow leaf curl virus (TYLCV) using antisense RNA. Plant Mol. Biol. 33: 351-357. 100

PAGE 113

101 Brown, J. K., and Bird, J. 1992. Whitefly-transmitted geminiviruses and associated disorders in the Americas and the Caribbean Basin. Plant Dis. 76: 220-225. Brown, J. K., Frohlich, D. R., and Rosell, R. C. 1995. The sweetpotato or silverleaf whiteflies: Biotype of Bemisia tabaci or a species complex? Ann. Rev. Entomol. 40: 511-534. Brown, J. K. Idris, A. M., Torres-Jerez, I., Banks, G. K., and Wyatt, S. D. 2001. The core region of the coat protein gene is highly useful for establishing the provisional identification of begomoviruses. Arch. Virol. 146: 1581-1598. Brunetti, A., Tavazza, R., Noris, E., Lucioli, A., Accotto, G. P., and Tavazza, M. 2001. Transgenically expressed T-Rep of Tomato yellow leaf curl Sardina virus acts as a trans-dominant-negative mutant, inhibiting viral transcription and replication. J. Virol. 75: 10573-10581. Coutts, R. H. A., Coffin, R. S., Roberts, E. J. F., and Hamilton, W. D. O. 1991. The nucleotide sequence of the infectious cloned DNA components of Potato yellow mosaic virus. J. Gen. Virol. 72: 1515-1520. Czosnek, H., and Laterrot, H. 1997. A worldwide survey of tomato yellow leaf curl viruses. Arch. Virol. 142: 1391-1406. Czosnek, H., Navot, N., and Laterrot, H. 1990. Geographical distribution of Tomato yellow leaf curl virus. A first survey using specific DNA probe. Phytopath. Medit. 29: 1-6. Debrot, E. 1976. Estudios sobre el virus del grabado del tabaco en siembras de tomate en Venezuela. Agron. Trop. 26: 321-335. Debrot, E., Herold, F., and Dao, F. 1963. Nota preliminar sobre un "mosaico amarillento" del tomate en Venezuela. Agron. Trop. 13: 33-41. Debrot, E., and Centeno, F. 1985a. Infection natural de la papa en Venezuela con mosaico amarillo del tomate un geminivirus transmitido por moscas blancas. Agron. Trop. 35: 125-138. Debrot, E., and Centeno, F. 1985b. Ocurrenecia del virus mosaico de las euphorbiaceas infectando a Euphorbia heterophylla L en Venezuela. Agron. Trop.l 35: 5-12. Debrot, E., and Ordosgoitti, F. 1975. Estudios sobre un mosaico amarillo de la soya en Venezuela. Agron. Trop. 25: 435-449. Deng, D., McGrath, P. F., Robinson, D. J., and Harrison, B. D. 1994. Detection and differentiation of whitefly-transmitted geminiviruses in plant and vector insects by polymerase chain reaction with degenerate primers. Ann. Appl. Biol. 125: 327-336.

PAGE 114

102 Doyle, J. J., and Doyle, J. L. 1987. A rapid DNA isolation from small amount of fresh leaf tissue. Phytochem. Bull. 19: 11-15. Duan, Y. P., Powell, C. A., Webb, S. E., Purcifull, D. E., Hiebert, E. 1997. Geminivirus resistance in transgenic tobacco expressing mutated BC1 protein. Mol. Plant Microbe Interact. 10:610-623. Engel, M., Fernandez, O., Jeske, H., and Frischmuth, T. 1998. Molecular characterization of a new whitefly-transmissible geminivirus infecting tomato in Panama. J. Gen. Virol. 79: 2313-2317. Fauquet, C. M., Bisaro, D. M., Briddon, R. W., Brown, J. K., Harrison, B. D., Rybicki, E. P., Stenger, D. C., and Stanley, J. (Study Group Chair). 2003. Revision of taxonomic criteria for species demarcation in the family Geminiviridae, and an updated list of begomovirus species. Arch. Virol. 148: 405 Fondong, V. N., Pita, J. S., Rey, M. E. C., Kochko, A. de, Beachy, R. N., and Fauquet, C. M. 2000. Evidence of synergism between African cassava mosaic virus and a new double-recombinant geminivirus infecting cassava in Cameroon. J. Gen. Virol. 81: 287-297. Gafni, Y., and Epel, B. L. 2002. The role of host and viral proteins in intraand inter-cellular trafficking of geminiviruses. Phyl. Mol. Plant Pathol. 60: 231-241. Garridi-Ramirez, E. R., Sudarshana, M. R., and Gilbertson, R. L. 2000. Bean golden yellow mosaic virus from Chiapas, Mexico: Characterization, pseudorecombination with other bean-infecting geminiviruses and germ plasm screening. Phytopathology 90: 1224-1232. Ghanim, M., Morin, S., Zeidan, M., and Czosnek, H. 1998. Evidence for transovarial transmission of tomato yellow leaf curl virus by its vector, the whitefly Bemisia tabaci. Virology 240: 295-303. Gutierrez, C. 2000. DNA replication and cell cycle in plants: learning from geminiviruses. EMBO J. 19: 792-799 Guzman, P., Arredondo, C., Emmatty, D., Portillo, R., and Gilbertson, R. 1997. Partial characterization of two whitefly-transmitted geminiviruses infecting tomatoes in Venezuela. Plant Dis. 81: 312. Hanafi, A., Bouharroud, R., Murphy, B., and Enkegaard, E. 2002a. Evaluating a new non-toxic pesticide for integrated control of Bemisia tabaci in protected agriculture in Morocco. Bulletin OILB SROP. 25: 89-92. Hilje, L. 2002. Manejo de Bemisia tabaci en America Central y el Caribe: la experiencia de un decenio. Manejo integrado de plagas y Agroecologia 65: 102-108.

PAGE 115

103 Hill, J. E., Strandberg, J. O., Hiebert, E., and Lazarowitz, S. G. 1998. Asymmetric infectivity of pseudorecombinants of Cabbage leaf curl virus and Squash leaf curl virus: implications for bipartite geminivirus evolution and movement. Virology 250: 283-292. Hou, Y., Sanders, R., Ursin, V. M., Gilbertson, R. L. 2000. Transgenic plants expressing geminivirus movement proteins: abnormal phenotypes and delayed infection by Tomato mottle virus in transgenic tomatoes expressing the Bean dwarf mosaic virus BV1 or BC1 proteins. Mol. Plant Microbe Interact. 13: 297-308 Hunter, W.B., Hiebert, E., Webb, S. E., Tsai, J. H., and Polston, J. E. 1998. Location of geminiviruses in whitefly Bemisia tabaci (Homoptera: Aleyrodidae). Plant Dis. 82: 1147-1151. Idris, A. M., and Brown, J. K. 1998. Sinaloa tomato leaf curl geminivirus: Biology and molecular evidences for a new subgroup III virus. Phytopathology 88: 648-657. Kashina, B. D., Mabagala, R. B., and Mpunami, A. A. 2002. Reservoir weed hosts of tomato yellow leaf curl begomovirus from Tanzania. Arch. Phytopathol. Plant Prot. 35: 269-278. Konate, G., Barro, N., Fargette, D., Swanson, M. M., and Harrison, B. D. 1995. Occurrence of whitefly-transmitted geminiviruses in crop in Burkina Faso, and their serological detection and differentiation. Ann. Appl. Biol. 126: 121-129. Kunik, T., Salomon, R., Zamir, D., Navot, N., Zeidan, M., Michelson, I., Gafni, Y., and Czosnek, H. 1994. Transgenic tomato plants expressing the Tomato yellow leaf curl virus capsid protein are resistant to the virus. Biotechnol. 12: 500-504. Lapidot, M., and Friedmann, M. 2002. Breeding for resistance to whitefly-transmitted geminiviruses. Ann. Appl. Biol. 140: 109-127. Lastra, R., and Uzcategui, R. 1975. Viruses affecting tomatoes in Venezuela. Phytopathology Z 84: 253-258. Latrakul, P., Valverde, R. A., and Landry, A. D. 2000. Biological and molecular properties of begomovirus from Dicliptera sexangularis. Phytopathology 90: 723-729. Laufs, J., Jupin, I., David, C., Schumacher, S., Heyraud-Nitschke, F. Gronenborn, B. 1995 a Geminivirus replication: Genetic and biochemical characterization of Rep protein function, a review. Biochimie 77: 765-773. Laufs, J., Traut, W., Heyraud, F., Matzeit, V., Rogers, S. G., Schell, J., and Gronenborn, B. 1995 b In vitro cleavage and joining at the viral origin of replication by the replication initiator protein of Tomato yellow leaf curl virus. Proc. Natl. Acad. Sci. USA. 92:3879-3883.

PAGE 116

104 Lazarowitz, S. G. 1992. Geminiviruses: Genome and structure and gene function. Rev. Plant Sci. 11: 327-349. Legg, J.P., and Thresh, M. J. 2000. Cassava mosaic virus disease in East Africa: a dynamic disease in a changing environment. Virus Res. 71:135-149. Lotrakul, P., Valverde, R. A., Clark, C. A., Sim, J., and De La Torre, R. 1998. Detection of a geminivirus infecting sweet potato in United States. Plant Dis. 82: 1253-1257. Mansoor, S., Briddon, R. W., Zafar, Y., and Stanley, J. 2003. Geminivirus disease complex: an emerging threat. Trends Plant Sci. 8: 128-134. Mehta-Prem, A., Wyman, J., Nakhka, M. K., and Maxwell, D. P. 1994 Polymerase chain reaction detection of viruliferous Bemisia tabaci (Homoptera: Aleyrodidae) with two tomato-infecting geminiviruses. J. Econ. Entomol. 87: 1285-1290. Monci, F., Sanchez-Campos, S., Navas-Castillo, J., and Moriones, E. 2002: A natural recombinant between the geminiviruses Tomato yellow leaf curl Sardinia virus and Tomato yellow leaf curl virus exhibits a novel pathogenic phenotype and is becoming prevalent in Spanish populations. Virology 303: 317-326. Morales F. J., and Anderson, P. 2001. The emergence and distribution of whitefly-transmitted geminiviruses in Latin America. Arch. Virol. 146: 415-441. Morales, F., Lastra, R., Uzcategui, R., and Calvert, L. 2001. Potato yellow mosaic virus: a synonym of Tomato yellow mosaic virus. Arch. Virol. 146: 2249-2253. Moriones, E., and Navas-Castillos, J. 2000. Tomato yellow leaf curl virus, an emerging virus complex causing epidemics worldwide. Virus Res. 71: 123-124. Nava, A. 1999. Deteccin de diez virus en zonas productoras de tomate (Lycopersicon esculentum Mill) en Venezuela. J. Agric. Univ. P.R. 83: 83-88. Nava, A., Ochoa, F., Trujillo, G., Geraud, F., Hernndez, L., Lastra, R., and Rivas, G. 1996. Deteccin de virus en zonas productoras de tomate (Lycopersicon esculentum Mill) en Venezuela. I Estados Aragua y Zulia. Rev. Fac. Agron. (LUZ). 13: 285-292. Nava, A., Trujillo, G., Chirinos, D., and Rivero, G. 1997. Deteccin de virus en zonas productoras de tomate (Lycopersicon esculentum Mill) en Venezuela. II Estados andinos (Mrida, Tchira y Trujillo). Rev. Fac. Agron. (LUZ). 14: 611-624. Nava, A., Trujillo, G., Chirinos, D., and Rivero, G. 1998 a Deteccin de virus en zonas productoras de tomate (Lycopersicon esculentum Mill) en Venezuela. III Estados centro occidentales (Lara, Portuguesa, Barinas y Cojedes). Rev. Fac. Agron. (LUZ). 15: 23-29.

PAGE 117

105 Nava, A., Trujillo, G., Chirinos, D., and Rivero, G. 1998 b Deteccin de virus en zonas productoras de tomate (Lycopersicon esculentum Mill) en Venezuela. IV Estado Zulia. Rev. Fac. Agron. (LUZ). 15: 135-141. Navas-Castillo, J., Daz, J. A., Snchez-Campos, S., and Moriones, E. 1998. Improvement of the print-capture polymerase chain reaction procedure for efficient amplification of DNA virus genomes from plants and insect vectors. J. Virol. 75: 195-198. Padidam, M., Beachy, R. N., and Fauquet, C. M. 1995. Classification and identification of geminiviruses using sequence comparisons. J. Gen. Virol. 76: 249-263. Patel, V. P., Rojas, M. R., Paplomatas, E. J., and Gilbertson, R. L. 1993. Cloning biologically active geminivirus DNA using PCR and overlapping primers. Nucleic Acids Res. 21: 1325-1326. Paximadis, M., Idris, A. M, Torres-Jerez, I., Villareal, A., Rey, M. E. C., and Brown, J. K. 1999. Characterization of tobacco geminiviruses in the Old and New World. Arch. Virol. 144: 703-717. Pico, B., Dez, M. J., and Nuez, F. 1999. Improved diagnostic techniques for Tomato yellow leaf curl virus in tomato breeding programs. Plant Dis. 83: 1006-1012. Pita, J. S., Fondong, V. N., Sangar, A., Otim-Nape, G. W., Ogwal, S., and Fauquet, C. M. 2001. Recombination, pseudorecombination and synergism of geminiviruses are determinant keys to the epidemic of severe cassava mosaic disease in Uganda. J. Gen. Virol. 82: 655-665. Polston, J. E., and Anderson, P. K. 1997. The emergence of whitefly-transmitted geminiviruses in tomato in the Western hemisphere. Plant Dis. 81: 1358-1369. Polston, J. E. and Bois, D. 1998. Occurrence of a strain of Potato yellow mosaic geminivirus infecting tomato in the eastern Caribbean. Plant Dis. 82: 126. Polston, J. E., McGovern, R. J., Stanley, P. A., Schuster, D. J. 1993. Host range of Tomato mottle virus, a new geminivirus infecting tomato in Florida. Plant Dis. 77: 1181-1184. Polston, J. E., Perring, T. M., and Dodds, J. A. 1989. Nucleic acid probes for detection and strain discrimination of cucurbit geminiviruses. Phytopathology 79: 1123-1126. Polston, J. E., Yang, Y., Sherwood, T., Bucher, C., Freitas-Astua, J., and Hiebert, E. 2001. Effective resistant to Tomato yellow leaf curl virus (TYLCV) in tomato and tobacco mediated by a truncated TYLCV Rep gene. 3 rd International Geminivirus Symposium. Norwich. UK. P. 82.

PAGE 118

106 Rampersad, S. N., and Umaharan, P. 2003. Detection of begomoviruses in clarified plant extracts: A comparison of standard, direct-binding, and immunocapture polymerase chain reaction techniques. Phytopathology 93: 1153-1157. Rani, R. V., Karthikeyan, A. S., Anuradha, S., and Veluthambi, K. 1996. Genome homology among geminiviruses infecting Vigna, cassava, Acalypha, Croton and Vernonia. Curr. Sci. 70: 63-69. Rath, P., Rajaseger, G., Goh, C. J., and Kumar, P. P. 1998. Phylogenetic analysis of Dipterocarps using Random Amplified Polymorphic DNA markers. Ann. Bot. 82: 61-65. Ribeiro, S. G., Ambrozevcius, L. P., Avila, A. C., Bezerra, I. C., Calegario, R. F., Fernandes, J. J., Lima, M. F., Mello, R. N., Rocha, H., and Zerbini, F. M. 2003. Distribution and genetic diversity of tomato-infecting begomoviruses in Brazil. Arch. Virol. 148: 281-295. Roberts, E. J. F., Buck, K. W., and Coutts, R. H. A. 1986. A new geminivirus infecting potatoes in Venezuela. Plant Dis. 70: 603. Roberts, E. J. F., Buck, K. W., and Coutts, R. H. A. 1988. Characterization of Potato yellow mosaic virus as a geminivirus with a bipartite genome. Intervirology 29: 162-169. Rojas, M. R., Gilbertson, R. L., Russell, D. R., and Maxwell, D. P. 1993. Use of degenerate primers in the polymerase chain reaction to detect whitefly-transmitted geminiviruses. Plant Dis. 77: 340-347. Rosell, R., Torres-Jerez, I., and Brown, J. K. 1999. Tracing the geminivirus-whitefly transmission pathway by polymerase chain reaction in whitefly extracts, saliva, hemolymph, and honeydew. Phytopathology 89: 239-246. Roye, M E., Henry, N. M., Burrell, P. D., McLaughlin, W. A., Nakhla, M. K., and Maxwell, D. P.2000. A new tomato-infecting begomovirus in Barbados. Plant Dis. 84: 1342. Rybicki, E. P. 1994. A phylogenetic and evolutionary justification for three genera of Geminiviridae. Arch. Virol. 139: 49-78. Salas, J., y Arnal, E. 2001. Bemisia tabaci (Gennadius, 1899) biotipo B, primer registro para Venezuela utilizando RAPDsPCR. Entomotrpica 16: 181-185. Salati, R., Nahkla, M. K., Rojas, M. R., Guzman, P., Jaquez, J., Maxwell, D. P., and Gilbertson, R. L. 2002. Tomato yellow leaf curl virus in the Dominican Republic: Characterization of an infectious clone, virus monitoring in whiteflies, and identification of reservoir hosts. Phytopatology 92: 487-496.

PAGE 119

107 Schuster, D. and Polston, J. E. 1999.Whitefly management guide: Tomato yellow leaf curl virus. Whitefly management guide. Summer: 6-7. Sinisterra, X. H., Polston, J. E., Abouzid, A. M., Hiebert, E. 1999. Tobacco plants transformed with a modified coat protein of tomato mottle begomovirus show resistance to virus infection. Phytopatology 89: 701-706. Stanley, J., Frischmuth, T., and Ellwood, S. 1990. Defective viral DNA ameliorates symptoms of geminivirus infection in transgenic plants. Proc. Natl. Acad. Sci. USA. 87: 6291-6295. Sunter, G., and Bisaro, D. M. 1991. Transactivation in a geminivirus: AL2 gene product is needed for coat protein expression. Virology 180: 416-419. Sunter, G., Hartitz, M. D., Hormudzi, S. G., Brough, C. L., and Bisaro, D. M. 1990. Genetic analysis of Tomato golden mosaic virus: ORF AL2 is required for coat protein accumulation while ORF AL3 is necessary foe efficient DNA replication. Virology 179: 69-77. Timmermans M. C. P., Das, O. P., and Messing, J. 1994. Geminiviruses and their uses as extrachromosomal replicons. Annu. Rev. Plant Physiol. Plant Mol. Biol. 45: 79-112. Torres-Pacheco, I., Garzon-Tiznado, J. A., Brown, J. K., Becerra-Flora, A., and Rivera-Bustamante, R. F. 1996. Detection and distribution of geminiviruses in Mexico and the Southern United States. Phytopathology 86: 1186-1192. Umaharan, P., Padidam, M., Phelps, R. H., Beachy, R. N., and Fauquet, C. M. 1998. Distribution and diversity of geminiviruses in Trinidad and Tobago. Phytopathology 88: 1262-1268. Urbino, C., Polston, J. E., Patte, C. P., Caruana, M-L. 2003. Characterization and genetic diversity of Potato yellow mosaic virus from the Caribbean. Arch. Virol. (in press). Uzctegui, R., and Lastra, R. 1978. Transmission and physical properties of the causal agent of Mosaico amarillo del tomate (Tomato yellow mosaic). Phytopathology 68: 985-988. Van den Heuvel, J. F. J. M., Hogenhout S. A., and van der Wilk, F. 1999. Recognition and receptors in virus transmission by arthropods. Trends Microbiol. 7: 71-76. Vidavsky, F., and Czosnek, H. 1998. Tomato breeding lines resistant and tolerant to Tomato yellow leaf curl virus issued from Lycopersicon esculentum. Phytopathology 88: 910-914. Wyatt, S. D., and Brown, J. K. 1996. Detection of subgroup III geminivirus isolates in leaf extracts by degenerate primers and polymerase chain reaction. Phytopathology 86: 1288-1293.

PAGE 120

BIOGRAPHICAL SKETCH Alba Ruth Nava was born on November 6, 1959, to Francisco Nava and Aida Fereira de Nava, in Maracaibo, Venezuela. She received a bachelors degree in agronomic engineering from the Facultad de Agronoma, at Universidad del Zulia (LUZ). She obtained her masters degree in plant breeding at Universidad Central de Venezuela (UCV) in 1986. From 1986 to 1989, she worked as assistant researcher in oil crop breeding at the Centro Nacional de Investigaciones Agropecuarias. From 1989 to 1991, she worked as assistant professor in the Genetics Department at UCV. From 1992 to 1998, she initiated a tomato virus research program at LUZ. In 1999, she obtained a scholarship from LUZ to continue her graduate studies toward a Doctor of Philosophy degree in the Plant Pathology Department at the University of Florida, where she worked under the guidance of Drs. Jane Polston and Ernest Hiebert. Upon completion of her Ph. D. degree, Alba will be returning to Venezuela, where she will work in tomato-virus research at LUZ. 108


Permanent Link: http://ufdc.ufl.edu/UFE0002181/00001

Material Information

Title: Detection, Characterization, and Distribution of Begomoviruses Infecting Tomatoes in Venezuela
Physical Description: Mixed Material
Copyright Date: 2008

Record Information

Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
System ID: UFE0002181:00001

Permanent Link: http://ufdc.ufl.edu/UFE0002181/00001

Material Information

Title: Detection, Characterization, and Distribution of Begomoviruses Infecting Tomatoes in Venezuela
Physical Description: Mixed Material
Copyright Date: 2008

Record Information

Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
System ID: UFE0002181:00001


This item has the following downloads:


Full Text












DETECTION, CHARACTERIZATION, AND DISTRIBUTION OF
BEGOMOVIRUSES INFECTING TOMATOES IN VENEZUELA


















By

ALBA RUTH NAVA


A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY

UNIVERSITY OF FLORIDA


2003

































Copyright 2003

by

Alba Ruth Nava















ACKNOWLEDGMENTS

I thank Dr. Jane Polston (chairman of the supervisory committee) for her support

and guidance during my graduate studies. I also thank Drs. Ernest Hiebert, Gail Wisler,

Susan Webb and Maria Gallo-Meagher (who served on the supervisory committee) for

their suggestions and contributions. Special thanks go to Christian Patte, Tracy

Sherwood, and Kristin Beckham for their technical assistance and friendship.

I thank my friends Yolanda, Denise, Abby, Zenaida, Maritza, Belkys, Francisco,

Olimpia, Adriana, Marlene, Juliana, Gustavo, Jorge, Basma, Abdulwahid, Hamed,

Vicente, Julia, Renato, Manjunath, Lucious, and Eugene for their support and constant

encouragement during the last 4 years.

I thank the Universidad del Zulia, for providing financial support to my studies.
















TABLE OF CONTENTS
Page

A C K N O W L E D G M E N T S ......... ...................................................................................... iii

LIST OF TABLES ....... .. ......................................... ...... ........... vii

LIST OF FIGURES ......... ........................................... ............ ix

ABSTRACT .............. .................. .......... .............. xi

CHAPTER

1 LITERATURE REVIEW ........................................................... ..................1

B egom ovirus D description .................................................................. .. ..................
Begomovirus Genome Structure and Viral-Encoded Proteins..............................1
Criteria for Species D em arcation........................................... ........................... 6
Impact of Begomoviruses on Agriculture.................. ..................6
M anagem ent of Begom viruses ............................................................................8
Detection and Discrimination of Begomoviruses..................................10
V ariability of B egom ovirus Species...................................................... ..................10
Variability of African cassava mosaic virus and Tomato yellow mosaic virus in
the O ld W orld .................. ............... ...... ................. ................11
Variability of Potato yellow mosaic virus in the New World .............................12
Begomoviruses in Tomato Crops in Venezuela .....................................................13
B background and O objectives .................................................................................. 14

2 DETECTION AND VARIABILITY OF BEGOMOVIRUSES IN TOMATO-
PRODUCTION AREAS OF ANDEAN STATES, VENEZUELA........................16

Introduction ............... .... ............ ............ .. ......................... 16
M materials and M methods ....................................................................... .................. 18
Survey ............... ........... .......................... ........................... 18
D N A extraction .................. ........................................ ........ ........ . ........... 19
Polymerase Chain Reaction (PCR) .................................... .. ............... 21
Restriction Enzyme Analysis of Purified PCR Products................................22
Cloning and DNA Sequence Determination ....................................... .......... 22
Com prison of DN A Sequence .................................. ............... ............... 23
P hy log en etic A n aly sis .............................................................. .....................2 3









R e su lts ............................ .... ........................ ........................................2 4
Comparison of DNA Extraction Protocols .................................... 24
PCR Amplification of Begomovirus Sequences from Field Samples.................25
Restriction Enzyme Analysis of Purified PCR Product................. ............28
D N A Sequence C om parison.......................................... ........... ............... 30
Phylogenetic A analysis ............................................... ........ .. ............ .... 35
D discussion ..................................... .................. ............... .......... 38

3 CHARACTERIZATION OF A NEW BEGOMOVIRUS FROM THE STATE OF
TRUJILLO, VENEZUELA ............ ..... ........ ............................... 43

Introdu action .....................................................................................................4 3
M materials and M methods .............. .......... ...................................................................44
Plant Sample and Extraction of Genomic DNA ....... .................................44
Obtaining Full-Length Sequences ............................................ ............... 45
O obtaining Infectious C lones ........................................................... ....... ........ ..47
Comparison of DNA and Amino Acid Sequences............................................48
P hy log en etic A n aly sis .............................................................. .....................4 8
R e su lts ............... .. ............ ....................... ................. ................ 5 1
Full-L ength Sequencing ................................................. .......... ............... 51
Comparison of DNA and Amino Acid Sequences.............................................51
Phylogenetic analysis ................................................ .............................. 53
D iscu ssio n ...................................... ................................................. 5 4

4 CHARACTERIZATION OF A NEW BEGOMOVIRUS FROM THE STATE OF
M ERID A V EN EZUELA .................................................. ............................. 61

Intro du action .....................................................................................................6 1
M materials and M methods .............. .......... ...................................................................63
Plant Sample and Extraction of Genomic DNA ....... .................................63
Obtaining Full-Length Sequences ............................................ ............... 63
Obtaining Full-Length Clones.... ................. ...... ...............65
Comparison of DNA and Amino Acid Sequences.................. ..................67
P hy log en etic A n aly sis .............................................................. .....................67
R esu lts................. ... ....... ........................... ................. ................. 6 8
Full-Length Sequencing and Cloning....................................... ............... 68
Comparison of DNA and Amino Acid Sequences............................................70
P hylogenetic A analysis ............................................... ........ .. ...... ............7 1
D iscu ssio n ...................................... ................................................. 7 2

5 DISTRIBUTION OF TWO BEGOMOVIRUSES IN TOMATO-PRODUCTION
AREA S OF VEN EZUELA ............................................... ............................. 79

Intro du action .....................................................................................................7 9
M materials and M methods ....................................................................... ..................8 1
S u rv ey ............... .. ....... .......................................................... .. ........ ..............8 1
DNA Extraction, Begomovirus Detection by PCR, and Hybridization .............82


v









P C R P ro b e ..................................................................................................... 8 3
H ybridization C conditions ............................ ............... .. .................. .... 84
Confirmation of Hybridization Results by PCR and Restriction Analysis .........85
Southern Blot Analysis, Cloning and Sequence Determination........................86
R e su lts ............. ... ...... ...... .. ... .. .. ...................... ...................... 8 7
Confirmation of Samples with Hybridization Signal to the 2.9-v Probe by PCR
and R estriction A analysis ............................ ............ .. ..................... .... 87
Confirmation of Samples with Hybridization Signal to the 57-v Probe by PCR
and R estriction A analysis ............................................................................ 87
Southern Blot Analysis, Sequence comparison................................................88
D iscu ssion ............... .................................... ............................89

6 CONCLUSIONS ......................................... .......... ...............96

LIST OF REFEREN CE S ......... .................................. ........................ ............... 100

B IO G R A PH ICA L SK ETCH ......... ................. ...................................... .....................108
















LIST OF TABLES


Table p

2-1. Acronyms and accession numbers of known begomoviruses used in the
phylogenetic analysis of partial sequences of samples from Andean states ............24

2-2. Comparison of two DNA extraction protocols through PCR results of samples
from Andean states, Venezuela using 4 primer sets to amplify partial sequence
of A and B components of begomoviruses ........... ........................................26

2-3. Polymerase chain reaction results using degenerate primers to amplify A and B
components of begomoviruses using Doyle and Doyle DNA extraction of
Venezuelan samples from tomato-production areas ............................................. 27

2-4. Expected restriction-fragment sizes (bp) in the fragment amplified by primer set
PAR1c496/PALlv1978 of the A component of known begomoviruses .................28

2-5. Restriction-fragment sizes in the fragment amplified by primer set
PAR1c496/PAL1v1978 in the A component of begomoviruses from Andean
states, in V enezuela ............................. ......... ....... ........... 29

2-6. Expected restriction-fragment sizes in the fragment amplified by primer set
PBL1v2040/PCRcl54 of the B component of known begomoviruses .................. 30

2-7. Restriction-fragment sizes in the fragment amplified by PBLlv2040/PCRc154 of
the B component of begomoviruses from Andean states, in Venezuela.................31

2-8. Nucleotide sequence identity (%) between partial sequences from PCR using
primer set PAR1c496/PALlv1978 of the A component of begomoviruses from
Andean states and four known begomoviruses............. .............. ............... 34

2-9. Motifs of the intergenic region and Iteron-related domain of Replication protein
(Rep IRD) of cloned sequences from Andean states and eight known
b egom oviru ses ...................................... ............................. ................ 36

2-10. Nucleotide sequence identity (%) between partial sequences from PCR using
primer set PBLlv2040/PCRc154 of the B component of begomoviruses from
Andean states and four known begomoviruses....... ... ........................................ 37

3-1. Specific primers designed to determine the full-length sequence of a new
begomovirus from Trujillo, Venezuela. ........... .............................. 47









3-2. Acronyms and accession numbers of known begomoviruses used for nucleic acid
and amino acid sequence comparison and phylogenetic analysis of 2.9-v virus. ....49

3-3. Percent of nucleic acid identity between 2.9-v virus sequence and 15 known
begom viruses ............................................................... .. .... ......... 52

3-4. Percent of similarity of amino acid sequences between proteins of 2.9-v virus and
proteins of 15 known begom oviruses............... ............................. ............... .53

4-1. Specific primers designed to determine the full-length sequence of 57-v virus from
M erida, V enezuela. ........................ ........................ .. .. .. ..... .... ........... 66

4 -2. Acronyms and accession numbers of known begomoviruses used for nucleic acid
and amino acid sequence comparison and phylogenetic analysis of 57-v virus. .....68

4-3. Percent of nucleotide sequence identity between 57-v virus sequence and
14-know n begom viruses .............................................. .............................. 71

4-4. Percent of similarity of amino acid sequences between proteins of 57-v virus and
proteins of 14-known begom viruses ........................................... ............... 72

5-1. PCR products and digestion fragments of samples with hybridization signal with
2 .9 -v p ro b e ...............................................................................................................9 1

5-2. PCR products and digestion fragments of samples with hybridization signal with
5 7 -v p rob e ...........................................................................92

5-3. Nucleotide sequence identity (%) of four partial sequences of the B component
w ith 57-v probe. .................................................... ................. 94
















LIST OF FIGURES


Figure p

1-1. Genom e structure of begom viruses. .......................................................................... 3

1-2. Hypothetical model for the functional organization of the replication origin in
begom oviru ses ................................................................................................... 5

2-1. Map of Venezuela indicating the states where plant tissue samples were
collected. ...................................................................20

2-2. Phylogenetic tree constructed based on the nucleotide sequence of the entire
fragment sequenced from amplification using primer set PAR1c496/PALlv1978
for the A com ponent ........ .... ........................ ........ ....... .. ...... ........ .... 39

2-3. Phylogenetic tree constructed based on the nucleotide sequence of the entire
fragment sequenced from amplification using primer set PBLlv2040/PCRc154
for the B com ponent .................. ............................ ........ .. ........ .... 40

3-1. Symptoms of sample 2.9-v collected in Trujillo state, Venezuela.............................45

3-2. Genome organization of 2.9-v sequence a bipartite begomovirus from Trujillo
state, Venezuela..........................................................................50

3-3. Phylogenetic tree based on the complete nucleotide sequence of the A component
of the 2.9-v sample from Trujillo state and fifteen bipartite begomoviruses...........55

3-4. Phylogenetic tree based on the complete nucleotide sequence of the B component
of the 2.9-v sample from Trujillo state and fifteen bipartite begomoviruses...........56

3-5. Phylogenetic tree based on the complete coat protein nucleotide sequence of the
2.9-v sample from Trujillo state and fifteen bipartite begomoviruses .....................57

3-6. Phylogenetic tree based on the complete replication-associatedprotein nucleotide
sequence of the 2.9-v sample from Trujillo state and fifteen bipartite
begom viruses ............................................................... .. .... ......... 58

4-1. Symptoms of sample 57-v collected in Merida state, Venezuela..............................64

4-2. Genome organization of 57-v sequence a bipartite begomovirus from Merida
state, V enezu ela .............................................................................. ............... 69









4-3. Phylogenetic tree based on the complete nucleotide sequence of the A component
of the 57-v sample from Merida state and fourteen bipartite begomoviruses..........73

4-4. Phylogenetic tree based on the complete nucleotide sequence of the B component
of the 57-v sample from Merida state and fourteen bipartite begomoviruses..........74

4-5. Phylogenetic tree based on the complete coat protein nucleotide sequence of the
57-v sample from Merida state and fourteen bipartite begomoviruses....................75

4-6. Phylogenetic tree constructed based on the complete replication-associatedprotein
nucleotide sequence of the 57-v sample from Merida state and fourteen bipartite
begom viruses ............................................................... ... ... ........ 76

5-1. Map of Venezuela indicating the states where tomato samples were collected for
this research and the localities where 2.9-v and 57-v viruses were detected using
specific probe labeled w ith a 32 ........................................ ......................... 90

5-2. Detection of 57-v virus in digested-PCR product from primers
PBLlv2040/PCRc154 of samples from four Venezuelan states:
103-v (Lara), 117-v (Zulia), 261-v (Aragua), 297-v and 307-v (Guarico) ..............94















Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy

DETECTION, CHARACTERIZATION, AND DISTRIBUTION OF
BEGOMOVIRUSES INFECTING TOMATOES IN VENEZUELA

By

Alba Ruth Nava

December 2003

Chair: Jane E. Polston
Major Department: Plant Pathology

Begomoviruses are members of the geminiviridae family. They are mono- or

bipartite single-stranded DNA plant viruses that are transmitted by whiteflies. This genus

belongs to the geminivirudae family. Begomoviruses are a limiting factor in the

production of vegetables and other crops in the tropics and subtropics worldwide. To

identify begomoviruses in tomatoes in Venezuela, symptomatic leaves were collected in

ten states from 1993 to 1998. Detection of Begomovirus was performed by polymerase

chain reaction (PCR) with four sets of primers. Begomoviruses were detected in 50% of

the samples. Samples from Andean states (Trujillo, Merida, and Tachira) were selected to

examine variability of begomovirus. Resulting sequences were placed in four groups

based on BLAST, GAP, and phylogenetic analyses. Two groups were considered new

begomoviruses based on low (<89%) nucleotide sequence identity and phylogenetic

analyses. The viruses were designated 2.9-v (from Trujillo) and 57-v virus (from

Merida). The full-length sequences of the A and B component of both viruses were









obtained and compared with known begomoviruses. The 2.9v virus was closely related to

Dicliptera yellow mottle virus (DiYMoV). However, the 57-v virus was a unique virus,

distantly, but most closely related, to DiYMoV. The hypervariable regions of the B

components of both viruses were used as specific probes to determine the distribution of

these viruses. The 2.9-v virus was detected in samples from Trujillo, Lara, and Zulia. The

57-v virus was detected in samples from Merida, Guarico, and Aragua. The distribution

of these viruses in such distant states could be explained by movement of infected

transplants or by the whitefly vector. We confirmed the high variability of begomoviruses

in the Andean states. We expect to find more distinct begomoviruses in samples from

other states included in the survey. The genomic characterization of these two new

begomoviruses and the generation of specific probes will allow the monitoring of the

current prevalence of these two begomoviruses in tomato commercial areas of Venezuela.

This information will also be used to develop a virus resistance program against these

new begomoviruses using genetic engineering approaches.














CHAPTER 1
LITERATURE REVIEW

Begomovirus Description

Begomoviruses are small (ca. 18-30 nm) plant viruses with single-stranded

circular DNA genomes that are encapsidated in twinned quasi-icosahedral particles. They

belong to the Geminiviridae family.

Begomoviruses are transmitted by whiteflies, and infect dicotyledonous plants;

their genomes can be mono- or bipartite (Lazarowitz, 1992). They cause significant and

often total yield losses of important food and industrial crops in tropical and subtropical

regions of the Western and Eastern Hemispheres (Morales and Anderson, 2001; Navas-

Castillo et al. 1998; Polston and Anderson, 1997). High incidences of begomoviruses are

associated with high populations of whiteflies and serious losses in several crops in the

Americas and the Caribbean Basin (Brown and Bird, 1992; Morales and Anderson, 2001;

Polston and Anderson, 1997).

Begomovirus Genome Structure and Viral-Encoded Proteins

Begomoviruses are mostly bipartite, but some Old World begomoviruses are

monopartite. Bipartite begomoviruses have two components, designated A and B. Each

component has -2,600 nt. The genes on the A component are involved in encapsidation

and replication, whereas the genes on the B component are involved in the movement of

virus through the plant, host range, and symptom expression (Figure 1-1) (Gafni and

Epel, 2002; Lazarowitz, 1992). One of the five genes in the A component, the coat

protein (CP) gene, is transcribed in the viral sense or clockwise direction. The other four









genes replication-associated protein (Rep), transcriptional activator protein (TrAP),

replication enhancer (REn), and AC4 are transcribed in complementary sense or

counterclockwise direction (Lazarowitz, 1992). The two sets of genes overlap and are

separated by an intergenic region (IR), which begins with the start codon of the Rep and

ends with the start codon of the CP. This region does not encode any protein and its

sequence varies widely among begomoviruses, except that there is a conserved GC-rich

inverted repeat sequence, which has the potential to form a stem-loop structure (-30 nt)

with the invariant nanomeric TAATATT()AC sequence or loop of the stem-loop

structure. The nanomeric sequence contains the initiation site () of rolling circle DNA

replication (Gutierrez, 2000; Laufs et al. 1995a), the TATA box, and the forward and

inverted repeats. In bipartite begomoviruses, the IR also contains an identical sequence of

-200 nt in the A and the B components called the common region (CR) (Lazarowitz,

1992). The CR sequence is different among different begomoviruses and is used to

identify the A and B components of the same virus.

The CP is required for encapsidation of progeny virions, vector transmission,

virion structure, and host specificity. For bipartite begomoviruses, the CP is not required

for either local or systemic viral spread. In contrast, in all monopartite begomoviruses,

the CP is essential for viral spread (Gafni and Epel, 2002).

The Rep is the only gene essential for replication, being required for transcription

of both A and B components (Arguello-Astorga et al. 1994). Begomoviruses replicate in

the nucleus of infected cells through a double-stranded DNA intermediate via a rolling

circle mechanism. The Rep protein has two functional targets or DNA elements in the

begomovirus origin of replication: 1) The nanomeric sequence, where the Rep introduces


















NSP
TrAP NSP


REn



B

IR


C4 V2
Re
-~2800 nt CP

TrAP


REn


Figure 1-1. Genome structure of begomoviruses. A) Bipartite begomovirus. B)
Monopartite bigomoviruses. IR = Intergenic region, CRA = Common region
A, CRB = Common region B, CP = Coat protein, TrAP = Transcriptiona-
activator protein, REn = Replication-enhancer protein, AC4 = AC4 protein,
Rep = Replication- associated protein, C4 = Symptoms expression V2 =
Movement protein MP = Movement protein, NSP = Nuclear- shuttle protein.









a site-specific nick to initiate virus replication for the rolling circle mechanism, and 2) a

tandemly repeated motif located at variable distances from the conserved hairpin

sequence, which is bound specifically by its cognate Rep protein. This motif functions as

a major recognition element of the replication origin in begomoviruses. These cis-acting

elements belong to a series of iterate DNA motifs called iterons (Arguello-Astorga and

Ruiz-Medrano, 2001).

A functional organization of the replication origin in begomoviruses has been

hypothesized (Figure 1-2) (Arguello-Astorga et al. 1994). The Rep protein binds to the

iterons associated with the TATA box, where a TATA-binding protein was previously

bound. A transcription factor binds to a cis-regulatory element, associated with the

5' border of the stem-loop sequence, and creates a nucleosome-free region in its

neighborhood. The transcriptional factor interacts with the TATA binding protein, by

means of its activation domain, looping the intervening DNA. This event would place the

stem-loop structure in an accessible position so that the Rep complex can nick the viral

(+) strand in the loop of the hairpin structure. The stem-loop structure may acquire a

cruciform structure as a consequence of the interactions with the transcription factor

or/and Rep (Arguello-Astorga et al. 1994).

TrAP is a transactivator of the expression of the CP and the nuclear shuttle protein (NSP)

genes (Sunter et al. 1990; Sunter and Bisaro, 1991). The TrAP along with the two

proteins encoded by the B component are indirectly involved in the systemic movement

of the virus through the plant (Gafni and Epel, 2002). The REn is not essential for viral

replication. However, viral DNA replicates at higher levels when REn is present (Sunter









et al. 1990). The AC4 gene is involved in symptom expression of monopartite

begomoviruses, but to date does not appear to have a function in bipartite begomoviruses.


1GBo. iS.L-*


SJL


Figure 1-2. Hypothetical model for the functional organization of the replication origin in
begomoviruses. Rep Compl = Replication complex, TBP Complex= TATA-
binding protein complex, TF = Transcription factor, S.L = Stem loop (Source:
Argtiello-Astorga et al. 1994).

The B component has two genes: the NSP gene, which is transcribed from the

viral-sense strand, and a movement protein (MP), which is transcribed from the

complementary-sense strand. The NSP is implicated in nuclear shuttling of the viral

genome, and MP is involved in cell-to-cell movement of the virus via plasmodesmata

(Gafni and Epel, 2002). The MP appears to be a symptom-inducing element or a

determinant of pathogenicity of bipartite begomoviruses. Mutation studies suggest that


^^^^wBAplex
c'cc^^TATA









the 3' region of the MP gene is associated with symptom development (Gafni and Epel,

2002).

Bipartite begomoviruses often spontaneously produce approximately half-sized

defective DNA B components that function as defective interfering (DI) DNA. The DI

DNA may have a biological role during infection to reduce the severity of the disease by

competing with the genomic components for cellular resources (Stanley et al. 1990).

Monopartite begomoviruses have small circular single-stranded DNA satellites, named

DNA P. These depend on begomoviruses for their proliferation and, in turn, they affect

the accumulation and symptom expression of begomovirus (Mansoor et al. 2003).

Criteria for Species Demarcation

Several taxonomic criteria for demarcating species of begomoviruses have been

proposed by the International Committee on Taxonomy of Viruses (ICTV) based on the

reliability and applicability of these criteria to the large number of characterized

begomoviruses (Fauquet et al. 2003). Nucleotide sequence comparison plays a much

greater role in determining taxonomic status. Thus, for comparative analyses, only full-

length DNA A sequences were considered, based on recombination events that readily

occur among begomoviruses (Fauquet et al. 2003; Pita et al. 2001). A cut-off value of

89% of nucleotide sequence identity (NSI) of the A component was established to

distinguish different species from strains (Fauquet et al. 2003).

Impact of Begomoviruses on Agriculture

Proliferation and rapid dissemination of begomoviruses that infect important food

and industrial crops in Latin America have been the consequence of drastic changes in

traditional cropping systems (Morales and Anderson, 2001), along with the introduction









of the B biotype of Bemisia tabaci beginning in the mid 1980s (Polston and Anderson,

1997). The B biotype of B. tabaci has displaced many indigenous biotypes, because of its

broader host range, higher fecundity, dispersal capacity, virus-transmission efficiency,

and resistance to insecticides traditionally used against whiteflies (Brown et al. 1995).

Begomoviruses have been reported as limiting factors in the production of several

crops in the Americas such as cotton, common bean, tomato, pepper, and cucurbits,

among others. (Morales and Anderson, 2001; Polston and Anderson, 1997). In the 1990s,

Cassava mosaic begomoviruses caused a major regional pandemic of Cassava mosaic

disease (CMD) (affecting parts of at least five countries in Africa) that led to massive

economic losses and destabilization of food security (Legg and Thresh, 2000). A key

factor in the genesis and spread of the pandemic was the recombination of two distinct

cassava mosaic begomoviruses to produce a novel and more virulent hybrid (Pita et al.

2001). Resistance was developed originally in Tanzania, providing effective CMD

control in current pandemic-affected areas of East Africa. (Legg and Thresh, 2000).

Tomato yellow leaf curl virus (TYLCV) is another example of an emerging virus

that causes epidemics worldwide with frequent losses of up to 100% (Moriones and

Navas-Castillo, 2000). In tomato, the virus causes prominent upward curling of leaflet

margins, reduction of leaflet area, yellowing of young leaves, and stunting of plants.

These symptoms were first reported in tomato crops from Israel in the late 1930's, then in

Middle Eastern countries from the 1960s to the present. Damage to tomato crops

attributed to TYLCV has been reported in the Middle East and Far East, Africa, Europe,

Caribbean Islands, Central America, Mexico, and the United States of America

(Moriones and Navas-Castillo, 2000).









Management of Begomoviruses

Begomovirus management strategies have been implemented in several locations

in Central America and the Caribbean after the occurrence of serious agricultural and

economic crises caused by begomovirus infection of many crops in the mid 1980s (Hilje,

2002). In the first decade of implementation, area-wide plant-protection campaigns were

initiated. These involved quarantine regulations and host-free periods in the Dominican

Republic, Mexico, and Cuba. Cultural practices (such as production of seedlings under

netting and the use of living ground covers in production fields) are the most novel

contributions of this action plan (Hilje, 2002).

Several practices have been implemented to control begomoviruses in tomato

crops, such as the destruction of infected crops at the end of the production cycle using

herbicide combined with oil to kill plants and whiteflies, and then burning the plants; use

of virus-free transplants; and removal of infected plants at the first sign of begomovirus

symptoms (Schuster and Polston, 1999). In greenhouses, the use of ultraviolet

(UV)-absorbing plastic sheets or (UV)-absorbing screens of 50-mesh density (Antignus,

2000) (which interferes with the visual behavior of B. tabaci) has been shown to reduce

begomovirus transmission and crop losses. The use of a 50-mesh screen prevents vector

entry, but it produces poor ventilation and overheating of the closed structures

(Morriones and Navas- Castillo, 2000). A good practice is to eradicate plants that can be

sources of inoculum, making it possible to reduce the primary spread of the virus

(Kashina et al. 2002; Schuster and Polston, 1999). One practice more accepted by

growers in many locations is chemical control of the insect vector, but it has a detrimental

environmental impact; and whitefly vector populations rapidly develop resistance to

insecticides. Insecticides that are not toxic to nontarget species reduce the impact on









important natural enemies such as Eretmocerus sp. and Diglyphus sp. parasitoids

compared to conventional insecticides (Hanafi et al. 2002).

Host resistance to begomoviruses is another means of management. Development

of resistant cultivars by classical plant breeding or genetic engineering requires time, a

good scale for evaluating symptom severity, inoculation protocols, and constant

adjustment of the resistance due to changes in begomovirus populations (Lapidot and

Friedmann, 2002). Some genes for resistance to certain begomoviruses that infect tomato

have been identified in wild species of Lycopersicon and have been transferred to

cultivated tomato. Thus, resistant cultivars and breeding lines have been generated.

Recently, immunity to infection to TYLCV was obtained from L. hirsutum. This

immunity was shown to be controlled by three additive recessive genes (Vidavsky and

Czosnek, 1998).

Another option to obtain resistance is using genetic engineering via pathogen-

derived resistance approaches, such as CP-mediated resistance (Kunik et al. 1994;

Sinisterra et al. 1999); MP-mediated resistance (Duan et al. 1997; Hou et al. 2000),

defective interfering viral DNA (Stanley et al. 1990); Rep gene in antisense orientation

(Bendahmane and Gronenborn, 1997); and expression of truncated viral Rep protein

(Brunetti et al. 2001; Polston et al. 2001). The two first approaches involve expression of

the CP and MP to inhibit viral proliferation. The last three approaches inhibit viral

replication by disrupting the activity of the Rep gene.

Currently for optimal control of begomoviruses in tomato it is necessary to use

several practices simultaneously. Development of new and improved methods of control

for begomoviruses depend on our understanding of the mechanisms involved in virus-









vector and virus-host plant recognition, and knowledge of the variant forms of the virus

in natural populations.

Detection and Discrimination of Begomoviruses

Begomoviruses have been detected in plants or insects by different techniques,

such as visualization of nuclear inclusion bodies by light microscopy, ultrastructural

localization of virions in plant cell by transmission electron microscopy, serological

assays using polyclonal or monoclonal antibodies (Hunter et al. 1998; Konate et al. 1995;

Pico et al. 1999; Polston et al. 1989), DNA hybridization assays (Lotrakul et al. 1998),

Polymerase chain reaction (PCR) (Deng et al. 1994; Ghanim et al. 1998; Lotrakul et al.

1998; Mehta-Prem et al. 1994; Pico et al. 1999; Rosell et al. 1999), immunocapture PCR

(Rampersad and Umaharan, 2003), and print-PCR (Navas-Castillo et al. 1998) among

others. Molecular cloning and DNA sequencing of viral genomes have become the tools

of choice, allowing virus identification and evaluation of relationships with other virus

isolates (Brown et al. 2001; Padidam et al. 1995; Paximadis et al. 1999; Rybicki, 1994).

Variability of Begomovirus Species

High diversity among begomovirus species is associated with mixed infections, in

which recombination and pseudorecombination events may explain the frequent

emergence of new begomoviruses. Recombination is the exchange of DNA between

similar DNA components, and pseudorecombination is the exchange of DNA

components (Polston and Anderson, 1997). Both events have been demonstrated in the

laboratory (Hill et al. 1998; Garrido-Ramirez et al. 2000) and under natural conditions

(Pita et al. 2001). Some isolates of African cassava mosaic virus (ACMV), TYLCV and

Potato yellow mosaic virus (PYMV) are good examples of recombination and

pseudorecombination (Monci et al. 2002; Pita et al. 2001; Umaharan et al. 1998).









Variability of African cassava mosaic virus and Tomato yellow mosaic virus in the
Old World

The situation of African cassava mosaic disease (ACMD) in Africa is increasing

in complexity due to the number and types ofbegomovirus isolates from different

locations within the continent. Four different species, Indian cassava mosaic virus,

ACMV, East African cassava mosaic virus (EACMV), and Siuh,, African cassava

mosaic virus, can cause ACMD. A comparison of sequences of African cassava

begomoviruses revealed that all the isolates of ACMV (irrespective of their geographical

origin) were clustered together with little or no variation in their genomic sequence.

However, the sequences of EACMV isolates were more genetically diverse than those of

ACMV due to the frequent occurrence of recombination among equivalent components

of different strains and species. Variation among EACMV isolates is so high that their

classification is becoming problematic (Pita et al. 2001). In addition, a synergistic

interaction between ACMV and EACMV has been reported in three countries in Africa:

Uganda, Cameroon, and Ivory Coast. This interaction has led to very severe

symptomatology and yield losses (Fondong et al. 2000; Pita et al. 2001).

The yellow leaf curl disease in tomato is causing important economic losses to

this crop worldwide. Phylogenetic analysis of the CP from 23 accessions of TYLCV

revealed the presence of seven groups of sequences now classified as seven species:

TYLCV-Israel, TYLCV-Sardinia, TYLCV-Thailand, TYLCV-China, TYLCV-Tanzania,

TYLCV- Nigeria, TYLCV-Saudi Arabia (Moriones and Navas-Castillo, 2000).

Recombination also contributes to the genetic diversity of emerging begomovirus

populations. In southern Spain, Tomato yellow leaf curl Sardinia virus (TYLCSV) and

TYLCV are distinct begomovirus species that co-exist in the field and contribute to the









epidemic of tomato yellow leaf curl disease. A natural recombinant between TYLCSV

and TYLCV has been detected and an infectious clone of a recombinant isolate

(ES421/99) was obtained and characterized. Analysis of its genome showed that

recombination sites are located in the intergenic region in which a conserved stem-loop

structure occurs at the 3'-end of the REn gene. The ES421/99 exhibited a wider host

range than TYLCSV and TYLCV, which might provide it with a selective advantage

over the parental genotypes. Field studies revealed that the recombinant strain is

becoming the predominant strain in the region in which it was detected (Monci et al.

2002).

Variability of Potato yellow mosaic virus in the New World

A relatively high variability of PYMV isolates has been reported from the

Caribbean, and from Central and South America: Four full-length and five partial

sequences of PYMV strains have been reported. Potato yellow mosaic virus-Venezuela

(PYMV-VE) was first reported in 1963 in Venezuela (Debrot et al. 1963). A bipartite

begomovirus closely related to PYMV-VE was reported to infect tomato in Panama

(Engel et al. 1998). The virus was molecularly characterized and called Potato yellow

Panama mosaic virus (PYMPV) (formerly named Tomato leaf curl virus-Panama). High

amino acid sequence similarity between PYMPV and PYVM-VE was determined for all

the open reading frames with the exception of the AC4 gene product (Engel et al. 1998).

Another bipartite begomovirus was reported to infect pepper, sweet pepper, okra, beans,

and several weeds in different locations in Trinidad (Umaharan et al. 1998). The virus

was fully sequenced and named Potato yellow mosaic Trinidad virus- Trinidad & Tobago

(PYMTV-TT). The full A component of PYMTV-TT has 85% nucleotide sequence

identity (NSI) with PYMV-VE; and was proposed to be a recombinant between either









PYMV-VE or PYMPV and Sida golden mosaic Honduras virus (SiGMHV) (Umaharan

et al. 1998). Potato yellow mosaic virus-[Guadeloupe] (PYMV-[GP]) was first reported

in 1998 as a strain of PYMV-VE (Polston and Bois, 1998). It was responsible for

epidemics of virus-like symptoms in tomato in Guadeloupe, Martinique, and Puerto Rico

(Polston and Bois, 1998). It was suggested that PYMV-[GP] could be a recent

introduction based on the high value of NSI between begomovirus sequences from distant

locations (Polston and Bois, 1998). Phylogenetic analysis of the A component of

PYMTV-[GP] revealed that this virus is closely related to PYMV-VE, and PYMTV-TT

(Urbino et al. in press). In addition, several partial sequences of PYMV-VE strains have

been reported such as PYMV-Martinique, PYMV-Puerto Rico, PYMV-Dominican

Republic, PYMV-VE strain tomato, and Tomato yellow mosaic virus (ToYMV) (Guzman

et al. 1997; Morales et al. 2001; Polston and Bois, 1998).

Begomoviruses in Tomato Crops in Venezuela

Venezuela was one of the first places worldwide where tomato begomoviruses

were identified. In 1963, Debrot and colleagues reported the presence of Tomato yellow

mosaic virus (ToYMV), the first begomovirus affecting tomato in Venezuela (Debrot et

al. 1963); and the second begomovirus of tomato known in the New World. Whitefly

transmission and physical properties of ToYMV were published by Uzcategui and Lastra

(1978). Later, a yellow mosaic disease affecting potato crops was reported in Venezuela

(Debrot and Centeno, 1985a). The causal agent of this disease was a bipartite

begomovirus named Potato yellow mosaic virus-Venezuela (PYMV-VE) (Roberts et al.

1988). The complete nucleotide sequence of PYMV-VE was generated (Coutts et al.

1991).









More recently, other begomoviruses in tomato in Venezuela have been reported,

based on partial sequences. One, from the state of Aragua, was considered an isolate or

strain of PYMV-VE (PYMV-VE strain tomato) (Guzman et al. 1997). The other (from

the states of Monagas, Guarico, and Portuguesa) was considered an undescribed virus and

tentatively named Tomato Venezuela virus (ToVEV) (Guzman et al. 1997). Recently the

genome of ToYMV was partially sequenced and found to be related to PYMV due to a

95.7% NSI between the two genomes. Thus, it was proposed that PYMV is not a distinct

species, but synonym of ToYMV (Morales et al. 2001). Therefore, it appears that there

is some variability of begomoviruses in tomato in Venezuela.

Background and Objectives

Tomato is one of the most important vegetable crops in Venezuela. Its production

is concentrated in the following states: Guarico, Lara, Aragua, Trujillo, Tachira and

Portuguesa. Since 1975, a yellow mosaic disease has been associated with annual losses

of up to 100% in Aragua and Lara states (Lastra and Uzcategui. 1975). Since the mid

1980s, there has been a lack of information regarding the importance of begomoviruses in

tomato fields in Venezuela. In the last 23 years, only three partial sequences have been

generated (Guzman et al. 1997: Morales et al. 2001). A survey of tomato-growing areas

was conducted from 1993 to 1998 with the purpose of identifying and determining the

incidence of RNA viruses in tomato in Venezuela. Samples (334) were collected from 10

states: Aragua, Barinas, Cojedes, Guarico, Lara, Merida, Portuguesa, Tachira, Trujillo,

and Zulia. Tomato aspermy virus, Tobacco mosaic virus, Potato virus Y and Cucumber

mosaic virus were detected in less than 18% of the samples (Nava et al. 1996, 1997,

1998a, 1998b). To study the begomovirus situation for tomato crops in Venezuela, the









samples collected from this survey were used to address the following general and

specific objectives

General objective: establish the identity, diversity and distribution of

begomoviruses that affect tomato crops in Venezuela.

Specific objectives:

* Detect begomoviruses in genomic DNA extracted from desiccated tomato tissues
using PCR.

* Analyze the variability ofbegomovirus sequences from Andean samples using
partial sequences of the A and B components amplified by PCR using degenerate
primers.

* Characterize two begomovirus, designed 2.9-v and 57-v viruses, which were found
in tomato in the Andean states of Trujillo and Merida.

* Generate full-length sequences of the A and B components of 2.9-v and 57-v
viruses.

* Determine the distribution of 57-v and 2.9-v viruses in tomato samples from ten
states in Venezuela by hybridization with specific probes.














CHAPTER 2
DETECTION AND VARIABILITY OF BEGOMOVIRUSES IN TOMATO-
PRODUCTION AREAS OF ANDEAN STATES, VENEZUELA

Introduction

The first report of a whitefly-transmitted virus affecting tomatoes in Venezuela

was published by Debrot et al. (1963). The tomato plants showed yellow mosaic,

stunting, upward cupping, and leaf deformation. In some fields these symptoms were

observed in more than 30% of the tomato crops in the state of Aragua. The disease was

called tomato yellow mosaic (TYM) and was shown to be caused by the begomovirus

Tomato yellow mosaic virus (ToYMV) (Debrot et al. 1963). Transmission and physical

properties of ToYMV have been described (Uzcategui and Lastra, 1978).

A yellow mosaic disease was also reported in potato in the state of Aragua in

1985 (Debrot and Centeno, 1985a). The causal agent of this disease was thought to be

ToYMV and the virus was characterized as a geminivirus. The disease was transmitted

by grafting, mechanically, and by whitefly to potato, Lycopersicon esculentum var.

cerasiforme (a common weed in Venezuela), and several other hosts of ToYMV (Debrot

and Centeno, 1985a). Later the causal agent of a yellow mosaic on potato was

characterized and named Potato yellow mosaic virus-Venezuela (PYMV-VE) (Roberts et

al. 1986; 1988). The full sequence of the A and the B components of PYMV, as well as

infectious clones, have been generated (Coutts et al. 1991; Roberts et al. 1988); and the

nucleic acid sequence was shown to be closely related to other begomoviruses from the

New World, especially in the coat protein (CP) gene regions (Coutts et al. 1991).









High incidences of TYM in tomato and other crops in Venezuela have been

correlated with high populations of the whitefly, Bemisia tabaci (Anzola and Lastra,

1985; Amal et al.1993a; 1993b; Debrot et al. 1963; Debrot and Centeno, 1985a, 1985b;

Debrot and Ordosgoitti, 1975). Annual rainfall patterns appear to be a major factor that

affect whitefly populations, with rapid population increases at the end of the rainy season

(November-December). The greatest number of whiteflies was captured at a height of 10

to 60 cm above ground, which coincided with the height of tomato plants (Arnal et al.

1993a). Recently the biotype B has been identified by ramdom amplified polymorphism

DNA (RAPD) analysis (Salas and Arnal, 2001).

There is evidence for other begomoviruses in tomato in Venezuela. Two partial

sequences of begomoviruses affecting tomato in Venezuela have been reported. One,

from the state of Aragua, was considered an isolate or strain of PYMV-VE (PYMV-VE

strain tomato) (Guzman et al. 1997). The other, from the states of Monagas, Guarico, and

Portuguesa, was considered an undescribed virus and named Tomato Venezuela virus

(ToVEV) (Guzman et al. 1997). Recently the genome of ToYMV was partially

sequenced and found to be related to PYMV based on 95.7% nucleotide sequence

identity (NSI). Thus it was proposed that PYMV is not a distinct species, but synonym of

ToYMV (Morales et al. 2001). There appears to be great variability of PYMV in

Venezuela. Much of what is known about begomoviruses in tomato is based on only

very few samples (selected from different states) which were collected and sent to

specialists outside of Venezuela for characterization and sequence analysis. Tomato is an

important crop in Venezuela and is produced commercially in 13 of the 23 states in

Venezuela. A more thorough study of the presence and diversity of begomoviruses is









needed that uses larger numbers of samples and includes samples from all the major

tomato-production areas in Venezuela.

In addition to begomoviruses, other viruses have been found to infect tomato in

Venezuela. A survey conducted in 1975 reported four viruses that were affecting

tomatoes in Aragua and Lara states: ToYMV, Tobacco etch virus, Tobacco mosaic virus,

and Cucumber mosaic virus. At flowering, all the plants were shown to be infected with

ToYMV alone or in combination with one of the other three viruses mentioned above

(Lastra and Uzcategui, 1975). Tobacco etch virus was found infecting tomato crops in

Yaracuy and Aragua states (Debrot, 1976). A survey of tomato fields in 10 Venezuelan

states from 1992 to 1998 reported that, in addition to the viruses reported previously,

tomato plants were infected with Potato virus Y, Potato virus X, Tomato aspermy virus,

Tomato ringspot virus, Tomato spotted wilt virus, Tobacco streak virus, and Zucchini

yellow mosaic virus (Nava, 1999; Nava et al. 1996, 1997, 1998a, 1998a).

Objectives of this research were

* Determine a reliable DNA extraction protocol for amplification of DNA from
desiccated tomato tissue of samples from the Andean states

* Detect begomoviruses in genomic DNA extracted from desiccated tomato tissue
using the polymerase chain reaction (PCR)

* Analyze the variability ofbegomovirus sequences from Andean samples using
partial sequences of the A and B components amplified by PCR using degenerate
primers.

Materials and Methods

Survey

Samples of leaf tissue from tomato plants showing symptoms of viral diseases

were collected from the main tomato-growing states in Venezuela from 1993 to 1998

(Nava et al. 1997, 1996, 1998a, 1998b). In general, mosaic, curly leaf, yellowing, vein









clearing, stunting, upward cupping, reduced leaf size, and chlorotic leaf margins were the

symptoms observed in the tomato plants in the fields. The samples were collected from

plants approximately 2 months after transplant to the field. The total number of samples

collected was 334. The samples were dried at room temperature for 2 weeks. Then they

were cut, rolled in a piece of tissue paper, placed into a vial that was 1/3 filled with silica

gel, and then stored at -200C. The survey covered 10 states: Aragua, Barinas, Cojedes,

Guarico, Lara, Merida, Portuguesa, Tachira, Trujillo, and Zulia (Figure 2-1). Aragua and

Lara have large fields and continuous tomato production. The Andean states (Merida,

Tachira and Trujillo) have small fields and continuous tomato production. The states of

Barinas, Cojedes, Guarico, Portuguesa, and Zulia have large fields and one season of

tomato production per year.

DNA xtraction

Genomic DNA was extracted from each sample using the protocol described by

Doyle and Doyle (1987). This protocol resulted in a DNA pellet with sign of oxidation

from a large number of the samples. Sixteen samples from the Andean states were

selected to compare Doyle and Doyle (1987) and another protocol developed to obtain

tannin- and polysaccharide-free genomic DNA from mature tissue of plants from genera

belonging to the Dipterocarpaceae (Rath et al. 1998).











Nueva Esparta


.- 1 .i-


Bolivar


SAndean states
m Aragua and Lara states
SCentral-Westrn plain states
SZulia states


Figure 2-1. Map of Venezuela indicating the states where plant tissue samples were
collected.

Samples from the Andean states were selected for more in-depth studies because

this region was likely to have the greatest variability of begomovirus sequences. The

higher diversity of sequences is expected because of the continuous (year-round)

production of tomatoes in this region; and because of some growers from the state of

Trujillo routinely obtain tomato transplants from the state of Lara, where several

begomoviruses have already been reported (Guzman et al. 1997; Lastra and Uzcategui,

1975). Some modifications were made to the Rath protocol: desiccated tissue was

rehydrated for 10 min in 2X CTAB (cetyltrimethylammonium bromide) buffer before

extraction; and phenol:chloroform:isoamyl in 25:24:1 (v:v:v) was used in place of









phenol:chloroform in 1:1 (v:v). The DNA extracted by both protocols was stored

at -800C.

Polymerase Chain Reaction (PCR)

Two sets of primers were used for each component of the viral genome. Primers

PARlc496 and PALlv1978 (Rojas et al. 1993) amplify -1100 bp of the A component.

This includes -690 bp of the replication associated protein (Rep), -300 bp of the region

between the beginning of the Rep and the beginning of the coat protein (CP), and -120

bp of the CP. Primers PCRv181 and PARlc496 (Rojas et al. 1993) amplify -300 bp of

the A component that contains -180 bp of region between the beginning of the loop and

the beginning of the CP and -120 bp of the CP. Primers PVL1v2040 and PCRcl54

(Rojas et al. 1993) amplify -600 bp of the B component, which consists of part of the

nuclear shuttle protein (NSP) (-160 bp), the entire hypervariable region and part of the

common region (CR). The primers JAP 58 (5' -TTCAGTGCCGAAGACCGAAG-3')


and JAP 59 (5' -ACGGGAAATGGGAGAGGAAG-3') are designed for PYMV-VE and

amplify a fragment of -750 bp of the B component, which comprises 420 bp of the

NSP, -20 bp of sequence between the end of the NSP and the movement protein (MP),

and 330 bp of the MP.

The PCR reactions for primer pairs PARlc496 and PALlv1978, and PVLlv2040

and PCRc154 were carried out as previously described (Rojas et al. 1993). For primers

PCRv181 and PAR1c496, DNA amplification parameters were 35 cycles of denaturation

for 1 min at 940C, primer annealing for 20 s at 600C, and primer extension for 30 s at

720C, with an initial denaturation at 940C for 2 mins and a final extension for 7 min at

720C. The DNA amplification parameters for primers JAP58 and JAP59 were 35 cycles









of denaturation for 30 s at 940C, primer annealing for 30 s at 60 C and primer extension

for 45 s at 75C, with an initial denaturing at 940C for 5 min and a final extension for 5

min at 72C. The PCR reactions were carried out in a Gene Amp PCR system 9700

(PE Applied Biosystems, Foster City, CA) thermocycler. All amplifications were

performed in volumes of 25 [tL containing 10 mM Tris-HCl (pH 9), 50 mM KC1 and

1% Triton X-100, 2.5 mM MgC12 except for primers JAP58 and JAP59, which

contained 2 mM MgC12, 250 [aM dNTPs, 0.5 mM spermidine, 1 [aM of each primer,

100 ng of genomic DNA, and 2.5 U of Taq polymerase. PCR products were

electrophoresed (1 h at 90 volts) in 1% agarose gels in Tris-acetate-EDTA buffer, pH 8.

Gels were stained with ethidium bromide (0.0015 mg/mL), viewed by a UV

transilluminator.

Restriction Enzyme Analysis of Purified PCR Products

The PCR products from the Andean samples were purified using the QIAquick

gel extraction kit (QIAGEN Inc. Valencia, CA). The purified PCR products generated

from primer set PARlc496 and PAL1v1978 (-1100 bp) were digested with BglII, EcoRI,

NcoI, and NdeI. The purified PCR products generated from primer set PVLlv2040 and

PCRc154 (-700 bp) were digested with EcoRI, HindIII, KpnI, NdeI, and XbaI. Digestion

conditions were performed according to manufacturer's instructions (New England

Biolab, Inc., Beverly, MA).

Cloning and DNA Sequence Determination

Ligation of purified PCR products amplified from Andean samples was

performed using the pGEM-T Easy vector system I (Promega Corporation, Madison, WI)

according to the manufacturer's instructions. Transformation was performed using

XL 1-Blue MRF Supercompetent Escherichia coli cells (Stratagene, La Jolla, CA). The









clones were screened with the same restriction enzymes used above. Nucleotide

sequences of plasmid DNA from clones were determined by automated sequence analysis

at the DNA Sequencing Core Laboratory, Interdisciplinary Center for Biotechnology

Research (ICBR) University of Florida, Gainesville.

Comparison of DNA Sequence

Nucleic acid sequences were analyzed using a Wisconsin Package Version 10.3,

[Accelrys (GCG), San Diego, CA]. Basic Local Alignment Search Tool (BLAST), was

used to search for similarities between a query sequence and all the sequences in the

database. The six begomovirus sequences from the database that had the highest

similarity to each Andean sequences were selected to perform GAP analysis. GAP was

used to obtain the values of NSI. GAP uses the algorithm of Needleman and Wunsch,

which considers all possible alignments and gap positions between two sequences and

creates a global alignment that maximizes the number of matched residues and minimizes

the number and size of gaps. Multiple alignments of Andean sequences and the

begomovirus sequences (selected by the similarity values) were performed using

PILEUP, which is the first step for phylogenetic analysis.

Phylogenetic Analysis

Phylogenetic analysis was carried out using PAUP* (Phylogenetic Analysis Using

Parsimony). A maximum parsimony and heuristic tree search were used to generate the

best tree. The initial tree was created by stepwise addition with tree-bisection-

reconnection as branch swapping method. The reliability of the tree was estimated by

performing 500 bootstrap repetitions. The begomoviruses selected for the phylogenetic

analysis are listed in table 2-1.









Table 2-1. Acronyms and accession numbers of known begomoviruses used in the
phylogenetic analysis of partial sequences of samples from Andean states


Species
Bean dwarf mosaic virus
Bean golden yellow mosaic virus-[Puerto Rico]
Chino del tomate virus
Chino del tomate virus-[IC]
Dicliptera yellow mottle virus
Jatropha mosaic virus
Macroptilium golden mosaic virus-[Jamaica]
Macroptilium yellow mosaic Florida virus
Potato yellow mosaic Panama virus
Potato yellow mosaic virus [Guadeloupe]

Potato yellow mosaic Trinidad virus-Trinidad
& Tobago
Potato yellow mosaic virus- [Dominican
Republic]
Potato yellow mosaic virus- Martinique

Potato yellow mosaic virus-Venezuela
Potato yellow mosaic virus-Venezuela strain
tomato
Rhynchosia golden mosaic virus
Sida golden mosaic Costa Rica virus
Sida golden mosaic Honduras virus
Tomato dwarf leaf curl virus
Tomato golden mottle virus
Tomato leaf curl Malaysia virus
Tomato leaf curl Sinaloa virus
Tomato mottle Taino virus

Tomato mottle virus-[Florida]
Tomato Venezuela virus


Acronym
BDMV
BGMV-[PR]
CdTV
CdTV-[IC]
DiYMoV
JMV
MGMV-[JM]
MaYMFV
PYMPV
PYMV-[GP]


PYMTV-TT

PYMV-[DR]

PYMV-Mart

PYMV-VE
PYMV-VE-[tom]

RhGMV
SiGMCRV
SiGMHV
ToDLCuV
ToGMoV
ToLCMV
ToLCSinV
ToMoTV

ToMoV-[FL]
ToVEV


Accession number
M88179
M10080
AF226664
AF101476
AF170101
AF324410
AF098940
AY044136
Y15033
AY120882,
Y120883
AF039031,
AF039032
AY126611,
AY126614
AY126610,
AY126612
D00940, D00941
AF026553

AF239671
X99550, X99551
Y11097
AF035225
AF138298
AF327436
AF131213
AF012300,
AF012301
L14461
AF026464


Results


Comparison of DNA Extraction Protocols

The Rath protocol (Rath et al. 1998) produced a whiter pellet, and more stable

DNA than the Doyle and Doyle DNA extraction protocol (Doyle and Doyle, 1987). A

greater number (44%) of samples using the Rath protocol produced the expected

fragment size after amplification, especially for primers that amplified the B component









(Table 2-2). DNA that was extracted using the Rath protocol could be amplified after

6 months of storage at 40C. There was no difference in PCR results when fresh tissue was

used for either DNA extraction protocol (ToMoV-[FL] and Tomato yellow leaf curl

virus-[Florida]), however when DNA was extracted from frozen tissue there was a

difference in intensity of the bands after electrophoresis of the PCR product from Potato

yellow mosaic virus-[Guadeloupe]. In addition, the stability of the DNA extracted with

the Doyle and Doyle protocol was low which resulted in inconsistent amplifications.

For several dehydrated tomato samples from the survey, brownish pellets were

obtained after using the Doyle and Doyle protocol. The color of the pellet was eliminated

with an additional incubation period with RNAse A and proteinase K, followed by

phenol: chloroform:isoamyl extraction, and cesium chloride precipitation.

PCR Amplification of Begomovirus Sequences from Field Samples

Approximately 50% of the samples produced a PCR product of -1100 bp using

the degenerate primers PAR1c496 and PAL1v1978 (Table 2-3). The primer set PCRv181

and PARlc496 was able to amplify a -300 bp fragment which is located within the

~1100 bp fragment from some but not all of the samples from which a ~1100 bp

fragment was obtained (Tables 2-2 and 2-3). Most likely this was due to sequence

differences at the binding site for PCRv181 which prevented annealing. It could also be

due to the presence of an inhibitor, which might have affected the performance of

PCRv181. The percentage of samples from which the B component primers amplified a

product was almost 50% less than the percentage obtained for the primer set

PARlc496/PALlv1978 used to amplify products from the A component (Table 2-3). The

results suggested that the primer sets PARlc496/PALlv1978 and PBLlv2040/PCRcl42














Table 2-2. Comparison of two DNA extraction protocols through PCR results of samples from Andean states, Venezuela using 4
primer sets to amplify partial sequence of A and B components of begomoviruses
State Sample
DSte S e A component B component
ID No.
t Primers Primers Primers
Primers JAP58/JAP59
PCRv181/PARlc496 PARlc496/PALlv1978 PBLlv2040/PCRc 54
D&D* R** D&D R D&D R D&D R


2.2-v +
2.4-v +
2.7-v
2.8-v
2.9-v
6-v +
8-v +
10-v +
12t-v +
28-v +
37-v +
56-v +
57-v +
58-v
82-v
ToMoV-[FL] +
TYLCV-FL] +
PYMV-[GP] +
Healthy tom.
Doyle and Doyle DNA


+
+
+

+

+
+



+
+
+


extraction; **R = Rath DNA extraction; + = Right PCR product was amplified; -


produced
t ToMoV [FL] = Tomato mottle virus [Florida]; TYLCV [FL] = Tomato yellow leaf curl virus [Florida]; PYMV-
mosaic virus [Guadeloupe]; Healthy tom.= Healthy tomato


+

+
+
+
+
+

+
+

+



+

- No PCR product was

[GP]= Potato yellow


Trujillo









Tachira



Merida
" 1
" i
" i


*D&D=












Table 2-3. Polymerase chain reaction results using degenerate primers to amplify A and B components of begomoviruses using Doyle
and Doyle DNA extraction of Venezuelan samples from tomato-production areas
No. and % of positive PCR samples


No. of fields/ A component B component
No. of No. of Primers Primers Primers Year of
State locations samples PCRv181/PARlc496 PARlc496/PALlv1978 PBLlv2040/PCRcl54 collection


64( 34%)
6( 75%)
0( 0%)
13 (100%)
21 ( 78%)
2( 9%)
1 ( 20%)
2( 8%)
6( 32%)
20( 63%)
135 ( 40%)


81 ( 46%)
6( 75%)
0( 0%)
13 (100%)
21 ( 78%)
4( 17%)
2( 40%)
2( 8%)
9( 48%)
21 ( 66%)
159( 48%)


44 (25%)
0( 0%)
0( 0%)
12 (92%)
8 (30%)
1 ( 4%)
1 (20%)
2( 8%)
6 (32%)
6 (19%)
80 (24%)


1995-1996
1995
1995
1998
1994
1994
1995
1993
1993
1993-1995


Aragua
Barinas
Cojedes
Guarico
Lara
Merida
Portuguesa
Tachira
Trujillo
Zulia
Total


15/ 4
2/2
1/ 1
3/ 2
10/ 8
11/ 6
2/ 2
10/ 7
6/ 6
20/12
80/50









are useful for studies of variability of begomoviruses because they were able to amplified

larger number of samples, the size of amplified a fragment was optimum for sequencing

determination, and the presence of part of the CR in the amplified fragment from the

B component enabled the confirmation that the sequences for the A and B components

belonged to the same virus.

Table 2-4. Expected restriction-fragment sizes (bp) in the fragment amplified by primer
set PAR1c496/PALlv1978 of the A component of known begomoviruses
Restriction Enzyme fragment sizes (bp)
Sample Idel NcoI EcoRI BglII
TYLCV-[ FL] 1000+291 1234+ 57 760+531
TYLCV IL sev 1015+288 1246+ 57
ToVEV 829+293 762+253+107 781+341 566+329+227
ToMoV-[FL] 844+294 1031+107 808+330
TLCV -944+345
SiGMV 846+294 667+317+156 810+330
PYMV-VE-[tom] 753+375 1021+107
PYMV-VE 1023+107
PYMTV-TT
PYMV-[GP] 1025+134 602+556
PHYVV 873+291 819+345 837+327
= No restriction site
TYLCV-[FL] = Tomato yellow leaf curl virus- [Florida]
TYLCV IL sev = Tomato yellow leaf curl virus -[Israel strain severe]
ToVEV = Tomato Venezuela virus (ToVEV)
ToMoV-[FL] = Tomato mottle virus-[Florida]
TLCV = Tobacco leaf curl virus
SiGMV = Sida golden mosaic Virus
PYMV-VE-[tom] = Potato yellow mosaic virus-Venezuela [strain tomato]
PYMV-VE = Potato yellow mosaic virus-Venezuela
PYMTV-TT = Potato yellow mosaic Trinidad virus-Trinidad & Tobago
PYMV-[GP] = Potato yellow mosaic virus [Guadeloupe]
PHYVV = Pepper huasteco yellow vein virus

Restriction Enzyme Analysis of Purified PCR Product

Purified PCR products amplified by primer set PARlc496/PALlv1978 for the A

component of the Andean samples were digested. Restriction patterns were compared

with those expected of the same region of selected begomoviruses (Table 2-4). None of










the Andean samples produced restriction-fragment sizes identical to those selected

begomoviruses (Table 2-5). However, the restriction-fragment sizes of each sample were

useful to distinguish the presence of single or mixed infections en each sample. There

were five samples that appeared to be infected with multiple begomoviruses, 2.2-v, 2.4-v,

8-v, 10-v, and 82-v (Table 2-5). Three of the mixed infections were confirmed by

sequence determination of different clones from the same sample (multiple sequences

from samples 2.2-v and 8-v could not be obtained).

Table 2-5. Restriction-fragment sizes in the fragment amplified by primer set
PAR1c496/PALlv1978 in the A component of begomoviruses from Andean
states, in Venezuela
Restriction Enzyme fragment sizes (bp)
State Sample BgflII EcoRI [NcoI Ndel
Trujillo2.2-v 600+600 1150 1000+160+116 1150+950+850+175+134
2.3-v 600 +600 1200+850+350 1100+1000+160+116 1100+950+900+500+300
134+260+220+175
6-v 600+600 1150 900+160+116 1150
2.4-v 600+350 1150+750+3501000+900+160+116 1150+1000+850
750+300+134
2.7-v 800+350 1150 900+160+108 1150
2.8-v 800+350 1150 900+160+108 1150
2.9-v 800+350 1200 900+160+108 1150
8-v 600+350+2501200 1100+116 1150+900+800+350
10-v 600+350+250750+350 1100+900+200+161+1301150+900+300
Tachira28-v 600+600 1200 900+185+130 1150+182
Merida 56-v 800+350 750+350 1000+185 1150+182
57-v 800+350 750+350 1000+185 1150
82-v 800+350 800+370 1000+900+185+130 900+300+182

There were fewer differences in restriction patterns among amplified products

using the primer set, PBLlv2040/PCRcl54, which amplifies a region of the B

component. This could be due to the short length of the PCR product (-700 bp) used for

this analysis, and/or the fact that this region has fewer restriction sites than other regions

of the genome (Table 2-6). The PCR products of three other primer sets for the B

component were evaluated for restriction analysis (data not shown). These generated









larger PCR products, included other regions of the genome and all were more suitable for

restriction analysis than the product generated by primer set PBLlv2040/PCRc154.

However, the primer set PBLlv2040/PCRc154 was selected for restriction analysis

because it amplified a fragment from almost all of the Andean samples that were positive

for the A component based on primer set PARlc496/PALlv1978 (except for sample 82-

v). In addition, this amplified fragment included part of the CR, which is necessary to

identify the A and B components of the same virus. The following enzymes were used in

the analysis: XbaI which had no restriction sites in any of the samples, HindlII which had

a restriction site in one sample, and Ndel and EcoRI which had one restriction site in

three and six of 12 samples, respectively (Table 2-7). The restriction analysis suggested

that there were several A component sequences but just one B component sequence in

three of the samples with mixed infections (samples 2.4-v, 10-v and 82-v).

Table 2-6. Expected restriction-fragment sizes in the fragment amplified by primer set
PBL1v2040/PCRcl54 of the B component of known begomoviruses
Restriction Enzyme fragment sizes (bp)
ba I Nde I Hind III EcoR I Kpn I
PYMV-GP -- 458+160 -
PYMV-VE -- 444+160 239+365
PYMTVTT 436+148 424+160
ToMoTV 396+231 484+143 331+296
- = No restriction site
PYMV-[GP] = Potato yellow mosaic virus [Guadeloupe]
PYMV-VE = Potato yellow mosaic virus-Venezuela
PYMTV-TT = Potato yellow mosaic Trinidad virus-Trinidad & Tobago
ToMoTV = Tomato mottle Taino virus

DNA Sequence Comparison

The sequence of 15 clones obtained from purified PCR products amplified by

primer set PAR1c496/PALlv1978 from 12 samples of Andean states were submitted for









Table 2-7. Restriction-fragment sizes in the fragment amplified by PBLlv2040/PCRc154
of the B component of begomoviruses from Andean states, in Venezuela
Restriction Enzyme fragment sizes (bp)
State Sample Xlbal del [HindlII [EcoRI [KpnI
Trujillo 2.2-v 700 700 700 550+200 700
2.3-v 700 700 700 700 700+400
2.4-v 700 700 700 500+200 700+500+200
2.7-v 700 550+180 700 700
2.8-v 700 700 700+400+289 700 700
2.9-v 700 700 700 700
6-v 700 700 700 700*+550+200 700+500+200
8-v 700 700 700 700*+550+200 700+500+200
10-v 700 700 700 700*+550+200 700+500+200
Tachira 28-v 700 700 700 700*+550+200 700+500+200
Merida 56-v 700 700+400+250 700 700 700
" 57-v 700 700+400+250 700 700 700

a BLAST search. These ~1100 bp sequences were clustered into four groups according to

the percentage of NSI by GAP analysis, each group having high NSI with a known

begomovirus (Table 2-8). There was a high value of NSI (from 94 to 99%) within the

sequences in groups 1 and 2. There was a high NSI among the cloned sequences of group

3 (99%) and 4 (98%). Group 1 sequences (2.2-v37, 6-v, 10-v4, 28-v5, 82-vl, and 82-v5)

were characterized by high NSI with PYMV-VE-[tom] (from 92 to 93 %). Group 2

sequences (2.4-v6, 2.4-v7, 10-vl, and 82-v7) were characterized by high NSI with

ToVEV (from 93 to 96%). Group 3 sequences (2.7-vl, 2.8-v and 2.9-vl) shared its

highest NSI with ToLCSinV (86%), group 4 sequences (56-v4 and 57-v) shared its

highest NSI with PYMTV-TT (82% and 83%, respectively). Sequences in groups 3 and 4

had the lowest percentage of NSI with known begomoviruses of the 4 groups. They were

therefore considered potential new species of begomoviruses (Table 2-8). Sequences

from clones 2.9-vl and 57-v were selected for complete sequencing of the genome.

Sequences belonging to group 1, related to PYMV-VE-[tom], were detected in

samples from all three Andean states. Sequences belonging to group 2, related to ToVEV,









were detected in samples from Merida and Trujillo. All the sequences belonging to group

3 were detected in samples from Trujillo. All the sequences belonging to group 4 were

detected in samples from Merida

The sequences of the 15 clones were aligned using PILEUP. The sequences had a

conserved nonameric motif 5' -TAATATTAC-3' in the intergenic region, located in the

loop of the conserved "hairpin" element, where Rep introduces a site-specific nick to

initiate virus replication via a rolling-circle mechanism (Laufs et al. 1995b). Two

inverted repeats, two forward repeats, and a TATA box were present before the stem loop

sequence in all the cloned sequences in group 1, the other group sequences showed

differences in the number of inverted and forward repeats, keeping the conserved

nonameric motif and the TATA box. These inverted repeats or interactive sequences

(iterons) have been reported to be specific binding sites for the Rep (Rep iteron-related

domain), to initiate the rolling circle replication process (Arguello-Astorga et al. 1994,

Arguello-Astorga and Ruiz-Medrano, 2001)

Sequences in groups 1 share the iteron core sequence, GGGGG, and the GSFSIK

Rep iteron-related domain with PYMV (Table 2-9). Sequences in group 2 appear to be

related to ToVEV. The core sequences of iteron and Rep iteron-related domain in this

group were similar to those of ToVEV, there was a nucleotide change resulting in a

different amino acid in the Rep iteron-related domain between sequences of group 2 and

ToVEV, (lysine for isoleucine due to substitution of A for T in the nucleic acid

sequence). The iteron and Rep iteron-related domain sequences of ToVEV were reported

as unique, because they did not fit in any other group of iteron and rep iteron-related

domains of begomoviruses from the New World (Arguello-Astorga and Ruiz-Medrano,









2001). Sequences in group 3 also share the iteron core sequence, GGGGG, and the

GSFSIK Rep iteron-related domain with PYMV. PYMV has two inverted and forward

repeats, however the sequences in group 3 have only one inverted repeat (Table 2-9).

This is a significant difference and implies that sequences in group 3 are unique. The

sequences in group 4 are distinctive, since they have one inverted and one forward repeat

and are the most similar to those of RhGMV (Table 2-9). Thus, cloned sequences

belonging to groups 3 and 4 are tentative new species ofbegomoviruses from the New

World.

Partial sequences of the B component (primer set PVL1v 2040/PCRc 54) were

clustered into four groups after BLAST search and GAP analysis (Table 2-10).

Sequences 2.2-vB6, 2.4-vB34, 2.7-vB3, 6-vB2, 8-vB12, 10-vB15, and 28-vB2 had high

percentages of NSI (92-95 %) with PYMV- VE. The value of NSI within this group

ranged from 87 to 99%. These sequences could be strains of PYMV-VE, but further

studies would have to be done to confirm this assumption. Most of the samples of this

group were separated into two groups based on the comparison of A component

sequences (Table 2-8). Thus, a similar comparison could not be made for the B

component sequences since B component sequences were not reported for

PYMV-VE-[tom] and ToVEV. Therefore, the B component sequences for these samples

clustered with PYMV-VE. The sequences 2.8-vB8 and 2.4-vB24 had lower NSI (86 %

and 88 %, respectively) with ToMoTV. The sequence 2.9-vB20 had 86 % NSI with

ToGMoV, and the sequences 56-vB7 and 57-vB2 had lower percentages of NSI (86 %)

with DiYMoV (Table 2-10).















Table 2-8. Nucleotide sequence identity (%) between partial sequences from PCR using primer set PARlc496/PALlv1978 of the A
component of begomoviruses from Andean states and four known begomoviruses
Virus group 1 Virus group 2 Virus group 3 Virus group 4

Clone 6-v 28-v5 82-vl 82-v5 10-v4 PYMV-VE [tom] 2.4-v7 82-v7 10-vl ToVEV 2.7-vl 2.9-vl ToLCSinV 57-v PYMTV-TT

2.2-v37 98.7 96.98 98.22 98.62 98.11 91.93

6-v 97.88 98.71 98.62 98.43 93.14

28-v5 98.15 94.33 97.75 91.67

82-vl 95.87 99.01 91.76

82-v5 98.91 92.53

10-v4 93.00

2.4-v6 96.04 99.52 97.58 93.78

2.4-v7 95.68 93.76 93.35

82-v7 97.63 93.62

10-vl 96.00

2.8-v 99.61 99.32 86.00

2.7-vl 99.06 86.00

2.9-vl 86.00

56-v4 98.77 82.00

57-v 83.00

Virus group 1 related to Potato yellow mosaic virus-Venezuela strain tomato (PYMV-VE-[tom]
Virus group 2 related to Tomato Venezuela virus (ToVEV)
Virus group 3 related to Tomato leaf curl Sinaloa virus (ToLCSinV)
Virus group 4 related to Potato yellow mosaic Trinidad virus (PYMTV-TT)









Phylogenetic Analysis

The phylogenetic tree of nucleic acid sequences of clones from partial sequences

of begomovirus A components is shown in Figure 2-2. The Andean sequences followed

the same tendency of clustering that was observed in the GAP analysis. Sequences

2.2-v37, 6-v; 28-v5, 10-v4, 82-vl, and 82-v5 are closely related to the PYMV virus

group. Even though this analysis was done with partial sequences it is possible to infer

that PYMV-VE strain tomato is distributed in the three Andean states. The data also

confirm that MGMV is a sister taxon of the clone group mentioned above. The sequences

10-vl, 2.4-v 6, 2.4-v7 and 82-v7 are closely related to ToVEV.

Sequences 57-v and 56-v4 as well as 2.7-vl, 2.8-v and 2.9-vl had RhGMV and

ToLCMV as sister taxa, respectively. These two groups of sequences clustered by

themselves, they are not closely related to any known begomovirus.

The sequences from the B component were clustered into 3 groups using phylogenetic

analysis (Figure 2-4). The largest group of B component sequences (28-vB2, 2.2-vB6,

10-vB 15, 6-vB2, 8-vB 12, 2.4-vB34, and 2.7-vB3) was most closely related to strains and

isolates of PYMV-VE. The next largest group of sequences (2.8-vB8, 2.9-vB20 and

2.4-vB24) was not closely related to any known begomoviruses. The last group of

sequences (56-vB7 and 57-vB2) was distantly related to BGYMV-[PR] and MaYMFV.

The latter two groups of sequences appeared to be those of new begomoviruses.

The partial A and B component sequences from the same sample did not always

cluster with the same sequences in the two phylogenetic trees. For example, sequences of

from samples, 2.4-v and 2.7-v, were clustered in different groups for the A and B

component trees (Figures 2-2 and 2-3). This is primarily due to the fact that there were

four clusters of partial sequences in the A component tree and three in the B component












Table 2-9. Motifs of the intergenic region and Iteron-related domain of Replication protein (Rep IRD) of cloned sequences from
Andean states and eight known begomoviruses


Sequence*
2.2-v37
6-v
10-v4
28-v5
82-vl
82-v5
PYMV-Mart
PYMV-[GP]
PYMV-[DR]
PYMV-Tom
PYMV-VE


10-vl
2.4-v6
2.4-v7
82-v7
ToVeV

2.7-vl
2.8-v
2.9-vl

56-v4
57-v
RhGMV


Invert repeat
CCCCCGATA
CCCCCGATA
CCCCCGATA
CCCCCGATA
CCCCCGATA
CCCCCGATA
CCCCCAATA
CCCCCAATA
CCCCCAATA
CCCCCAATA
CTCCCAATA


Group
1
1
1
1
1
1


* Virus names and accession number are described in material and methods.
Iteron core and Rep IRD motif are shown in italic.


Invert repeat
CCCCCAATT
CCCCCAATT
CCCCCAATT
CCCCCAATT
CCCCCAATT
CCCCCAATT
CCCCCAATA
CCCCCAATA
CCCCCAATA
CCCCCAATT
CCCCCTATT

TGCACCGATT
TGCACCGATT
TGCACCGATT
TGCACCGATT
TGCACCGATT

CCCCCAATT
CCCCCAATT
CCCCCAATT

ACACCAATT
ACACCAATT
ACCCCGATT


Forward repeat
AATTGGGGG
AATTGGGGG
AATTGGGGG
AATTGGGGG
AATTGGGGG
AATTGGGGG
AATTGGGGG
AATTGGGGG
AATTGGGGG
AATTGGGGG
AATTGGGGG

AATTGGGGCA
AATTGGGGCA
AATTGGGGCA
AATTGGGGCA
AATTGGGGCA

AATCGGGGG
AATCGGGGG
AATCGGGGG

AATCGGTGT
AATCGGTGT
TATCGGTGT


Forward repeat
AACTGGGGG
AACTGGGGG
AACTGGGGG
AACTGGGGG
AACTGGGGG
AACTGGGGG
AACTGGGGG
AACTGGGGG
AACTGGGGG
AACTGGGGG
AACTGGGGG

AATTGGGGTC
AATTGGGGTC
AATTGGGGTC
AATTGGGGTC
AAATGGGGTC

AACTGGGGG
AACTGGGGG
AACTGGGGG



TATCGGTAT


Rep IRD
MPRKGSFSIK
MPRKGSFSIK
MPRKGSFSIK
MPRKGSFSIK
MPRKGSFSIK
MPRKGSFSIK
MPRKGSFSIK
MPRKGSFSIK
MPRKGSFSIK
MPRKGSFSIK
MPRKGSFSIK

MPPPKHFRLN
MPPPKHFRLN
MPPPKHFRLN
MPPPKHFRLN
MPPPKHFRIN

MPRKGSFSIK
MPRKGSFSIK
MPRKGSFSIK

MPTARAFKIN
MPTARAFKIN
MPQPRRFRIN














Table 2-10. Nucleotide sequence identity (%) between partial sequences from PCR using primer set PBLlv2040/PCRc154 of the B
component of begomoviruses from Andean states and four known begomoviruses
Virus group 1 Virus group 2 Virus group 3 Virus group 4
Clone 2.7-vB3 6-vB2 2.2-vB6 8-vB12 10-vB15 28-vB2 PYMV-VE 2.4-vB24 ToMoTV ToGMoV 57-vB2 DiYMoV
2.4-vB34 99.01 95.37 92.00 91.43 91.91 91.22 94
2.7-vB3 95.04 91.90 90.92 91.58 90.89 93
6-vB2 94.00 91.90 93.21 92.88 92
2.2-vB6 88.00 95.10 92.10 92


87.56


87.33
93.20


94.86 86
86


56-vB7
57vB2
Virus group 1 related to Potato yellow mosaic virus-Venezuela (PYMV-VE)
Virus group 2 related to Tomato Mottle Taino virus (ToMoTV)
Virus group 3 related to Tomato golden mottle virus (ToGMoV)
Virus group 4 related to Dicleptera yellow mottle virus (DiYMoV)


96.55 86
86


8-vB12
10-vB15
28-vB2

2.8-vB8
2.4-vB24

2.9-vB20









tree. This is probably a result of the presence of fewer B component sequences in the

GenBank database. Greater number of sequences allow for greater precision in

phylogenetic analysis.

Discussion

Begomoviruses were detected in almost of samples collected from tomato plants

in commercial fields in Venezuela. Begomoviruses were detected in nine out of the ten

Venezuelan states included in this study. These are Aragua, Barinas, Guarico, Lara,

Merida, Portuguesa, Tachira, Trujillo and Zulia states. This study reports the presence of

begomoviruses in tomato plants for the first time in the states of Barinas, Merida,

Tachira, and Trujillo state. Until this study, tomato-infecting begomoviruses had only

been reported from Lara, Aragua, Guarico, Monagas, Portuguesa, and Zulia (Debrot et al.

1963; Guzman et al. 1997; Lastra and Uzcategui, 1975; Nava et al. 1996). The greatest

number of samples positive for at least one begomovirus came from the states of Barinas,

Guarico, Lara and Zulia.

There appeared to be at least four different begomoviruses in the Andean states.

Three types of analyses: GAP, iteron and Rep iteron-related domain comparison, and

phylogenetic analysis of the partial sequences of the A component supported the presence

of four groups of sequences (Tables 2-8, 2-9, 2-10 and Figure 2-2). The individual

sequences remained in the same groups regardless of the analysis. All four begomovirus

sequence groups appear to be New World in origin. This is the first report of

begomovirus sequences in tomato plants from the Andean states of Venezuela. The

variety of begomovirus sequences found in this study is significant, as these were found










ToLCMV
2.8-v
'OD 2.9-vl
2.7-vl
PYMTV-TT
ToMoTV
ToMoV-[FL]
JMV
ToLCSinV
CdTV-[IC]
CdTV
SiGMHV
100 SiGMCRV
8o BDMV
RhGMV
57-v
L56-v4
ToVEV
100 10-vl
1002.4-v6
2.4-v7
82-v7
MGMV-[JM]
-PYMV-VE
-- PYMV-VE-[tom]
96 PYMV-Mart
10,oPYMV-[GP]
SPYMV-[DR]
2.2-v37
-96-v
S28-v5
1 10-v4
8082-vl
82-v5


100.00
Figure 2-2. Phylogenetic tree constructed based on the nucleotide sequence of the entire
fragment sequenced from amplification using primer set
PARlc496/PALlv1978 for the A component. Tree was generated using
PAUP program. A single most parsimonious tree was predicted by a heuristic
search with stepwise addition, random branch-swapping, tree-bisection-
reconnection options (500 replication for bootstrapping). Bootstrap indices are
shown at each node. Scale bar references branch length as frequency of
changes per site











MaYMFV
BGYMV-[PR]
-- 60
56-vB7
100
S57-vB2
SiGMCRV
DIY1IV
ToGMoV
2.8-vB8
72.9-vB20
2.4-vB24
ToDLCV
93- ToMoV-[FL]
100
ToMoTV
PYMV-VE
PYMTV-TT
PYMPV
S 10 o PYiV-[GP]
9"PYMV-[DR]
100
Loo IPYMV-[MAR]
28-vB2
100
2.2-vB6
I- 10-vB15
6-vB2
L 8-vB12
6r72.4-vB34
100
2.7-vB3

100.00

Figure 2-3. Phylogenetic tree constructed based on the nucleotide sequence of the entire
fragment sequenced from amplification using primer set
PBL v2040/PCRc154 for the B component. Tree was generated using PAUP
program. A single most parsimonious tree was predicted by a heuristic search
with stepwise addition, random branch-swapping, tree-bisection-reconnection
options (500 replication for bootstrapping). Bootstrap indices are shown at
each node. Scale bar references branch length as frequency of changes per
site.









in just 15 samples which were positive for a begomovirus from a total of 68 plants

sampled. Similarly, numerous variations in begomovirus sequences have been reported

from tomato plants in Brazil (seven species) and Trinidad & Tobago (three species)

(Ribeiro et al. 2003; Umaharan et al. 1998).

Analysis of the partial sequence of the A and B components from group 1 and 2

suggests that these begomovirus components are closely related to ToVEV, PYMV-VE-

[tom], and PYMV-VE (Tables 2-8 and 2-10, Figures 2-2 and 2-3), which were previously

reported in Venezuela (Debrot et al. 1963; Guzman et al. 1997; Roberts et al. 1986).

PYMV-VE and ToVEV were first reported in Venezuela in tomato samples from Aragua

state, and PYMV-VE-[tom] was also first reported in Venezuela in tomato growing areas

in Monagas, Guarico and Portuguesa states (Debrot et al. 1963; Guzman et al. 1997;

Roberts et al. 1986). This suggests that these viruses represented by these partial

sequences are established in Venezuela. This study has found PYMV-VE, and PYMV-

VE-[tom], are widely distributed within the Andean states. ToVEV was found in two

Andean states (Trujillo and Merida). The Andean states are separated from the states of

Aragua, Monagas, Guarico and Portuguesa by more than 370 Km. It is unexpected that

these distant tomato production regions share some of the same begomoviruses. The

dissemination of these begomoviruses among states might be possible by the movement

of Bemisia tabaci biotype B or more likely by the movement of infected plants.

Analysis of partial sequences also suggests that there are two unique and

uncharacterized begomoviruses present in samples from Trujillo and Merida states

(groups 3 and 4). Only the full-length genome sequences of the A component are

considered for comparative analysis and 89% NSI has been proposed to demarcate









species. Since recombination events have been shown in this genus, partial sequences are

not sufficient to distinguish new species (Fauquet et al. 2003). Studies are in progress to

obtain complete sequences of the genomes of begomoviruses from appropriate samples to

determine if they are indeed new species of begomoviruses.

This study has shown the presence of four begomoviruses in tomato in Andean

states in Venezuela. Two of these viruses are tentatively considered new species of

begomoviruses. In addition, a high percent of samples collected from tomato fields were

positive for the presence of at least one begomovirus. These data suggest that

begomoviruses may be a bigger concern for the Venezuelan tomato industry than was

previously believed.














CHAPTER 3
CHARACTERIZATION OF A NEW BEGOMOVIRUS FROM THE STATE OF
TRUJILLO, VENEZUELA

Introduction

Begomoviruses are members of the Geminiviridae family, and are characterized

by their circular single-strand DNA genomes enclosed within a distinctive geminate

capsid. Begomoviruses are transmitted to dicotyledonous plants by a whitefly vector,

Bemisia tabaci (Fauquet et al. 2003). Begomoviruses have been reported in the Americas

since the middle 1960s. Several epidemics have been reported which caused serious

problems in different crops in the Americas (Engel et al. 1998; Morales and Anderson,

2001; Polston and Anderson, 1997), especially after the introduction of biotype B of

B. tabaci and after changes in traditional cropping systems (Morales and Anderson, 2001;

Polston and Anderson, 1997).

Four begomoviruses have been reported affecting tomatoes in Venezuela. Tomato

yellow mosaic virus (ToYMV) was the first begomovirus reported to affect tomatoes in

Aragua state, Venezuela (Debrot et al. 1963). Recently a partial sequence of ToYMV was

generated and considered closely related to Potato yellow mosaic virus-Venezuela

(PYMV-VE) (Morales et al. 2001). PYMV-VE was reported as the causal agent of a

yellow mosaic disease affecting potato crops (Coutts et al. 1991; Roberts et al. 1988). A

tomato strain of PYMV-VE and Tomato Venezuela virus were also reported to affect

tomatoes in Venezuela (Guzman et al. 1997).









The presence of two new begomoviruses has also been reported in tomatoes from

Andean states of Venezuela (Chapter 2). One of these two begomoviruses was detected in

the state of Trujillo. Based on the nucleotide sequence identity (NSI) (86%) of partial

sequences, this virus is distantly related to Tomato leaf curl Sinaloa virus (ToLCSinV).

The partial sequences were obtained by the polymerase chain reaction (PCR) using

primers PARlc496 and PAL1v1978 (Rojas et al. 1993). Using the newly established

criteria to distinguish species of begomoviruses (Fauquet et al. 2003), the full-length

sequence of the genomic components must be determined to confirm that the virus

detected from Trujillo is a new and unique begomovirus.

The objectives of this research were:

* Characterize a begomovirus that was found in tomato in Trujillo state, Venezuela

* Generate full-length sequences of the A and B components of this new virus.

Materials and Methods

Plant Sample and Extraction of Genomic DNA

Young leaves of a symptomatic tomato plant were collected as part of a survey of

viruses infecting tomato in Trujillo state, Venezuela (Nava et al. 1997). The symptoms on

this plant included curly leaves, chlorotic leaf margins, foliar deformation and reduced

leaf size (Figure 3-1). This sample, designated 2.9-v, appeared to be an undescribed

begomovirus based on comparison of DNA partial sequences (Chapter 2). Genomic DNA

was extracted from frozen, desiccated leaf tissue (Rath et al. 1998) for use in genomic

characterization studies.
































Figure 3-1. Symptoms of sample 2.9-v collected in Trujillo state, Venezuela. Curly
leaves, yellow margin of the leaves, mosaic and reduction of the size of the
leaves are shown

Obtaining Full-Length Sequences

Degenerate primers PAR1c496/PALlv1987 and PBL1v2040/PCRcl54 (Rojas et

al. 1993), were used to generate partial sequences of A and B components from sample

2.9-v. From these partial sequences, two sets of specific primers (JAP122/JAP123 and

JAP128/JAP129) (Table 3-1) were designed to amplify the remaining sequence of the A

and B components. A Wisconsin Package Version 10.3, [Accelrys (GCG), San Diego,

CA] was used to design the primers. The PCR parameters were: 35 cycles of denaturation

for 1 min at 940C, primer annealing for 1 min at 570C (JAP122/JAP123) or 62C

(JAP128/JAP129) and primer extension for 1 min at 720C, with an initial denaturation at

94C for 5 min and a final extension of 7 min at 720C. PCR reactions were carried out in

a PE Applied Biosystems GeneAmp PCR System 9700 thermocycler (PE Applied









Biosystems, Foster City, CA). Genomic DNA (-100 ng) extracted from desiccated tissue

was amplified in a volume of 25 tL containing 10 mM Tris-HCl (pH 9), 50 mM KC1 and

1% Triton X-100, 1.5 mM MgC12 (JAP122/JAP123) or 1.25 mM MgC12

(JAP128/JAP129), 250 pM dNTPs, 1.0 mM spermidine, 1 tM of each primer, and

1.25 U of Taq polymerase. PCR products were electrophoresed (1 h at 90 volts) in 1%

agarose gels in Tris-acetate-EDTA buffer, pH 8. Gels were stained with ethidium

bromide (0.0015 mg/mL), viewed with a UV transilluminator. The PCR product was

purified using a QIAquick gel extraction kit (QIAGEN Inc. Valencia, CA), cloned using

the pGEM- T Easy system (Promega Corporation, Madison, WI), and transformed into

XL 1-Blue supercompetent Escherichia coli cells (Stratagene, La Jolla, CA) according to

the manufacturer's instructions. Nucleotide sequences were determined (Ana-Gen

Technologies, INC. Atlanta, GA). Internal unique primers (JAP134/JAP 135) for the A

and (JAP138/JAP139) the B component (Table 3-1) were designed to complete the viral

genome. The PCR conditions for the internal unique primers were: 35 cycles of

denaturation for 1 min at 940C, primer annealing for 1 min at 570C, and primer extension

for 1 min at 720C, with an initial denaturation at 940C for 5 min and a final extension for

7 min at 72C. All amplifications were performed as previously described above, using

2 mM MgC12 for primer set JAP134/JAP135 and 1.5 mM MgC12 for primer set

JAP138/JAP 139, and plasmid DNA diluted 1:100 (v/v) as a template. The plasmid DNA

was derived from clones of the PCR products generated by primer sets JAP122/JAP123

and JAP128/JAP129.









Table 3-1. Specific primers designed to determine the full-length sequence of a new
begomovirus from Trujillo, Venezuela.
Primer Sequence Amplified fragment size(bp)
/DNA component
JAP122 5'-GTCGAACCGGAAAGACAATG-3' 1479 bp / A component
JAP123 5'-ACAGGCCCATGTACAGGAAG-3'
JAP128 5'-ATATGAATCGGGGGAACTGGG-3' 2005 bp / B component
JAP129 5'-GCAGCACAATTAACGGCAAG-3'
JAP134 5'-TATGCCAGTAACGAGCAGTC-3' 528 bp / A component
JAP135 5'-ACCTCCCAAATAAAAACGCC-3'
JAP138 5'-CCCAATTAAATGACCTGGTTCG-3' 654 bp / B component
JAP 139 5'-ACGTTCATCAATTACCCTGTTC-3'
JAP150 5'-TACACTGCAGGGCCCTTTGAG-3' 2525 bp / Full B component
JAP 151 5'-GAGATTCATAGGGCCCAGTCCA-3'
JAP166 5'-GCAATGGCCACTTAGATAG-3' 2560 bp / Full A component
JAP167 5'-CGTAGTGGCCATCTTGA-3'
Restriction-enzyme sites are underlined

Obtaining Infectious Clones

Two sets of overlapping primers, JAP166/JAP167 and JAP150/JAP151

(Table 3-1) were designed to amplify a PCR product of the full-length sequence of each

component, using genomic DNA extracted from sample 2.9-v. The primers

JAP166/JAP167 overlap at anMscI site, which is located in the replication associated

protein (Rep) gene in the A component sequence of 2.9-v (Figure 3-2 A). The primers

JAP150/JAP 151 overlap at an Apal site in the movement protein (MP) gene of the

B component sequence of 2.9-v (Figure 3-2 B). The PCR conditions for the overlapping

primers were: 10 cycles of denaturation for 30 s at 94C, primer annealing for 30 s at

52C for primer set JAP166/JAP167 and 60C for primer set JAP150/JAP151, and primer

extension for 2.5 min at 72C, following by 25 cycles in which gradually the extension

time was increased in 10 s/cycle, an initial denaturation at 94C for 2 min and a final

extension for 7 min at 72C. PCR reactions were carried out in a MJ Research PTC- 200

DNA Engine thermocycler (MJ Research, Inc. Waltham, Massachusetts). Genomic DNA









(-100 ng) extracted from desiccated tissue was amplified in a volume of 25 [L

containing 10 mM Tris-HCl (pH 9), 50 mM KC1 and 1% Triton X-100, 2.5 mM MgC12,

100 pM dNTPs, 0.4 [M of each primer, and 1.25 U of Expand High Fidelity Plus Taq

DNA polymerase (Roche Diagnostics Corporation, Indianapolis, IN). The PCR products

were purified using a QIAGEN kit, then ligated using the pGEM-T Easy system, and

transformed into XL-1 Blue supercompetent E. coli cells.

Comparison of DNA and Amino Acid Sequences

The full-length sequences of the A and B components were analyzed using the

Wisconsin Package Version 10.3, Accelrys (GCG). Similarities between a query

sequence and all the sequences in the database were established using Basic Local

Alignment Search Tool (BLAST). Full-length sequences of 15 begomoviruses were

selected based on their high percentage of identity with the A and B component

sequences obtained from 2.9-v sample as established by BLAST. The values of NSI and

amino acid sequence similarity were generated using GAP analysis. Multiple sequence

comparisons between the 15 known begomoviruses and the sequence from 2.9-v sample

were performed using PILEUP.

Phylogenetic Analysis

Phylogenetic analysis was performed using PAUP* (Phylogenetic Analysis Using

Parsimony) [Wisconsin Package Version 10.3, Accelrys (GCG)]. A maximum parsimony

and heuristic tree search was used to create the best tree with stepwise addition, tree-

bisection-reconnection, and random branch-swapping options. The reliability of the tree

was estimated by performing 500 bootstrap repetitions. The selected 15 sequences of

begomoviruses (Table 3-2) were used for DNA and amino acid sequence comparison,

and also for phylogenetic analysis.









Table 3-2. Acronyms and accession numbers of known begomoviruses used for nucleic
acid and amino acid sequence comparison and phylogenetic analysis of
2.9-v virus.


Species
Bean calico mosaic virus


Acronym
BCaMV


Bean dwarf mosaic virus
Bean golden mosaic virus- Brazil
Cabbage leaf curl virus
Cucurbit leaf curl virus-[Arizona]

Chino del tomate virus-[H8]

Dicliptera yellow mottle virus

Pepper golden mosaic virus
Potato yellow mosaic Trinidad virus-Trinidad &
Tobago
Sida golden mosaic Costa Rica virus
Sida golden mosaic Honduras virus
Sida golden mosaic virus
Tomato chlorotic mottle virus-[Brazil]

Tomato mottle virus- [Florida]
Tomato mottle Taino virus


BDMV
BGMV-[BZ]
CaLCuV
CuLCuV-
[AZ]
CdTV-[H8]

DiYMoV

PepGMV
PYMTV-TT

SiGMCRV
SiGMHV
SiGMV
ToCMoV-
[BZ]
ToMoV-[FL]
ToMoTV


Accession number
AF110189,
AF110190
M88179, M88180
M88686, M88687
U65529, U65530
AF256200,
AF327559
AF101476,
AF101478
AF139168,
AF170101
U57457, AF499442
AF039031,
AF039032
X99550, X99551
Y11097, Y11098
U77963, AF039841
AF90004,
AF491306
L14460, L14461
AF012300,
AF012301















AC4 (2007 2261 nt)

Rep (1345 2418 nt)

JAP122 (1729 nt)
PAL1v1978 (1724 nt)
JAP 166 (1684 nt)
JAP 167 (1682 nt)


TrAP (1029 1421


P4


JAP128 (2446 nt)


JAP129 (1926 nt)


PBL1v2040 (1893 nt)

MP (1220 2059 nt)


JAP151 (1228 nt)


JAP123 (250 nt)
PAR1c496 (255 nt)

CP (138 881 nt)

JAP134 (684 nt)



REn (884 1279 nt)
JAP135 (1212 nt)
nt)



CRc154 (2517 nt)





JAP139 (588 nt)

S NSP (285 1016 nt)


JAP138 (1242 nt)


JAP150 (1279 nt)


Figure 3-2. Genome organization of 2.9-v sequence a bipartite begomovirus from Trujillo
state, Venezuela. A) Open reading frames of the A component; B) Open
reading frames of the B component. Start and end position of each ORF,
primers and binding sites are shown. CP = Coat protein, TrAP =
Transcriptional- activator protein, REn = Replication-enhancer protein, AC4
= AC4 protein, Rep = Replication- associatedprotein, MP = Movement
protein, NSP = Nuclear- shuttle protein.









Results

Full-Length Sequencing

The full-length sequences of the A and B components were obtained from sample

2.9-v. The common region shared by both components contains 159 nt. The NSI value of

this common region was 97.16%, and therefore both genomes belong to the same

bipartite begomovirus, designated 2.9-v virus. The genome organization of 2.9-v virus

was similar to those of bipartite begomoviruses from the New World. The A component

has five putative genes (Figure 3-2A) and the B component has two putative genes

(Figure 3-2B). Conserved sequences for begomoviruses were present in the intergenic

region of the A component, such as a nonameric motif 5'-TAATATTAC-3', one inverted

repeat (5'-CCCCCAATT-3'), two forward repeats (5'-AAYYGGGGG-3'), and a TATA

box, which is located before the hairpin element sequence. The core sequences of the

iteron or forward repeat (5'-GGGGG-3'), and the iteron-related domain of the replication

protein (GSFSIK) were most similar to those of PVMV-VE, ToMoTV, and DiYMoV

(Argtello-Astorga and Ruiz-Medrano, 2001).

Full-length sequences of each component of 2.9-v virus were generated using

overlapping primers JAP150/JAP151 and JAP166/JAP167. Full-length clones were

constructed.

Comparison of DNA and Amino Acid Sequences

The values of NSI between 2.9-v virus and known begomoviruses for the full-

length sequence of the A component were low (68.5-78.39%) (Table 3-3). The NSI

values are lower than the cut-off value to distinguish different species of begomoviruses

(Fauquet et al. 2003), which confirms that 2.9-v virus is an uncharacterized begomovirus.

The highest NSI value for the full 2.9-v A component (78.39%) and Nuclea-shuttle









protein (NSP) gene (70.35%) was with the NSP of SGMCRV (Table 3-3). The highest

NSI value (82.17%) for the 2.9-v coat protein (CP) gene was with the CP of PepGMV.

The highest NSI value for 2.9-v transcriptional activator protein (TrAP), and replication

enhancer protein (REn) genes were with those of BCaMV. The highest NSI value for

2.9 AC4 gene were with those of PYMTV-TT and CdTV-[H8]. The sequence of the

replication associated protein (Rep) gene and the common region had 86.53 % and

78.52% of NSI with ToMoV-[FL], respectively. The highest value of NSI for the

movement protein (MP) gene was obtained between 2.9-v virus and DiYMoV

(Table 3-3).

Table 3-3. Percent of nucleic acid identity between 2.9-v virus sequence and 15 known
begomoviruses
A B
Virus* Comp Comp CR CP TrAP REn AC4 Rep MP NSP
SiGMCRV 78.39 67.31 61.81 79.57 74.41 80.51 85.66 78.02 78.21 70.35
PYMTV-TT 77.95 64.91 67.85 78.76 76.54 81.36 89.92 79.11 76.31 64.82
SiGMHV 77.55 64.60 71.97 78.76 73.97 78.09 88.37 78.49 77.02 70.35
BDMV 77.50 64.54 68.84 81.45 74.22 78.09 87.98 77.74 76.67 68.51
ToMoTV 76.17 63.90 78.52 79.44 70.61 77.33 88.76 78.73 77.86 68.19
DiYMoV 76.08 67.55 70.80 81.37 76.61 76.13 83.33 77.16 78.65 69.27
CdTV-[H8] 76.02 65.30 69.53 80.16 73.27 77.89 89.88 77.75 77.22 69.88
ToMoV-[FL] 75.26 65.93 65.65 78.68 71.72 77.38 86.82 86.53 77.22 67.58
BGMV-[BZ] 74.62 64.41 68.84 79.38 78.97 81.20 72.94 72.42 73.55 67.75
ToCMoV-[BZ] 74.21 61.81 68.64 78.98 75.13 78.69 82.17 73.53 70.82 65.14
SiGMFV-A1 74.07 64.25 72.00 78.87 73.70 75.44 85.99 74.86 77.58 68.64
BcaMV 72.52 61.55 62.59 79.65 81.03 82.46 56.91 69.16 71.31 66.21
CaLCuV 72.10 66.39 51.10 81.96 74.67 75.13 56.08 68.05 76.51 69.57
CuLCuV-AZ 70.88 61.71 45.93 80.29 76.23 77.89 53.72 66.35 72.12 64.98
PepGMV 68.50 62.54 43.18 82.17 74.81 75.63 NA 63.65 74.49 69.11
* Virus names and accession numbers are described in material and methods.
NA = Not available
A Comp = A component, B Comp = B component, CR = Common region, CP = Coat
protein, TrAP = Transcriptional-activator protein, REn = Replication-enhancer protein,
AC4 protein = AC4, Rep = Replication-associatedprotein, MP = Movement protein, NSP
= Nuclear-shuttle protein.









Amino acid sequence similarities between 2.9-v virus and known begomoviruses

were higher (ranging from 32.24% to 94.38%) than the NSI values for almost all the open

reading frames, except for AC4 (Table 3-4). Higher values of similarity were obtained for

the CP, followed by the similarity values of the Rep, REn, TrAP, and AC4. The values of

similarity were higher for the MP than those for the NSP (Table 3-4).

Phylogenetic analysis

The relationships between the 2.9-v virus sequence and 15 known begomoviruses

were analyzed using the full-length A and B component sequences as well as the

sequences of the CP and Rep genes. The phylogenetic trees of the full-length sequences

of the A and B components revealed that 2.9-v sequence is in the DiYMoV clade and

Table 3-4. Percent of similarity of amino acid sequences between proteins of 2.9-v virus
and proteins of 15 known begomoviruses
Virus* CP TrAP Ren AC4 Rep MP NSP
SiGMCRV 92.77 69.77 81.82 73.26 85.03 86.83 75.68
PYMTV-TT 93.57 71.52 82.71 82.56 87.18 87.54 69.73
SGMHV 91.57 67.69 79.70 76.74 84.60 87.54 79.26
BDMV 94.38 71.54 81.95 77.90 84.55 88.97 76.03
ToMoTV 91.97 66.92 80.45 79.10 82.40 87.90 72.72
DiYMoV 93.57 67.69 80.45 82.56 83.01 86.48 74.31
CdTV-[H8] 93.57 67.69 80.45 82.56 83.08 86.47 74.31
ToMoV-[FL] 91.57 66.92 81.20 74.12 81.06 89.68 71.56
BGMV-[BZ] 94.78 78.46 81.20 51.77 79.11 85.41 74.88
ToCMoV-[BZ] 91.97 72.31 82.71 66.28 81.25 83.63 72.48
SiGMFV-A1 91.57 67.18 76.52 73.26 83.38 86.61 75.34
BcaMV 93.98 78.46 84.96 34.15 73.28 86.43 72.73
CaLCuV 94.38 71.54 75.94 32.94 71.26 86.12 73.85
CuLCuV-AZ 93.17 81.20 50.00 NA 71.71 81.45 65.14
PepGMV 93.57 72.31 79.70 NA 69.65 84.84 74.77
= Virus names and accession numbers are described in material and methods.
NA = Not available.
CP = Coat protein, TrAP = Transcriptional-activator protein, REn = Replication-enhancer
protein, Rep = Replication-associated protein, AC4 = AC4 protein, MP = Movement
protein, NSP = Nuclear-shuttle protein.









clustered with CaLCuV and PepGMV (Figures 3-3 and 3-4). The phylogenetic trees for

the Rep and the CP confirm that 2.9-v virus is closely related to DiYMoV (Figures 3-5

and 3-6).

Discussion

The full-length sequences of the DNA-A and DNA-B of 2.9-v virus were generated. The

A and the B components had 2,560 nt and the 2,525 nt, respectively (Figure 3-2).

Although, the B component DNA of 2.9-v virus appears to be smaller than other

bipartite-begomovirus genomes, smaller B DNA components have been reported for

several begomoviruses (Abouzid et al. 1992a; Abouzid, et al. 1992b; Coutts et al. 1991;

Umaharan et al. 1998). The inferred-genome organization of 2.9-v virus was similar to

those of bipartite begomoviruses from the New World (Figure 3-2). The A component

contains four genes in complementary sense (Rep, TrAP, REn, AC4) and one gene (CP)

in viral sense. The B component contains two genes MP and NSP (Gutierrez, 2000). The

conserved sequences in the IR are similar to those of other New World bipartite

begomoviruses. The core of the forward repeat or iteron core (GGGGG) and the Rep

iteron-related domain sequences (GSFSIK) are similar to those of PYMV-VE. PYMV-

VE has two inverted and forward repeats (Arguello-Astorga et al. 2001). However, one

inverted and two forward repeats were observed in the IR of 2.9v virus. These results

confirm that 2.9-v virus is a unique begomovirus from the New World.

The NSI values for the full-length of the A component between of 2.9-v virus

sequence and 15 known begomoviruses were always less than 89% (Table 3-3), which

confirmed that 2.9-v virus sequence is a distinct begomovirus (Fauquet et al. 2003). The












2.9-v A


CaLCuV


100 PepGMV
100
1 CuLCuV-AZ
-- 61
BcaMV

BGMV-[BZ]

ToCMoV-[BZ]

51- PYMTV-TT
51

ToMoTV
100
86
I ToMoV-[FL]

98 CdTV-[H8]

SiGMHV
79
100
SiGMV
96
ISiGMCRV
98
S BDMV



100.00


Figure 3-3. Phylogenetic tree based on the complete nucleotide sequence of the A
component of the 2.9-v sample from Trujillo state and fifteen bipartite
begomoviruses.The tree was generated using the PAUP program. A single
most parsimonious tree was predicted by a heuristic search with stepwise
addition, random branch-swapping, tree-bisection-reconnection options (500
replication for bootstrapping). Bootstrap indices are shown at each node. Scale
bar references branch length as frequency of changes per site











BGMV-[BZ]


ToCMoV-[BZ]

- BcaMV


CuLCuV-AZ


PepGMV


100.00


Figure 3-4. Phylogenetic tree based on the complete nucleotide sequence of the B
component of the 2.9-v sample from Trujillo state and fifteen bipartite
begomoviruses. The tree was generated using the PAUP program. A single
most parsimonious tree was predicted by a heuristic search with stepwise
addition, random branch-swapping, tree-bisection-reconnection options (500
replication for bootstrapping). Bootstrap indices are shown at each node. Scale
bar references branch length as frequency of changes per site







57



CuLCuV-AZ

BCaMV

ToCMoV-[BZ]

BGMV-[BZ]

2.9-v A

DiYMoV

CaLCuV
56
PepGMV

ToMoV-[FL]

SiGMCRV

BDMV

52 ToMoTV
-- 79
PYMTV-TT

CDTV-[H8]

58 SiGMHV
100
SiGMV


100.00


Figure 3-5. Phylogenetic tree based on the complete coat protein nucleotide sequence of
the 2.9-v sample from Trujillo state and fifteen bipartite begomoviruses.The
tree was generated using the PAUP program. A single most parsimonious tree
was predicted by a heuristic search with stepwise addition, random branch-
swapping, tree-bisection-reconnection options (500 replication for
bootstrapping). Bootstrap indices are shown at each node. Scale bar references
branch length as frequency of changes per site











BCaMV

- CuLCuV-AZ

-- CaLCuV

PepGMV


100.00

Figure 3-6. Phylogenetic tree based on the complete replication-associatedprotein
nucleotide sequence of the 2.9-v sample from Trujillo state and fifteen
bipartite begomoviruses.The tree was generated using the PAUP program. A
single most parsimonious tree was predicted by a heuristic search with
stepwise addition, random branch-swapping, tree-bisection-reconnection
options (500 replication for bootstrapping). Bootstrap indices are shown at
each node. Scale bar references branch length as frequency of changes per site









values of NSI were noticeably higher for the CP. This is not surprising since the CP

sequence is very conserved within the begomovirus genus (Brown et al. 2001; Padidam

et al. 1995). The CP determines vector specificity in begomoviruses. This fact explains

why almost all the begomoviruses are transmitted by whiteflies (Timmermans et al.

1994; van den Heuvel et al. 1999).

Full-length clones for the A and B components have been constructed from

purified PCR product using overlapping primer methodology (Patel et al. 1993).

Determination of the host range using biolistic inoculation is in progress. Symptoms in

inoculated plants will be recorded and correlated with those observed in the 2.9-v sample.

Mixed infection can occur in the field, thus this correlation may be of little practical use.

Therefore, the viral characterization of 2.9-v virus was based on molecular data.

The 2.9-v virus is a begomovirus from the New World closely related to DiYMoV

based on phylogenetic trees of the full-length sequence of the A and the B components

(Figures 3-3 and 3-4). DiYMoV and 2.9-v virus appeared to have a common ancestor,

based on phylogenetic trees of the full-length of the A and B components, and of the Rep

gene (Figures 3-3, 3-4, and 3-6). The begomoviruses clustered with the 2.9-v virus do not

have tomato as a host with the exception of PepGMV (Torres-Pacheco et al. 1996)

(Figures 3-3, 3-4, and 3-6). The 2.9-v virus is probably a distinct species, albeit related to

DiYMoV. The phylogenetic tree of the sequence of the CP was not congruent with the

other trees generated in this research (Figure 3-5). However, the relationship between the

2.9-v virus and DiYMoV was maintained in this tree (Figure 3-5). Remarkably, 2.9-v

virus has the longest branch length in the CP tree, which implies that 2.9-virus is more

divergent from the other begomoviruses in that clade (Baldauf, 2003) (Figure 3-5).






60


Possibly, the 2.9-v virus evolved and acquired the ability to infect tomato crops. These

results confirm that 2.9-v is a distinct begomovirus.

Since the 2.9-v virus is an undescribed bipartite begomovirus from Trujillo state,

Venezuela, it is necessary to determine its distribution in other states where tomato crops

are grown in Venezuela. Thus, it is possible to infer the actual importance of this virus in

Venezuela. It could be the first step to initiate a management program ofbegomoviruses

in tomato crops.














CHAPTER 4
CHARACTERIZATION OF A NEW BEGOMOVIRUS FROM THE STATE OF
MERIDA, VENEZUELA

Introduction

Begomoviruses are members of the family Geminiviridae, they have circular

single-strand DNA genomes packaged within twinned particles, and are transmitted by

the whitefly, Bemisia tabaci, to dicotyledonous plants (Fauquet et al. 2003).

Begomoviruses are a limiting factor of many crops in the Americas, particularly with the

introduction of the polyphagous biotype B of Bemisia tabaci and changes in traditional

cropping systems (Morales and Anderson, 2001; Polston and Anderson, 1997). In

addition, the misuse of insecticides has increased the development of pesticide-resistant

populations of this whitefly and the elimination of natural enemies of the whitefly

resulting in epidemics of geminivirus-like symptoms in the Americas (Engel et al. 1998;

Morales and Anderson, 2001; Polston and Anderson, 1997).

A begomovirus has been reported affecting tomatoes in Venezuela. Tomato

yellow mosaic virus (ToYMV) was the first begomovirus reported to affect tomatoes in

Venezuela (Debrot et al. 1963). The symptoms produced by this virus were described as

yellow mosaic, stunting, and upward cupping and deformation of the leaves (Debrot et al.

1963). Whitefly transmission and physical properties of ToYMV were published in 1978

(Uzcategui and Lastra, 1978).

A yellow mosaic disease affecting potato crops was first reported in Venezuela

(Debrot and Centeno, 1985a). The causal agent of this disease was determined to be a









bipartite begomovirus and named Potato yellow mosaic virus-Venezuela (PYMV-VE)

(Roberts et al. 1988). The nucleotide sequence of PYMV-VE indicated that it is a

begomovirus of the New World (Coutts et al. 1991). Several strains of PYMV have been

reported from Central and South America, and the Caribbean. A bipartite begomovirus

closely related to PYMV-VE was reported to infect tomato in Panama (Engel et al.

1998). The virus was molecularly characterized and called Potato yellow Panama mosaic

virus (PYMPV) (formerly named Tomato leaf curl virus-Panama). High amino acid

sequence similarity between PYMPV and PYVM-VE was determined for the open

reading frames with the exception of the AC4 gene product. Venezuela was proposed as

the possible origin of PYMPV because of the close relationship between these two

viruses (Engel et al. 1998). Another bipartite begomovirus was reported to infect pepper,

sweet pepper, okra, beans, and several weeds in different locations in Trinidad

(Umaharan et al. 1998). The virus was named Potato yellow mosaic Trinidad virus-

Trinidad & Tobago (PYMTV-TT). The A component of PYMTV-TT has 85% nucleotide

sequence identity (NSI) with PYMV-VE and, and may be a recombinant between either

PYMV-VE or PYMPV and Sida golden mosaic Honduras virus (SiGMHV) (Umaharan

et al. 1998). Potato yellow mosaic virus-[Guadeloupe] (PYMV-[GP]) was first reported

in 1998 as a strain of PYMV-VE (Polston and Bois, 1998). It was responsible for

epidemics of virus-like symptoms in tomato in Guadeloupe, Martinique, and Puerto Rico.

It was suggested that PYMV-[GP] could be a recent introduction based on high value of

NSI between begomoviruses sequences at distant locations (Polston and Bois, 1998).

Phylogenetic analysis of the A component of PYMTV-[GP] revealed that this virus is

closely related to PYMV-VE, and PYMTV-TT (Urbino et al. in press).









Although partial sequences of the A component have been used to identify

PYMV-VE strains such as PYMV-Martinique, PYMV-Puerto Rico, PYMV-Dominican

Republic, Tomato yellow mosaic virus and PYMV-VE strain tomato (Guzman et al.

1997; Morales et al. 2001; Polston and Bois, 1998), only full-length sequences of the A

component are considered for determining taxonomic status in begomoviruses, and the

species demarcation would be < 89% of NSI. Thus, virus isolates with >89% of NSI may

be considered to be the species and should have the same name, irrespective of the host

from which they were derived. Further designation into strains may be justified by

demonstration of biological differences (Fauquet et al. 2003).

The objectives of this research were:

* Characterize a new begomovirus that was found in tomato in Merida state
Venezuela

* Generate full-length sequences of the A and B components of this new virus.

Materials and Methods

Plant Sample and Extraction of Genomic DNA

A leaf sample from a plant with mottled and severely deformed leaves (Figure 4-

1) was collected as part of a survey of viruses in tomato in Merida state, Venezuela (Nava

et al. 1997). This sample, designated 57-v, appeared to be a unique begomovirus based

on results from a study which compared partial DNA sequences amplified from tomato

samples collected from 10 states in Venezuela (Chapter 2). Genomic DNA was extracted

from frozen, desiccated leaf tissue (Rath et al. 1998).

Obtaining Full-Length Sequences

Partial sequences of A and B components from sample 57-v were generated using

degenerate primers PAR1c496/PAL1v1987 and PBL1v2040/PCRcl54 (Rojas et al.





























Figure 4-1. Symptoms of sample 57-v collected in Merida state, Venezuela. Mottling and
severe deformation of the leaves were observed.

1993). From these partial sequences two sets of specific primers (JAP124/JAP125 and

JAP130/JAP131) (Table 4-1) were designed to amplify the remainder of the A and B

component sequences. The specific primers were designed using a Wisconsin Package

Version 10.3, [Accelrys (GCG), San Diego, CA]. The PCR parameters were: 35 cycles of

denaturation for 1 min at 940C, primer annealing for 1 min at 550C (JAP124/JAP125) or

62C (JAP130/JAP131) and primer extension for 1 min at 720C, with an initial

denaturation at 940 for 5 min and a final extension of 7 min at 720C. PCR reactions were

carried out in a PE Applied Biosystems GeneAmp PCR System 9700 thermocycler (PE

Applied Biosystems, Foster City, CA). Genomic DNA (-100 ng) extracted from

desiccated tissue was amplified in a volume of 25 tL containing 10 mM Tris-HCl (pH 9),

50 mM KC1 and 1% Triton X-100, 1.5 mM MgCl2 (JAP124/JAP125) or 1.75 mM

MgCl2 (JAP130/JAP131), 250 pM dNTPs, 1.0 mM spermidine, 1 tM of each primer,

and 1.25 U of Taq polymerase. PCR products were electrophoresed (1 h at 90 volts) in









1% agarose gels in Tris-acetate-EDTA buffer, pH 8. Gels were stained with ethidium

bromide (0.0015 mg/mL)and viewed with a UV transilluminator. The PCR product was

purified using QIAquick gel extraction kit (QIAGEN Inc. Valencia, CA), cloned using

the pGEM- T easy system (Promega Corporation, Madison, WI), and transformed into

XL 1-Blue supercompetent Escherichia coli cells (Startagene, La Jolla, CA) according to

manufacturer's instructions. Nucleotide sequences were obtained from the partial

sequences and combined (Ana-Gen Technologies, INC. Atlanta, GA).

Internal unique primers (JAP136/JAP137) for the A and (JAP140/JAP141) the B

component (Table 4-1) were designed to complete the viral genome. The PCR conditions

for the internal unique primers were: 35 cycles of denaturation for 1 min at 940C, primer

annealing for 1 min at 550C (JAP136/JAP137) or 57C (JAP140/Japl41), and primer

extension for 1 min at 720C, with an initial denaturation at 940C for 5 min and a final

extension for 7 min at 72C. All amplifications were performed as previously described

above, using 1.5 mM MgC12 for both sets of primers, and plasmid DNA diluted 1:100

(v/v) as the template. The plasmid DNA was derived from clones of the PCR products

generated by primer sets JAP124/JAP125 and JAP130/JAP131.

Obtaining Full-Length Clones

Two sets of overlapping primers, JAP154/JAP155 and JAP168/JAP169

(Table 4-1), were designed to amplify a PCR product of the full-length sequence of each

component, using genomic DNA from sample 57-v. The primers JAP154/JAP155

overlap in a KpnI site, site in the movement protein (MP) gene of the B component

sequence of 57-v (Figure 4-2 B). The PCR conditions for the overlapping primers were:

35 cycles of denaturation for 1 min at 940C, primer annealing for 2 min at 53C









(JAP154/JAP155) or 57C (JAP168/JAP169), and primer extension for 3 min at 720C,

with an initial denaturation at

Table 4-1. Specific primers designed to determine the full-length sequence of 57-v virus
from Merida, Venezuela.
Primer Sequence Amplified-fragment size (bp)
/DNA component
JAP124 5'-CAGTCGTTCCTCCAATTATTCC-3' 1621 bp / A component
JAP125 5'-TGATATGCAAGAATGGGAG-3'
JAP130 5'-CAGTTTCCTTCCACTGCTGC-3' 2079 bp / B component
JAP131 5'-ATTGGAGTCTCTCAACTCTCTC-3'
JAP136 5'-ATGCCAGCAATGAGCAAG-3' 446 bp / A component
JAP 137 5'-GAGTGTACCACATCCAAATCAG-3'
JAP140 5'-CAGTGATAGGGGGAACAAAGGG-3' 857 bp / B component
JAP141 5'-GTGTTAAACGTGCAGATGGG-3'
JAP154 5'-TCACGCGGGTACCTTTTAT-3' 2575 bp / Full A component
JAP155 5'-TTCGCACTGGTACCTAGACA-3'
JAP168 5'-GATAGCAGGACCCCAGTCT-3' 2543 bp / Full B component
JAP169 5'-TTGGGGTCCCTGCACATTAG-3'
Restriction-enzyme sites are underlined

94C for 5 min and a final extension for 10 min at 720C. PCR reactions were carried out

in a PE Applied Biosystems GeneAmp PCR System 9700 (PE Applied Biosystems,

Foster City, CA) thermocycler. Genomic DNA (-100 ng) extracted from desiccated

tissue was amplified in a volume of 25 tL containing 10 mM Tris-HCl (pH 9), 50 mM

KC1 and 1% Triton X-100, 3.0 mM MgC12 (JAP154/JAP155) or 2.0 mM MgC12

(JAP168/JAP169), 250 pM dNTPs, 1.0 mM spermidine, 1 tM of each primer, and 1.25

U of Expand High Fidelity Plus Taq DNA polymerase (Roche Diagnostics Corporation,

Indianapolis, IN). PCR product of the full-length sequence was ligated into pGEM-T

Easy system and transformed in XL-1 Blue supercompetent E. coli cells. The ligation

reaction was most efficient if done the same day that the PCR was performed. It was

assumed that the A overhangs (necessary for cloning with pGEM T Easy) on the ends of

the PCR products were lost during storage. Thus, the formation of blunt-ended fragments









is reduced. In addition, exposure to UV light must be limited as much as possible to

reduce the production of pyrimidine dimers. High efficiency in the cloning was obtained,

when the PCR products of the full-length sequences were not exposed to UV and not

frozen.

Comparison of DNA and Amino Acid Sequences

The full-length sequences of the A and B components were analyzed using the

Wisconsin Package Version 10.3, Accelrys (GCG). Similarities between a query

sequence and all the sequences in the database were established using Basic Local

Alignment Search Tool (BLAST). Full-length sequences of 14 begomoviruses were

selected based on their high percentage of identity with the A and B component

sequences obtained from sample 57-v as established by BLAST. The values of NSI and

amino acid sequence similarity were generated using GAP analysis. Multiple sequence

comparisons between the fourteen known begomoviruses and the sequence from sample

57-v were performed using PILEUP.

Phylogenetic Analysis

Phylogenetic analysis was carried out using PAUP* (Phylogenetic Analysis Using

Parsimony) [Wisconsin Package Version 10.3, Accelrys (GCG)]. A maximum parsimony

and heuristic tree search was used to generate the best tree with stepwise addition, tree-

bisection-reconnection, and random branch-swapping options. The reliability of the tree

was estimated by performing 500 bootstrap repetitions. The selected 14 sequences of

begomoviruses (Table 4-2) were used for DNA and amino acid sequence comparison,

and also for phylogenetic analysis.









Table 4 -2. Acronyms and accession numbers of known begomoviruses used for nucleic
acid and amino acid sequence comparison and phylogenetic analysis of
57-v virus.
Species Acronym Accession number
Bean golden mosaic virus Brazil BGMV-[BZ] M88686, M88687
Cabbage leaf curl virus CaLCuV U65529, U65530
Chino del tomate virus [IC] CdTV-[IC] AF101476,
AF101478
Dicliptera yellow mottle virus DiYMoV AF 139168,
AF170101
Pepper golden mosaic virus PepGMV U57457, AF499442
Potato yellow mosaic Panama virus PYMPV Y15033, Y15034
Potato yellow mosaic virus [Guadeloupe] PYMV-[GP] AY120882,
AY120883
Potato yellow mosaic Trinidad virus-Trinidad & PYMTV-TT AF039031,
Tobago AF039032
Potato yellow mosaic virus-Venezuela PYMV-VE D00940, D00941
Sida golden mosaic Costa Rica virus SiGMCRV X99550, X99551
Tomato chlorotic mottle virus-[Brazil] ToCMoV- AF90004, AF491306
[BZ]
Tomato golden mosaic virus-yellow vein TGMV-YV K02029, K02030
Tomato mottle Taino virus ToMoTV AF012300,
AF012301
Tomato rugose mosaic virus -[Uberlandia] ToRMV-[Ube AF291705,
AF291706


Results

Full-Length Sequencing and Cloning

The full-length sequences of an A and the B component from tomato sample 57-v were

generated. A region of 134 nt was common between the A and B components and the

NSI value of this common region was 99.25%, which confirms that both components

conform to the bipartite begomovirus of 57-v sample. The genome structure deduced

from nucleotide sequence resembled that of other bipartite begomoviruses from the New

World, with five and two putative genes for the A (Figure 4-2A) and the B components

(Figure 4-2B), respectively. Conserved begomovirus sequences were present in the A

component, such as a nonameric motif 5 -TAATATTAC-3' in the intergenic region (IR),









A


AC4 (2041 2301 nt)


Rep (1421 2452 nt)



JAP125 (1852 nt)
PAL1v1978 (1758 nt)


JAP155 (1495 nt) -

JAP154 (1494 nt)





B
JAP131 (2439 nt)



JAP130 (1975 nt)


PBL1v2040 (1927 nt)



MP (1215 2093 nt)
JAP168 (1367 nt)

JAP169 (1364 nt)


TrAP (1072 145


PCRcl5


JAP124 (231 nt)
PAR1c496 (298 nt)


CP (181 924 nt)



JAP136 (728 nt)




REn (927 -1322 nt)

37(1174 nt)
5 nt)




4 (2534 nt)



JAP141 (468 nt)





NSP (383 1153 nt)


JAP140 (1325 nt)


Figure 4-2. Genome organization of 57-v sequence a bipartite begomovirus from Merida
state, Venezuela. A) Open reading frames of the A component; B) Open
reading frames of the B component. Start and end position of each ORF,
primers and binding sites are shown. CP = Coat protein gene, TrAP =
Transcriptional-activator protein, REn = Replication-enhancer protein, AC4
= AC4 protein, Rep = Replication-associatedprotein, MP = Movement
protein, NSP = Nuclear-shuttle protein









this motif is located in the loop of a "hairpin" element, where the Rep protein introduces

a site-specific nick to initiate virus replication via the rolling-circle mechanism (Laufs et

al. 1995). One inverted repeat (5'-ACACCAATT-3'), one forward repeat

(5'-AATCGGTGT-3'), and a TATA box were present before the stem loop sequence. The

core sequences of the iteron or forward repeat (5'-GGKGT-3'), and the iteron-related

domain of the replication protein (FRIN) were most similar to those of BGMV-[BZ], and

Rhynchosia golden mosaic virus (Arguello-Astorga and Ruiz-Medrano, 2001).

Full-length genomic PCR product was generated using primer sets

JAP154/JAP155 and JAP168/JAP169 for the A and B components, respectively. Full-

length clones were constructed for both genomic A and B components.

Comparison of DNA and Amino Acid Sequences

The percentages of NSI between 57-v sample and known begomoviruses for the

full-length sequence of the A component were low (70-77%) (Table 4-3), which

indicated that 57-v sequence was a distinct begomovirus species (Fauquet et al. 2003).

The highest NSI value for the full 57-v A component (77%), replication-enhancer

protein (REn) gene (81.82 %), Rep (87.84 %), and the common region (79.85 %) were

with PYMTV-TT (Table 4-3). The highest NSI value (75.78 %) for the 57-v

transcriptional- activator protein gene (TrAP) was with TGMV-YV. PYMTV-TT and

CdTV-[CI] had the highest values of NSI with the AC4 sequence of the begomovirus

from the 57-v sample. The sequence of the coat protein (CP) had 84.08 % of NSI with

ToCMoV-[BZ] (Table 4-3). Regarding the B component sequence and its genes, the

highest values of NSI were always obtained between 57-v sequences and DiYMoV.









Table 4-3. Percent of nucleotide sequence identity between 57-v virus sequence and 14-
known begomoviruses
Virus A Comp B Comp CR CP TrAP REn AC4 Rep MP NSP
PYMTV-TT 77.07 65.94 79.85 80.11 73.70 81.82 87.84 79.69 76.56 68.75
SiGMCRV 76.32 65.15 63.3. 80.78 73.96 78.79 NA 79.36 76.68 73.57
TGMV-YV 76.27 64.06 65.87 82.86 76.82 78.79 83.46 74.65 79.28 69.79
ToRMV-[Ube] 75.83 64.82 62.69 81.86 75.78 79.80 85.71 76.65 76.11 70.44
ToCMoV-[BZ] 75.40 63.46 74.44 84.08 75.78 81.82 77.31 75.97 72.24 70.83
CdTV-[IC] 75.28 63.84 68.18 79.12 73.64 79.55 87.88 77.97 76.07 72.40
DiYMoV 74.53 66.98 59.70 82.93 72.92 77.78 82.35 75.87 79.07 74.09
PYMV-VE 74.49 64.75 62.12 81.26 73.39 79.45 NA 75.09 76.76 68.87
BGMV-[BZ] 74.26 65.01 69.92 80.65 75.78 80.80 72.94 73.74 74.52 73.39
TTMV 74.10 63.43 71.43 79.92 71.32 78.03 86.43 77.10 76.64 70.82
PYMV-[GP] 74.04 64.73 68.42 79.65 73.13 79.70 81.40 74.69 76.30 68.75
PYMPV 73.85 64.25 64.44 80.32 73.90 81.06 77.52 74.11 74.94 70.04
CaLCuV 72.17 64.76 58.12 82.66 75.52 78.53 54.65 65.09 76.56 72.91
PepGMV 70.14 64.71 55.17 82.79 74.48 79.04 NA 63.71 74.74 72.01
* Virus names and accession numbers are described in material and methods.
NA = Not available. A Comp = A component, B Comp = B component, CR = Common
region, CP = Coat protein gene, TrAP = Transcriptional-activator protein, REn =
Replication-enhancer protein, A C4 = A C4 protein, Rep = Replication-associated protein,
MP = Movement protein, NSP = Nuclear-shuttle protein.

The values of similarity of the amino acid sequences between 57-v virus and

known begomoviruses were higher (ranged from 70% to a 100%) for almost all the open

reading frames, except for AC4 (Table 4-4). Higher values of similarity were obtained for

the CP, followed by the similarity values of the Rep, REn, TrAP, and AC4. The values of

similarity were higher for the MP than those for the nuclear- shuttle protein (NSP)

(Table 4-4).

Phylogenetic Analysis

The relationships between 57-v sequence and known begomoviruses were

analyzed using the genomic A and B components sequences as well as the sequences of

the coat protein (CP) and Rep genes. The phylogenetic trees of the sequences of the A

and B components revealed that 57-v virus is a New World begomovirus (Figures 4-3

and 4-4). The B component sequence of 57-v is clustered with DiYMoV, CaLCuV and









PepGMV (Figure 4-4). The phylogenetic trees for the Rep and the CP confirm that 57-v

is closely related to bipartite begomoviruses from the New World (Figures 4-5 and 4-6).

Table 4-4. Percent of similarity of amino acid sequences between proteins of 57-v virus
and proteins of 14-known begomoviruses
Virus* CP TrAP REn AC4 Rep MP NSP
PYMTV-TT 93.95 67.19 85.61 71.77 89.80 87.42 70.82
SGMCRV 94.35 68.75 84.09 NA 88.08 88.09 73.93
TGMV-YV 95.55 71.88 85.61 72.73 83.43 89.73 71.60
ToRMV-[Ube] 93.15 73.44 83.33 72.73 84.30 87.08 75.10
ToCMoV-[BZ] 93.15 71.09 85.61 60.00 85.18 83.33 73.15
CdTV-[IC] 93.95 67.18 84.09 74.71 86.63 86.74 73.93
DiYMoV 93.15 69.53 78.03 65.88 84.01 87.75 75.30
PYMV-VE 93.95 68.75 82.58 NA 83.14 86.73 70.43
BGMV-[BZ] 95.16 76.56 86.36 52.94 82.27 85.03 72.73
ToMoTV 93.54 66.41 83.33 71.76 84.01 89.46 72.76
PYMV-[GP] 93.95 66.41 100.00 62.35 83.14 87.42 70.04
PYMPV 93.55 68.75 84.85 62.35 84.85 87.42 71.60
CaLCuV 95.16 70.31 81.06 28.57 71.47 85.71 75.10
PepGMV 93.55 71.09 83.33 NA 70.21 85.52 75.49
* = Virus names and accession numbers are described in material and methods.
NA = Not available. CP = Coat protein, TrAP = Transcriptional-activator protein, REn =
Replication-enhancer protein, AC4 = Putative AC4 protein, Rep = Replication-associated
protein, MP = Movement protein, NSP = Nuclear-shuttle protein.

Discussion

The full-length sequences of A and B components of 57-v virus were generated.

The A and the B components contained 2,575 nt and the 2543 nt, respectively

(Figure 4-2). Although, the genome of 57-v virus seems to be smaller than other bipartite-

begomovirus genomes, there are several bipartite begomoviruses with short genomes

such as Tomato mottle virus-[Florida] (Abouzid et al. 1992a), CaLCuV (Abouzid, et al.

1992b), PYMV-VE (Coutts et al. 1991), PYMTV-TT (Umaharan et al. 1998), among

others. The inferred-genome organization of 57-v virus was similar to those of bipartite

begomoviruses from the New World (Figure 4-2). The A component contains four genes

in the complementary sense (Rep, TrAP, REn, AC4) and one gene (CP) in the viral sense.















CaLCuV
100
PepGMV

DiYMoV

TGMV-YV

57-v A

BGMV-[BZ]

91 ToRMV-[UBE]
100
ToCMoV-[BZ]

SiGMCRV

-90 8 ToMoTV
85
CdTV-[IC]
96
PYMPV
100
PYMTV-TT

76 PYMV-[GP]
-- 98
-- FPYMV-VE


100.00



Figure 4-3. Phylogenetic tree based on the complete nucleotide sequence of the A
component of the 57-v sample from Merida state and fourteen bipartite
begomoviruses. The tree was generated using the PAUP program. A single
most parsimonious tree was predicted by a heuristic search with stepwise
addition, random branch-swapping, tree-bisection-reconnection options (500
replication for bootstrapping). Bootstrap indices are shown at each node. Scale
bar references branch length as frequency of changes per site











57-v B


PepGMV


- ToCMoV-[BZ]

ToRMV-[UBE]


PYMTV-TT


100.00


Figure 4-4. Phylogenetic tree based on the complete nucleotide sequence of the B
component of the 57-v sample from Merida state and fourteen bipartite
begomoviruses. The tree was generated using the PAUP program. A single
most parsimonious tree was predicted by a heuristic search with stepwise
addition, random branch-swapping, tree-bisection-reconnection options (500
replication for bootstrapping). Bootstrap indices are shown at each node. Scale
bar references branch length as frequency of changes per site











ToLCrV


ToCMoV-[BZ]


100.00

Figure 4-5. Phylogenetic tree based on the complete coat protein nucleotide sequence of
the 57-v sample from Merida state and fourteen bipartite begomoviruses. The
tree was generated using the PAUP program. A single most parsimonious tree
was predicted by a heuristic search with stepwise addition, random branch-
swapping, tree-bisection-reconnection options (500 replication for
bootstrapping). Bootstrap indices are shown at each node. Scale bar references
branch length as frequency of changes per site












CaLCuV
100
PepGMV

SiGMCRV

ToCMoV-[BZ]

DiYMoV

57-v A

--- ToMoTV
97
ToLCrV

BGMV-[BZ]

ToRMV-[UBE]

TGMV-YV

PYMTV-TT

PYMPV
58

PYMV-[GP]
100
PYMV-VE



100.00


Figure 4-6. Phylogenetic tree constructed based on the complete replication-associated
protein nucleotide sequence of the 57-v sample from Merida state and
fourteen bipartite begomoviruses. The tree was generated using the PAUP
program. A single most parsimonious tree was predicted by a heuristic search
with stepwise addition, random branch-swapping, tree-bisection-reconnection
options (500 replication for bootstrapping). Bootstrap indices are shown at
each node. Scale bar references branch length as frequency of changes per site









The B component contains two genes MP and NSP (Gutierrez, 2000). The

conserved sequences in the IR are similar to those of bipartite begomoviruses. However,

one inverted and one forward repeat was observed in the IR. Usually begomoviruses have

two inverted and two forward repeats. The sequence of the iteron-related domain of

replication protein is similar to those of BGMV-[BZ], PYMTV-TT, Sida golden mosaic

Honduras virus, Rhynchosia golden mosaic virus, Tomato mosaic Havana virus

[Quivican], and Cowpea golden mosaic virus-[Brazil] (Argiiello-Astorga et al. 2001).

The NSI values for the A component between of 57-v virus sequence and

14-known begomoviruses were always less than 89% (Table 4-3), which confirmed that

57-v virus is a distinct begomovirus (Fauquet et al. 2003). The highest values of NSI

were observed for the CP, this corroborates that the CP sequence is more conserved than

the remainder of the genome of begomoviruses (Brown, et al. 2001; Padidam et al.

1995).

Full-length clones for the A and B components were generated from purified PCR

product using overlapping primer methodology (Patel et al. 1993). Biolistic inoculation

will be conducted to determine host range of this new begomovirus from Merida state,

Venezuela. This will corroborate that the full-length clones are in fact infectious and

representative of the virus.

The phylogenetic analysis revealed that the 57-v virus is not closely related to

PYMV-VE or to other PYMV species. The 57-v virus is an undescribed begomovirus

related to DiYMoV based on phylogenetic trees of the full-length sequence of the A and

the B components (Figures 4-3 and 4-4). A similar relationship was observed in the

phylogenetic trees of the sequence of the CP and Rep genes (Figures 4-5 and 4-6).






78


DiYMoV was first reported to be closely related to BGMV isolates. In addition, tomato

was reported as a no host of DiYMoV (Lotrakul et al. 2000). In some way, the DiYMoV

lost the ability to recognize tomato plants as a host in its evolution process.

Since the 57-v virus is an undescribed bipartite begomovirus, it is necessary to

determine its distribution in other states where tomato crops are grown in Venezuela. The

knowledge of 57-v virus sequence and its distribution in Venezuela could be the first step

to initiate a management program of begomoviruses in tomato crops.














CHAPTER 5
DISTRIBUTION OF TWO BEGOMOVIRUSES IN TOMATO-PRODUCTION AREAS
OF VENEZUELA

Introduction

Begomoviruses are circular single strand DNA virus. Most begomoviruses,

particularly those found in the New World, have a bipartite genome. The A component

gives rise to a single virion-sense mRNA that codes for the coat protein (CP) while

complementary transcription results in four mRNAs that code for proteins related to

replication of the virus, replication-associated protein (Rep), replication enhancer

protein (REn, former AC3), transcriptional activator protein (TrAP, former AC2) and

AC4, which function is still unknown (Gutierrez, 2000; Lazarowitz, 1992). There is a

non-coding region between the Rep and the CP, called intergenic region. This region has

conserved sequences shared by both the A and B components of the bipartite

begomoviruses. These conserved sequences are identified as a stem-loop motif, TATA

box, forward, and inverted repeats, which are involved with replication and bi-directional

transcription (Argtiello-Astorga and Ruiz-Medrano, 2001).

The B component has the genes related to the movement of the virus. The A

components of begomoviruses share a higher degree of sequence homology than the B

components. This is due in large part to the presence of the coat protein (CP) and

conserved sequences in the Rep. The amino acid sequence of the CP is more conserved

than the remainder of the genome, this is probably because of the many constraints

placed on the CP for viral structure, vector transmission, host specificity and possibly









other unknown functions (Padidam et al. 1995). In contrast, the B component DNA

sequences exhibit a high degree of sequence divergence, especially the hypervariable that

is located between the beginning of the movement protein and the common region of the

A and B components (Rani et al. 1996).

Begomoviruses cause serious diseases in vegetables and fiber crops and have

emerged as important viral pathogens in tropical and subtropical regions (Brown et al.

2001; Polston and Anderson, 1997). There is a need for accurate and simple

methodologies for rapid and accurate detection of begomoviruses in order to develop

disease management strategies. Biological assays to identify begomoviruses have been

very difficult, since many begomovirus are not mechanically transmissible. Serology has

been of limited use for begomovirus characterization because of the low titer of antigens,

the coat protein amino acid sequence the target for serological probes is highly

conserved, cross-reaction of antibodies with heterologous antigens, and developmental or

environmental regulation of antigen production (Brown et al. 2001; Czosnek and

Laterrot, 1997; Rojas et al. 1993; Wyatt and Brown, 1996). DNA-DNA hybridization

assays, polymerase chain reaction (PCR), molecular cloning and DNA sequencing of

viral genomes have become the tool of choice, allowing one to accurately identify the

virus and to evaluate its relationship to other virus isolates (Padidam et al. 1995; Polston

et al. 1989; Rybicki, 1994). Even though hybridization is less sensitive than PCR, this

technique offers the possibility to evaluate many samples in less time and with less cost

than PCR.

Probes that include most of the intergenic region have been used to detect only

isolates closely related to TYLCV-Is (Czosnek and Laterrot, 1997). Full-length sequences









have been used to detect a wide range of begomoviruses, especially when the probes

include the conserved regions (Czosnek and Laterrot, 1997). The A components of

begomoviruses have been used as a general probe, and the B components have been used

as specific probes because of the high degree of sequence divergence (Rani et al. 1996).

Begomoviruses have been reported in Venezuela since 1963 (Debrot et al 1963).

Four begomoviruses have been described affecting tomato crops in different states in

Venezuela, Potato yellow mosaic virus-Venezuela (PYMV-VE) (Coutts et al. 1991;

Roberts et al. 1986; Roberts et al. 1988), Tomato yellow mosaic virus (ToYMV) (Debrot

et al. 1963; Lastra and Uzcategui, 1975; Morales et al. 2001), Potato yellow mosaic

virus-VE strain tomato, and Tomato Venezuela virus (ToVEV) (Guzman et al. 1997).

Recently, two new begomoviruses have been reported in a survey conducted in 1994 in

the states of Trujillo and Merida (Chapters 3 and 4). The begomoviruses were designated

as 2.9-v and 57-v virus. To determine the distribution of these two new begomoviruses in

tomato-growing areas of Venezuela, the following objectives were formulated:

* Determine the distribution of 57-v virus and 2.9-v virus in tomato samples from 10
states in Venezuela by hybridization with specific probes

* Confirm the results of hybridization using virus-specific probes by PCR followed
by restriction analysis

* Confirm the results of hybridization using virus-specific probes by sequence
analysis of selected plant samples.


Materials and Methods

Survey

Tomato leaves with begomovirus-like symptoms were collected in 10 states in

Venezuela. The sampling as well as the preservation of the samples has been previously

described (Chapter 2).









DNA Extraction, Begomovirus Detection by PCR, and Hybridization

Genomic DNA was extracted from freeze-dried tissue (Rath et al. 1998).

Previously, begomoviruses were detected by PCR using three sets of degenerate primers

and one set of primers specific for Potato yellow mosaic virus (PYMV) (Chapter 2).

Based on DNA sequence comparison and phylogenetic analysis, two new begomoviruses

designated 2.9-v (from Trujillo) and 57-v (from Merida) were reported (Chapter 2).

For the hybridization, genomic DNA was spotted onto nylon membranes. Only

those samples that were positive for at least one out of four PCR reactions, which amplify

partial sequences of the A or B components (Chapter 2), were used for the hybridization

assay. TE-8 (120 [tL) was placed in a chilled 1.5 mL tube and 40 ptL of genomic DNA

were added to the tube. To denature the DNA, 20 p.L of 1 M NaOH (final concentration

0.1 M) were added and mixed well with a pipette, and incubated for 10 min at room

temperature. Then, the DNA was neutralized by adding 20 pL of 3 M sodium acetate

(pH 4.5), inverting the tube several times and incubating for 10 min at room temperature.

Nylon membrane, Hybond N + (0.45 ptm) (Amersham Pharmacia Biotech Inc,

Pisscataway, NJ), was presoaked first in H20 for 5 min and then in TAE for 10 min.

After that, the membrane was placed in a blotting manifold with a layer of water-soaked

filter paper (Whatmann 3MM) beneath the nylon membrane. Then, 20 ptL of the DNA

sample were loaded per well. Genomic DNA extracted from healthy tomato leaves was

used as a negative control, and DNA extracts from samples 57-v and 2.9-v were used as

positive controls. The membranes were placed on dry filter paper and crosslinked twice

for 1200 s at 120 /J/cm2 in a UV Stratalinker 1800 (Stratagene, La Jolla, CA). The

membranes were kept in a plastic bag at 40C until hybridization.









PCR Probe

The hypervariable region (-300 nt) of the B component was used as virus specific

probes, this region was selected based on nucleotide sequence identity (NSI) value

among partial sequences of 12 different DNA samples from Andean states (Chapter 2) as

well as previous reports, in which the intergenic region, partial or full-length sequences

of the B component have been used as specific probe to distinguish virus strains

(Czosnek et al. 1990; Czosnek and Laterrot, 1997; Mansoor et al. 2003; Polston et al.

1993; Rani et al. 1996).

Primer sets were designed to amplify this region of virus 2.9-v and 57-v using a

Wisconsin Package Version 10.3, Accelrys (GCG), San Diego, CA. Primer set

JAP176/JAP177 (5' -AGTAAAATTAGCCCGCCAG-3' and


5' -TGCCACTCCAAGCTCCTATC-3') amplifies the hypervariable region of the 2.9-v


virus. Primer set JAP180/JAP181 (5 -AGGTTGCGCAGCTAAATG-3' and 5' -


CTGCCATAAATTATCCCTTCTC-3 ') amplifies the hypervariable region of the 57-v

virus.

DNA amplification parameters for both sets of primers were as follows: 35 cycles

of denaturation for 1 min at 940C, primer annealing for 1 min at 570C, and primer

extension for 1 min at 750C, with an initial denaturing at 940C for 5 min and a final

extension for 5 min at 720C. PCR reactions were carried out in a PE Applied Biosystems

GeneAmp PCR System 9700 thermocycler (PE Applied Biosystems, Foster City, CA).

All amplifications were performed in volumes of 25 [aL containing 10 mM Tris-HCl (pH

9), 50 mM KC1 and 1% Triton X-100, 1.5 mM MgC12, 150 [aM dNTPs (dCTP was









labeled with, approximately 50 [tCi dCTP, [a 32 P] 3000 Ci/mmol, aqueous solution), 0.5

mM spermidine, 1 [tM of each primer, 1 pL of 1:100 dilution of plasmid DNA from

clones 57-vB2 and 2.9-vB20 [purified-PCR products, using PBL1v2040/PCRc154

primers (Rojas et al. 1993), from samples 2.9-v and 57-v cloned in pGEM-T easy

system], and 1.25 U Taq polymerase. The amplified DNA fragments were filtered

through Sephadex columns, which were made with 0.5 mL plastic tubes with a

26.5 gauge-needle hole in the bottom and 1/5th of the tubes filled with glass wool. The

columns were placed into a 1.5 mL tubes and 0.5 mL of Sephadex in TAE was added and

spun down twice for 30 s, changing the 1.5 mL tubes each time. The probes 2.9-v and 57-

v were read in a Bioscan/QC-4000 XER (Bioscan Inc. Washington, DC), boiled for 2

min, and kept on ice until hybridization.

Hybridization Conditions

The membranes were rolled between two layers of mesh and placed into roller

bottles with 25 mL of prehybridization solution (pH 7.4) for 4 h at 65C. The

prehybridization solution contained 0.25 M NaPO4, 0.001% Ficoll (w/v), 0.001% PVP-40

(w/v), 0.001% BSA (w/v), 1% SDS (w/v), 1 mM EDTA, 100 [tg/mL of denatured-salmon

sperm DNA. The probe was added to 15 mL of the prehybridization solution. The

membranes were hybridized for 24 h at 650C. The membranes were washed twice in each

of the following solutions, 2X SSC for 5 min at room temperature, 0.2X SSC with 1%

SDS for 15 min at 650C, and 0.1X SSC with 1% SDS for 5 min at room temperature. The

membranes were exposed to Kodak BioMax MS film (Sigma-Aldrich St. Louis, MO)

with two intensifying screens at -80C for 72 h and 120 h for probe 57-v and 2.9v,

respectively.









Confirmation of Hybridization Results by PCR and Restriction Analysis

All the samples that gave a positive hybridization signal with either probe, were

tested by restriction analysis of the PCR products using degenerate primers and the virus

specific primers for 2.9-v and 57-v. Specific primer sets: JAP176/JAP177,

JAP122/JAP134 (Chapter 3) and JAP138/JAP139 (Chapter 3) for 2.9-v virus, were used

to amplify a fragment of -300 bp (hypervariable region); a fragment of -1,045 bp that

contains the end of the Rep (-384 bp), the TrAP, the REn and the 3 end of CP (-197


bp); and a fragment of -654 bp that contains the 3' end of the movement protein (MP)

(-22 bp), the intergenic region and the 3' end of the nuclear shuttle protein (NSP) (-428

bp), respectively. Specific primer sets: JAP180/JAP181, JAP136/JAP137 (Chapter 4) and

JAP140/JAP141 (Chapter 4) for 57-v virus, were used to amplify a fragment of -300 bp

(hypervariable region); a fragment of -446 bp that contains the 3 end of TrAP (-105

bp), the REn (-247 bp) and part of the CP (-199 bp); and a fragment of -857 bp that

contains the 3' end of the MP (-113 bp), the intergenic region (-56 bp) and the 3' end of

the NSP (-688 bp), respectively. Primer sequences and PCR conditions were previously

described (Chapter 3 and 4).

PCR products of -700 bp generated from primers PBL1v2040 and PCRc154

(Rojas et al. 1993) of samples that hybridized with 2.9-v probe were digested using four

restriction enzymes, Ndel, EcoRI, Mfel and BglII. Same PCR products of -700 bp of

samples that hybridized with 57-v probe were digested using Ndel, EcoRI, Mfel and

BglII. Digestion conditions were carried out according to the manufacturer's instructions

(New England Biolab, Inc. Beverly, MA)









Southern Blot Analysis, Cloning and Sequence Determination

Purified PCR products from primers PBLlv2040 and PCRc 154 of five samples

(103-v, 117-v, 261-v, 297-v and 307v) that hybridized to the 57-v probe were amplified

using primers PBLlv2040 and PCRc154 (Rojas et al. 1993), a 700 bp PCR product was

obtained and digested with EcoRI. The digestion product was run in agarose gel and

viewed with Ethidium bromide. The gel was blotted on charged nylon membrane

(0.45 micron) (Osmionics INC. Westborough, MA) using turboblotterTM (Schleicher &

Schuell, Keene, New Hampshire). The 57-v probe was labeled with digoxigenin using

PCR-Dig probe synthesis kit (Roche Diagnostics Corporation, Indianapolis, IN) and the

primers JAP180/JAP181. Visualization of the probe-target hybrids was performed by

chemiluminescent assay. The blot was exposed to blue medical X-ray film (Diagnostic

Imaging Inc. Jacksonville, FL) for 30 s at room temperature.

The amplified 700 bp products of samples 103-v, 117-v, 261-v and 307-v were

cloned in pGEM-T Easy vector system I (Promega Corporation, Madison, WI) according

to the manufacturer's instructions. Transformation was performed using XL1-Blue MRF

Supercompetent Escherichia coli cells (Stratagene, La Jolla, CA). The clones were

screened by PCR using primers PBLlv2040 and PCRcl54 and by digestion with EcoRI.

The clones with EcoRI restriction site were sequenced at Ana-Gen Technologies, Inc.

Atlanta, GA. Nucleic acid sequences were analyzed using a Wisconsin Package Version

10.3 [Accelrys (GCG), San Diego, CA]. Basic Local Alignment Search Tool (BLAST)

was used to search for similarities between a query sequence and all the sequences in the

database. GAP analysis was used to obtain the values of NSI.









Results

Confirmation of Samples with Hybridization Signal to the 2.9-v Probe by PCR and
Restriction Analysis

The 2.9-v virus was detected in Trujillo, Lara and Zulia states (Figure 5-1). Strong

hybridization signals were detected in three samples from Trujillo and a very weak signal

was observed in four samples from Lara and Zulia. The restriction analysis of these

samples revealed that they had similar restriction sites as the 2.9-v virus, based on the

fragment sizes from Mfel digestion, with the exception of sample 71-v from Zulia (Table

5-1).

Primer set JAP176/JAP177 was unable to generate an expected size fragment

from two of the seven samples that hybridized to the 2.9-v probe (Table 5-1). Primer set

JAP176/JAP177 amplified an approximately 300 bp sequence from the hypervariable

region, a region that was present in the virus specific probe. An expected size PCR

product was obtained when specific internal primers for 2.9-v virus, JAP122/JAP134,

were used for all of the samples, which hybridized to the 2.9-v probe. In some cases extra

bands were also amplified (i.e. samples 92-v and 107-v). When specific primers

(JAP138/JAP139) for the A component of virus 2.9-v were used, the expected PCR

product was generated in all the samples with the exception of 92-v and 107-v samples

(Table 5-1).

Confirmation of Samples with Hybridization Signal to the 57-v Probe by PCR and
Restriction Analysis

The virus 57-v was detected by hybridization using specific probe in samples

from five out of 10 states covered in this survey. The virus was detected in Merida, Lara,

Zulia, Guarico, and Aragua (Figure 5-1). The restriction analysis of Aragua samples

revealed that eight of 11 samples had more than one virus based on the digestion with









EcoR I. The samples 268-v and 270-v had BsrDI and Ndel restriction sites, respectively

(Table 5-2). No samples had the same restriction sites as the 57-v sample, from which the

57-v probe was made (Table 5-2).

Primers JAP180/JAP181 were able to amplify an expected size fragment from

samples from Aragua and Guarico, which hybridized either strongly or weakly to the

57-v probe. However, they did not amplify any fragments from the samples from Lara

and Zulia. When specific internal primers, JAP136/JAP137, for the A component were

used, PCR products of the expected size were amplified but other bands were also

amplified in 11 out of 24 samples with the exception of 2 samples from Merida, and

seven samples from Aragua (Table 5-2). The samples from Lara, Zulia and Guarico states

did not show the expected PCR product when specific internal primers of the B

component, JAP140/JAP141, were used. The samples from Aragua state presented the

expected sized PCR products with those primers but multiple bands were also observed

(Table 5-2).

Southern Blot Analysis, Sequence comparison

An EcoRI restriction site was present in the PCR product of 700 bp from primers

PBLlv2040/PCRcl54 of samples 103-v, 117-v, 261-v and 307-v samples, this digestion

produced two bands, 500 and 200 bp, and also a band of 700 bp was observed on the gel

(Figure 5-2-A). Therefore, it was evident that these samples had mixed infection. The

57-v probe labeled with digoxigenin hybridized only with the fragment of 700 bp

(Figure 5-2B). The BLAST analysis of the sequences with EcoRI site showed high NSI

values with PYMV sequences from different countries. In addition, the sequences had

high value of NSI within them (Table 5-3) and with partial sequences from Andean states