|UFDC Home||myUFDC Home | Help|
This item has the following downloads:
1 CHARACTERIZATION OF THE LETHAL HO ST-PATHOGEN INTERACTION BETWEEN TOBACCO MILD GREEN MOSAIC VI RUS AND TROPICAL SODA APPLE By JONATHAN ROBERT HORRELL A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE UNIVERSITY OF FLORIDA 2007
2 2007 Jonathan R. Horrell
3 To my Father, who has always inspired the best in me
4 ACKNOWLEDGMENTS I thank all the people in the Pl ant Pathology Department at th e University of Florida for giving me the chance to further my education. I thank Dr. Francis Ze ttler, without whose encouragement I would have never endeavored to undertake graduate studies. I especially thank my advisor, Dr. Raghavan Charudattan, for hi s positive outlook, infectious enthusiasm, and support. Thanks, also, to the other members of my graduate advisory committee: Dr. Jane Polston, Dr. Ernest Hiebert, a nd Dr. Dennis Lewandowski. I a dditionally thank: Mark Elliott and Jim DeValerio for their guidance; Kris B eckham and Rob England for their invaluable practical expertise and guidance; Dr. Carlye Bake r, Dr. Jeff Jones, Dr. Harry J. Klee, Dr. WenYuan Song, and Dr. Jeff Rollins, for technical expertise and advice; Eldon Philman and Herman Brown for greenhouse maintenance, insect control, and general camaraderie; and Gene Crawford for always having a good story to tell. Furthe rmore, I express gratit ude and thanks to Dr. Deborah M. Mathews (U.C., Riverside) and Dr. L ynn Bohs (University of Utah, Salt Lake City) for technical and material assistan ce. Finally, I give my sincerest thanks to the taxpayers of the State of Florida, withou t whose contribution my graduate scho ol education would not have been possible.
5 TABLE OF CONTENTS page ACKNOWLEDGMENTS...............................................................................................................4 LIST OF TABLES................................................................................................................. ..........7 LIST OF FIGURES................................................................................................................ .........8 LIST OF ABBREVIATIONS........................................................................................................11 ABSTRACT....................................................................................................................... ............17 CHAPTER 1 LITERATURE REVIEW.......................................................................................................19 Host Biology................................................................................................................... ........19 Pathogen Biology............................................................................................................... .....24 Host-Pathogen Interaction......................................................................................................38 Cases of virus-induced lethal syst emic hypersensitive response (LSHR).......................42 Cases of gene-for-gene interactions between tobamoviruses and Solanaceae................45 2 IDENTIFICATION OF COLU MNEA ISOLATE PV-0113.................................................55 Introduction................................................................................................................... ..........55 Materials and Methods.......................................................................................................... .59 Results........................................................................................................................ .............64 Discussion..................................................................................................................... ..........66 3 COLUMNEA ISOLATE PV-0113: SUBTYPE, SATELLITE, AND DISEASE PROGRESS....................................................................................................................... .....76 Introduction................................................................................................................... ..........76 Materials and Methods.......................................................................................................... .78 Results........................................................................................................................ .............81 Discussion..................................................................................................................... ..........84 4 DISEASE PHENOTYPE AND COMPARIS ON OF PV-0113 WITH OTHER CULTURES....................................................................................................................... .....93 Introduction................................................................................................................... ..........93 Materials and Methods.......................................................................................................... .95 Results........................................................................................................................ .............98 Discussion..................................................................................................................... ........101
6 5 CROSS-PROTECTION EFFECTS......................................................................................110 Introduction................................................................................................................... ........110 Materials and Methods.........................................................................................................111 Results........................................................................................................................ ...........113 Discussion..................................................................................................................... ........113 6 TMGMV PV-0113: PURIFICATION LYOPHILIZATION, AND RNA EXTRACTION.....................................................................................................................116 Introduction................................................................................................................... ........116 Materials and Methods.........................................................................................................117 Results........................................................................................................................ ...........120 Discussion..................................................................................................................... ........121 7 DILUTION ENDPOINT AND BIOLOGICAL ACTIVITY OF PV-0113..........................123 Introduction................................................................................................................... ........123 Materials and Methods.........................................................................................................127 Results........................................................................................................................ ...........127 Discussion..................................................................................................................... ........128 8 CLONING OF PV-0113 COAT A ND MOVEMENT PROTEIN GENES.........................138 Introduction................................................................................................................... ........138 Materials and Methods.........................................................................................................141 Results........................................................................................................................ ...........147 Discussion..................................................................................................................... ........148 9 TRANSIENT EXPRESSION OF PV-0113 COAT PROTEIN............................................158 Introduction................................................................................................................... ........158 Materials and Methods.........................................................................................................161 Results........................................................................................................................ ...........169 Discussion..................................................................................................................... ........172 10 GENERAL DISCUSSION...................................................................................................189 11 SUMMARY AND CONCLUSIONS...................................................................................196 LIST OF REFERENCES.............................................................................................................198 BIOGRAPHICAL SKETCH.......................................................................................................214
7 LIST OF TABLES Table page 1-1 Strains/isolates of Tobacco m ild green mosaic virus (TMGMV)......................................51 1-2 Size of TMGMV genes and gene products........................................................................52 1-3 Relationship between genotype and disease phenotype in Capsicum sp. peppers............52 2-1 Species-specific primers used in revers e transcriptase polymerase chain reaction...........68 2-2 Experimental host range and symptoms caused by TMGMV PV-0113............................69 2-3 Enzyme linked immunosorbent assay ( ELISA) using cross-absorbed antiserum.............69 3-1 Scale for rating disease pr ogress in TMGMV-infected tr opical soda apple (TSA)...........86 4-1 Symptoms of TMGM V on various hosts.........................................................................105 5-1 Cross-protection study.....................................................................................................115 7-1 Latin square design used in N. tabacum experiments......................................................130 8-1 Useful products produced by us ing transgenic tobamoviruses........................................150 9-1 Tobacco mild green mosaic virus-containing tobamovirus chimeras..............................176 9-2 Bim-Lab-inoculated treatment plants..............................................................................176
8 LIST OF FIGURES Figure page 1-1 Tropical soda apple (TSA ) colonizing pastureland...........................................................53 1-2 Tropical soda apple infestations.........................................................................................53 1-3 Genome organization map of Tobacco mild green mosaic virus (TMGMV)...................54 2-1 Symptoms of TMGMV on Nicotiana tabacum cv. Samsun ( nn )...................................70 2-2 Symptoms of TMGMV on Eryngium planum ...................................................................70 2-3 Isolate PV-0113 reacts with rabb it anti-TMGMV serum in SDS media...........................71 2-4 Isolate PV-0113 reacts with rabbit anti-TMGMV serum in non-SDS media....................72 2-5 Isolate PV-0113 does not react with rabbit anti-Tomato mosaic virus (ToMV) serum....73 2-6 Isolate PV-0113 reacts weakly with rabb it anti-Tobacco mosaic virus (TMV) serum.....74 2-7 Reverse transcriptase polymerase chai n reaction (RT-PCR) products of PV-0113..........75 3-1 Disease progress in TSA plants inoculated with TMGMV PV-0113................................88 3-2 Agarose gel analysis of TMGMV RT-PCR products........................................................89 3-3 Comparison of dsRNA extracts resolved in a polyacrylamide gel....................................90 3-4 Comparison of dsRNA extracts resolved in an agarose gel...............................................91 3-5 Imuunodiffusion analysis with anti-Satelli te tobacco mosaic virus (STMV) serum.........92 4-1 Genetic map of TMV/TMGMV chimera 30 B................................................................106 4-2 Symptoms produced in TSA s eedling by 30-B TMV/TMGMV chimera.......................107 4-3 Isolate PV-586-inoculated TSA rec overing with mosaic after dieback...........................108 4-4 Photographs of symptoms by various tobamoviruses on TSA........................................109 6-1 Comparison of the activity of tw o preparations of TMGMV PV-0113...........................122 7-1 Local lesion development over time, Preparation D....................................................131 7-2 Local lesion development over time, Preparation Q....................................................131 7-3 Local lesions produced on Samsun ( NN ) by TMGMV PV-0113 preparation D.........132
9 7-4 Local lesions produced on Samsun ( NN ) by TMGMV PV-0113 preparation Q.........132 7-5 Average of all local le sions produced on Samsun ( NN )...............................................133 7-6 Disease progress on TSA (1 plant per dilution) inoculated with preparation D..............134 7-7 Disease progress on TSA (1 plant per dilution) inoculated with preparation Q..............135 7-8 Average local lesions per half-leaf pr oduced by (refrigerated) preparation D.............136 7-9 Average local lesions per half-leaf produced by (lyophilized) preparation D.............137 8-1 Map of the proprietary cloning vector, pDONR221........................................................151 8-2 Constructs used in TMGMV PV-0113 research..............................................................152 8-3 Nucleotide alignment of TMGMV op en reading frame (ORF) 3 sequences...................153 8-3 Nucleotide alignment of TMGM V ORF 3 sequences (continued)..................................154 8-4 Amino acid alignment of TM GMV movement protein (MP).........................................155 8-5 Nucleotide alignment of TMGMV ORF 4 sequences.....................................................156 8-6 Amino acid alignment of TMGMV coat protein (CP).....................................................157 8-7 Nucleotide alignment showing the Gibbs TMGMV-specific motif................................157 9-1 Map of TMV/TMGMV chimera, 30B.............................................................................177 9-2 Nucleotide alignment between 30B CP domain and PV-0113 ORF4.............................178 9-3 Amino acid alignment between 30B CP and PV-0113 CP..............................................179 9-4 Symptoms on TSA leaf following in oculation with the 30B construct...........................180 9-5 Local lesions on Nicotiana sylvestris following inoculation with TMGMV PV-0113...181 9-6 Symptoms on N. sylvestris following inoculation with TMV.........................................182 9-7 Local lesions on N. sylvestris following inoculation with 30B.......................................183 9-8 Western blot showing CP accumulation in TSA.............................................................184 9-9 Solanum viarum inoculated with 30B GFPc3..................................................................184 9-10 Nicotiana benthamiana inoculated with sap containing 30B GFPc3..............................185 9-11 Epifluorescent microscopy of 30B GFPc3 in N. benthamiana ........................................186
10 9-12 Epifluorescent microscopy of 30B-GFPc3 in TSA.........................................................187 9-13 Necrotic areas on N. sylvestris in response to infiltration with pE1-D2..........................188
11 LIST OF ABBREVIATIONS aa Amino acid(s) AAB Association of Applied Biologists AjMV Araujia mosaic virus ATCC American type culture collection Avr Avirulence gene BLAST Basic local alignment search tool BMV Brome mosaic virus BSA Bovine serum albumen cDNA Complementary DNA CFMMV Cucumber fruit mottle mosaic virus CGMMV Cucumber green mottle mosaic virus CMI Commonwealth Mycological Institute CMMoV Cactus mild mottle virus CMV Cucumber mosaic virus CP Coat protein CREC Citrus Research and Education Center C-terminal Carboxy-terminal cr-TMV Crucifer-infecting tobamovirus CTMV-W Crucifer-infecting t obamovirus, wasabi strain DPI Days post-inoculation dH2O Distilled water DEPC Diethylpyrocarbonate
12 DNA Deoxyribonucleic acid dNTP Deoxynucleoside triphosphate DMSO Dimethyl sulfoxide DSMZ Deutsche Sammlung von Mikroorganismen und Zellkulturen (German Collection of Microorganisms and Cell Cultures) dsRNA Double-stranded RNA DTT Dithiothreitol eGFP Enhanced green fluorescent protein ELISA Enzyme linked immunosorbent assay ER Extreme resistance; endoplasmic reticulum GFP Green fluorescent protein GFPc3 Cycle 3 green fluorescent protein HLFPV Hibiscus latent Fort Pierce virus HLSV Hibiscus latent Singapore virus HR Hypersensitive-response IFAS Institue of Food and Agricultural Sciences kDa kiloDalton (equivalent to 1,000 atomic mass units) KGMMV Kyuri green mottle mosaic virus LB Lysogeny broth (a/k/a Lu ria-Bertani broth); left border LBA Lysogeny broth agar (a /k/a Luria-Bertani agar) LDS Laemmli dissociation solution LL Local lesions
13 LRR Leucine-rich repeat LSHR Lethal systemic hypersensitive response MarMV Maracuja mosaic virus MP Movement protein mRNA Messenger RNA NBS Nucleotide-binding site NCBI National Center for Biotechnology Information NCR Non-coding region nt Nucleotides N-terminal Amino terminal NTLV Nigerian tobacco latent virus NTR Non-translated region ObPV Obuda pepper virus ORF Open reading frame ORMV Oilseed rape mosaic virus ORSV Odontoglossum ringspot virus PaMMV Paprika mild mottle virus PBST Phosphate-buffered saline + Tween 20 PCD Programmed cell death PCR Polymerase chain reaction PLRV Potato leafroll virus PMMoV Pepper mild mottle virus PNP p-nitrophenylphosphate
14 PR Pathogenesis-related PVP-40 Polyvinyl pyrrolidone PVX Potato virus X virus PVY Potato virus Y virus R-gene Resistance gene RB Right border RMV Ribgrass mosaic virus RNA Ribonucleic acid RT Reverse transcription; reverse transcriptase RT-PCR Reverse transcriptase-polymerase chain reaction S Svedberg units SDS Sodium dodecyl sulfate SEL Size exclusion limit (plasmodesmata) SFBV Streptocarpus flower breaking virus SHMV Sunn-hemp mosaic virus SHR Systemic hypersensitive response SLL Single local lesion SMV Soybean mosaic virus SOC Super optimal catabolite-repression SPAR Single primer amplification reaction ssRNA Single-stranded RNA STMV Satellite tobacco mosaic virus Taq Thermophilus aquaticus
15 TBST Tris-buffered saline + Tween 20 TEV Tobacco etch virus TIR Toll-interleukin-1 receptor TMGMV Tobacco mild green mosaic virus TMGMV-L Large-type TMGMV TMGMV-NZ Nagel-Zettler is olate of TMGMV; PV-0113 TMGMV-S Small-type TMGMV TMV Tobacco mosaic virus TMV-Cg Crucifer-infecting t obamovirus, garlic strain TMV-Ob Pepper-infecting tobamovirus, Obuda strain TMV-p Tobacco mosaic virus, petunia strain ToMV Tomato mosaic virus ToMoV Tomato mottle virus TRV Tobacco rattle virus TSA Tropical soda apple TSAMV Tropical soda apple mosaic virus TSWV Tomato spotted wilt virus TVCV Turnip vein clearing virus USDA-GRIN U.S. Department of Agriculture Germplasm Resources Information Network UTR Untranslated region UV Ultraviolet WPI Weeks post-inoculation WT Wild-type
16 wtGFP Wild-type green fluorescent protein YEP Yeast-extract and peptone ZGMMV Zucchini green mottle mosaic virus
17 Abstract of Thesis Presen ted to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Master of Science CHARACTERIZATION OF THE LETHAL HO ST-PATHOGEN INTERACTION BETWEEN TOBACCO MILD GREEN MOSAIC VIRU S AND TROPICAL SODA APPLE By Jonathan Robert Horrell December 2007 Chair: Raghavan Charudattan Major: Plant Pathology Tropical soda apple ( Solanum viarum Dunal) develops symptoms of lethal systemic hypersensitive response (LSHR) in response to mech anical inoculation with Tobacco mild green mosaic virus (TMGMV; isolate DSMZ PV-0113) Sub-isolates of PV-0113 exhibited no variation in disease phenot ype induced on inoculated tr opical soda apple (TSA) and Nicotiana tabacum cv. Samsun ( nn ). Physical properties, host-ra nge analysis, immunodiffusion, enzyme linked immunosorbent assa y (ELISA), reverse-transcriptase polymerase chain reaction (RTPCR), and sequence analysis of the viral coat pr otein and movement protein genes indicate that PV-0113 is a small type isolate (TMGMV-S). Immunodiffusion using antiserum against Satellite tobacco mosaic virus (STMV) coat pr otein and analysis of double-stranded RNA from infected plants indicate that STMV is not present in this isolate. Sequences of the coat protein and movement protein genes of PV-0113 were near ly identical to 2 other sequences reported for TMGMV. A nucleotide sequence motif observed in PV-0113 was identical (100% homology) to a motif reported as unique to, and present in, all TMGMV sequences so far examined. The pathogen and the disease phenotype it causes in TSA were studied to learn more about lethal plant-virus interactions and to explore the possibility of exploiting this interaction to manage TSA. Symptoms of disease in TSA plan ts infected with PV0113 appear approximately
18 5 days post-inoculation as epinas ty and local lesions on the inocul ated leaves, followed by loss of turgor and abscission of both inoc ulated and uninoculated leaves at 6 days, leading to complete defoliation of the plant within 30 days. The st em becomes necrotic over a period 3 to 4 weeks post-inoculation as the entire pl ant dies. Lethal systemic hypersensitive response was not observed in TSA following inoculation with othe r common tobamoviruses. Inoculation of TSA with a culture of large type TMGMV, containi ng STMV produced necrotic symptoms that were initially similar to those produced by inocul ation with PV-0113, but the plants survived, recovering with mosaic symptoms. TSA plan ts inoculated with TMV are cross-protected against challenge with TMGMV. TSA plants inoculated with the TMV/TMGMV chimera 30B, which expresses the large type TMGMV coat protein, developed mosaic but not necrosis, suggesting that the coat protein of PV-0113 alone is not an elicit or of hypersensitive response in TSA. These results are consistent with the hypot hesis that the lethal pl ant-pathogen interaction between TSA and PV-0113 is highly speci fic to this virus-host combination.
19 CHAPTER 1 LITERATURE REVIEW Host Biology Tropical soda apple ( Solanum viarum Dunal), a prickly member of the nightshade family (Solanaceae), has spread from its natural range in South America (S.E. Brazil, N.E. Argentina, Paraguay and Uruguay), to become a problematic w eed in Central and North America, as well as Africa, Asia, and the Caribbean (Mullahey et al., 1997; Cuda et al., 2002 ). An herbaceous perennial, S. viarum is 1-2 m high, with alternate, simple lobed, pubescent leaves covered with sharp prickles perpendicular to the adaxial and ab axial surfaces. The stem s also are covered in prickles. White, tomato-like flowers, borne te rminally in clusters, develop into green to greenish-white fruit, camouflaged with waterm elon-like patterning. These mature to a solid mustard-yellow color, measur e about 2-3 cm in diameter, and contain numerous seeds surrounded by a spongy mesocarp. Containing the ster oidal glycoalkaloid so lasodine, the fruit are mildly poisonous to humans, although it woul d require about 200 fruit to be lethal. Solasodine has been used in the production of steroid-based medicines; tropical soda apple was once cultivated for this purpose and a body of rela ted literature exists. Production has declined in recent years as better sources of solasodine have been discovered by the pharmaceutical industry (Cuda et al., 20 02; Levin et al., 2005). Tropical soda apple, known in Brazil as ju bravo (angry appl e) (Lorenzi, 2000), and hereinafter referred to as S. viarum or TSA, is a member of the family Solanaceae; section Acanthophora; subgenus Leptostemonum (Levin et al., 2005). Synonyms for this plant appearing in lit erature include Solanum khasianum var. chatterjeeanum Sengupta, S. chloranthum DC, and S. viridiflorum Schlechtendal (USDA ARS, 2003). Solanaceae the nightshade family, is a diverse, successful family belonging to the Solanale s, an order it shares
20 with the morningglory family, Convolvulaceae Today Solanaceae exist on all continents except Antarctica, and in almost every habitat: tropical and temperate, from rainforest to desert (Anonymous, 2002). The family contains many species of great practical a nd aesthetic benefit to humankind, including the tomato, potat o, pepper, and petunia. It al so contains many species that are weeds, including the nightsh ades, horsenettles, and TSA. Some, such as tobacco ( Nicotiana sp.) and angels trumpet ( Datura sp.) might be placed in both categories. Samples of TSA collected in the United States and subjected to single primer amplification reactions (SPAR) and chloroplast DNA sequenci ng showed no variation, suggesting that the current infestation is the result of a single introduction, or multiple introductions from a common source (Kreiser et al., 2004). Recent phylogene tic analyses using sequence data from two nuclear and two chloroplast regions place S. viarum in a clade along with S. aculeatissimum S. incarceratum and S. myriacanthum (Levin et al., 2005, 2006). Th e close taxonomic relationship between S. viarum and S. aculeatissimum is also suggested by the original confusion of both species as S. khasianum with TSA later recognized as S. khasianum var. chatterjeeanum Sengupta. Morphologically, TSA may be distinguished from these relatives by the presence of hairs on the flower ovary (Dr. Lynn Bohs, professo r, Department of Biology, University of Utah, Salt Lake City, personal communication). Tropical soda apple has become established as a noxious weed of pastureland in the United States, especially in the state of Florida. Fi rst collected from Glades Co., FL in 1988, (Coile, 1993), TSA rapidly expanded its range from 10,000 h ectares in 1990 to infest over half a million in 1995 (Mullahey et al., 1998). While infesta tion by TSA remains great est in the state of Florida, it also occurs in Alabama, Arkansas, Georgia, Illinois, Louisi ana, Mississippi, North Carolina, Pennsylvania, Puerto Rico, South Caro lina, Tennessee, and Texas. Plants reproduce
21 either by seed or asexually from the roots. In the Northern Hemisphere, TSA blooms from May to September (Ferrell, 2006). Plants growi ng in Florida may produce 40,000 to 50,000 seeds per individual, about 20% of which remain dormant usually for one month (Mullahey et al., 1997; Medal et al., 2004). Favorable c onditions allow plants to reach maturity in 130 days (Ferrell, 2006). Long prickles on the plant discourage grazi ng, but the ripe fruits are consumed by cattle, pigs, and wildlife. Animals indiscriminate enough to eat the leaves may find themselves poisoned (Porter et al., 2003). The seeds survive digestion and are disseminated in the animals' droppings (McGovern et al., 1994), causing dense stands of TSA (Figur e 1-1, Figure 1-2) to form where cattle congregate ( under shade trees, palm hammocks, feeding stations, etc.) that eventually deny access to the cattl e. Plants and seed can also be disbursed by movement of contaminated hay and grass seed, generating plants in more urban areas such as lawns, roadsides, etc. (Mullahey et al., 1998). Like many invasives, TSA takes advantage of open niches in disturbed environments, and has been observed spreading into areas disturbed by phosphate mining (Albin, 1994). Tropical soda apple is causing economic da mage, mostly through th e loss of grazing land and reduced stocking rates, as well as the expe nse of control and certif ication programs. By 1993, TSA infestation had cost the cattle industr y in Florida 11 million dollars, reducing Florida beef production by about 1% annually (Mullahey et al., 1998). TSA is on both the Federal and Florida noxious weed lists, and is listed as a Category I invasive species by the Florida Exotic Pest Plant Council, defined as "invasive exotic s that are altering native plant communities by displacing native species, changing community struct ures or ecological func tions, or hybridizing with natives" (FLEPPC, 2005). So me states are discussing ship ment regulations that would mandate cattle be quarantined until all viable TSA seed has passed through their digestive
22 system; to quarantine these cattle on a feedlot fo r two weeks would be co stly and result in the cattle suffering increased stress-related illness and weight loss, (Ed Jennings, extension agent, Sumter Co., FL, personal communication). Furt hermore, TSA exerts ecological pressure by displacing native species where it occurs in the hammocks and natural areas of Florida. Approximately 10% of the land infested with TSA is wooded with oak/palm hammocks and cypress stands (Cuda et. al., 2002). Field studies conducted in Fl orida and Brazil have found at least 45 plant pathogens attacking TSA (Charudattan and DeValerio, 1996). Of these, several viruses of agricultural importance, including Cucumber mosaic virus (CMV ), Potato leafroll virus (PLRV), Potato virus Y (PVY), Tobacco etch virus (TEV), Tomato mo ttle virus (ToMoV), and Tomato mosaic virus (ToMV) have been recovered (McGovern et al ., 1994). Additionally, TSA has been infected artificially with Potato virus X (PVX) and To mato spotted wilt virus (TSWV) (Chagas et al., 1978). TSA supports several crop-damaging insect s of economic importance, including tobacco hornworm, tomato hornworm, Colorado potato beetle, tobacco budworm, tomato pinworm, green peach aphid, silverleaf whitefly, soybean looper, and the southern green stinkbug. As a host of these pests and pathogens, TSA may be causing economic damage to Florida vegetable growers (Cuda et al., 2002). Conventional methods of control for TSA requi re repeated mowing and applications of a chemical herbicide. Acifluorfen, clopyralid, dicamba, fluroxypyr, picloram, and triclopyr each provided >90% control of TSA s eedlings with minimal damage (> 10%) to the non-target plant, bahiagrass, observed 145 days afte r treatment (Akanda et al., 1997). Infested pastures should be mowed at least every 60 days to prevent flower ing/fruit set. Remedy (triclopyr) has been the most commonly used treatment for TSA, genera lly applied at a rate of 1.1 kg/ha, or Roundup
23 (glyphosate) applied as a 3% solution. None of the conventional herb icide treatments have proven completely satisfactory for control of TSA. Glyphosate causes damage to forage grasses, and triclopyr has little residua l activity for controlling TSA seed lings emerging post-application. The cost of applying dicamba at rates necessary fo r control of TSA is prohibitive (Ferrell et al., 2006). Control is complicated by the need to sp ot-treat under trees and on hammocks, and must be repeated as new seeds germinate. Complete control by these methods will take 1-3 years, and cost approximately $185 per hect are (Mislevy et al., 1996, 1997; Mu llahey et al., 1998; Sturgis and Colvin, 1996). Aminopyralid, (4-amino-3,6-dichlo ropyridine-2-carboxyl ic acid), a chemical developed by Dow Agrosciences, has been shown to provide > 96% control of TSA, including emergent TSA, for up to 335 days after applicati on, when applied at the rate of 80 g per hectare. Based on this efficacy and the long residual effect in the soil, aminopyralid has been regi stered for TSA control (Ferrell et al., 2006; Anonymous, 2007). Recent experimentation has shown that aminopyralid, (marketed as Milestone and, in co mbination with 2,4D, as ForeFront R&P), although welltranslocated in most plants, is not well translocated within TSA and requires complete and thorough coverage at a rate 0.11% to be effective (Ferrell an d Sellers, 2007). In contrast, inoculation of a single l eaf with a few micrograms of SolviNi x, a virus-based biological control agent described in this thesis, is often highly effective. A tortoise beetle, Gratiana boliviana (Chrysomelidae), found associated with TSA in South America, showed promise in controlling the weed by feeding on th e leaves, and, after host-specificity studies were certified by USDA-A PHIS, was released in Florida for control of TSA in May, 2003 (Cuda et al., 2002; Medal, et al., 2004). However, the field performance of
24 this beetle has been less than satisfactory (R Charudattan, professor, Department of Plant Pathology, University of Florid a, personal communication). Pathogen Biology Few virus-based approaches to biological control of weeds ha ve been proposed before. The non-tobamovirus, Araujia mosaic virus (AjM V), discovered producing disease in the South American Araujia angustifolia was proposed for control of milk weed vine (= strangler vine), Morrenia odorata in Florida (Charudattan et al., 1976, 1978) and more recently, for moth plant, Araujia sericifera in New Zealand (Landcare Research, 2003). The miteand grafttransmissible rose rosette disease has been proposed as a biological control for multiflora rose ( Rosa multiflora ), upon which it causes leaf reddening, shor tened petioles, and death (Epstein et al., 1997.) Although a causal agent has yet to be identified, vi rus-like partic les and doublemembrane bound inclusion bodies have been observe d in tissue symptomatic with rose rosette disease. (Silvestro and Chapman, 2004; Rohozinsk i et al., 2001). A tobamovirus, TMV Alke strain, isolated from Physalis alkekengi caused disease when applied to Solanum carolinense a North American species which had been infesting tea plantations in the former Soviet Republic of Georgia. Systemic symptoms produced by the virus were severe, and included mosaic, epinasty, leaf abscission, and enhanced drought stress (reported to the West by Izhevsky, 1979). Tobacco mild green mosaic virus (TMGMV) was proposed as a bioherbicide for Echium plantagineum in Australia, because systemic infection of E. plantagineum by TMGMV was found to cause systemic yellowing and mosaic, incr eased leaf senescence, and reduced leaf and seed production (Randles, 1986). Based on this research, a graduate student, M. Pettersen, (Plant Pathology Department, University of Florida) conducted an evaluation of Tobacco mosaic virus (TMV), TMGMV, and ToMV as biocontrol agents against TSA. When mechanically inoculated, all three viruses were
25 able to systemically infect TSA (unpublished data). TMGMV induced a necrotic response followed by wilting and leading to the eventual deat h of the plants (Pettersen et al., 2001). TMV and ToMV produced mosaic and mosaic plus mottl ing symptoms, but did not cause plant death. Subsequently, I tested Pepper mild mottle vi rus (PMMoV), Sunn-hemp mosaic virus (SHMV), and Tropical soda apple mosaic virus (TSAMV), but none produced systemic, lethal symptoms on TSA. Many collections of TMGMV have been made by scientists around the world (Table 1-1). In a study sampling gesneriad plants for a latent virus infection (Zettler and Nagel, 1983), an isolate identified as TMGMV was recovered from tissues of the Columnea hybrid Oneidan ( C. crassifolia x C. allenii ) originating from Ohio (Dr. F.W. Zettl er, University of Florida, personal communication). This isolate was propagated in tobacco and deposited as accession PV-0113 at the German Collection of Microorganisms and Ce ll Cultures (DSMV). It was this culture, reliably lethal to TSA, and casually referred to as the Nagel-Zettler or NZ strain of TMGMV, that was used in the research described in this thesis. Tobacco mild green mosaic virus belongs to the genus, Tobamovirus This group includes TMV, the first discovered, and probably most studied, virus in history. It was the first virus to be purified and crystallized (Stanle y, 1935), as well as the first vi rus to be observed by electron microscopy (reviewed by Hull, 2002). TMV was the first RNA virus to have its genome completely sequenced (Goelet et al., 1982). Tobamoviruses are mechanically transmitted and are not known to be vector-transmitted. Some tobamoviruses may be seed-transmitted, due to autoinoculation when the seedling germinates through the contaminated seed coat (R ast, 1979). Tobamovirus-infected plant debris may allow the virus to survive from planting to planting, but probably do es not contribute to
26 spread (Boubourakas et al., 2004). Nevertheless, research indi cates that tobamoviruses are successful and highly ubiquitous. Reports exist, for example, of recovery of TMV and ToMV from forest soils in New York State (Fillhart et al., 1998). Control of these viruses typically relies on proper sanitation and th e use of resistant varieties (Agr ios, 1997; Letschert et al., 2002). In addition to PMMoV, TMV, TMGMV, and ToMV, the genus Tobamovirus also includes the economically important species Odontoglossu m ringspot virus (ORSV) and Cucumber green mottle mosaic virus (CGMMV). ORSV, PMMoV, TMGMV, TMV, and ToMV ar e known to occur or have occurred in Florida (Alfieri et al., 1994; Ba ker and Zettler, 1988; Kucharek et al., 2003; Lamb et al., 2001). Other tobamoviruses detected in Florida include Hibiscus latent Fort Pierce virus (HLFPV) (Kamenova and Adkins, 2004), Maracuja mosaic virus (MarMV) (Alfieri et al., 1994; Song et al., 2006), TSAMV (Adkins et al., 2007.), and an unspecified tobamoviru s producing ringspot in Opuntia sp. cactus (El-Gholl et al., 1997). In addition to the subdivision of TMV, new sp ecies are continually discovered: Nigerian tobacco latent virus (NTLV), r ecovered from a mixed infecti on in Nigerian tobacco in 2003 appears to be somewhat distantly related to mo re common tobamoviruses. Of these, it most closely resembles TMGMV (Lapido et al., 2003 ). Recently, two species discovered in Malvaceous hosts: Hibiscus latent Singapore vi rus (HLSV) (Allen et al ., 2005) and HLFPV have been described (Kamenova and Adkins, 2004). C actus mild mottle virus (CMMoV) (Min et al., 2006) and Streptocarpus flower breaking virus (SFBV) (Heinze et al., 2006) are two more species recently described from infected ornamentals. Tobamoviruses are positive-sense, single strande d RNA viruses. Virions are hollow tubes, roughly 18 nm in diameter x 300 nm long, compos ed of approximately 95% protein, and 5%
27 RNA. Each virion consists of multiple (~2,130) slipper-shaped coat protein (CP) subunits, wound into a right-handed he lix around a single, 6.5 Kb-long RNA strand, somewhat like kernels of corn on a cob. Each CP subunit is composed of about 158 amino acids (TMV); about 50% are arranged into -helixes and 10% into -structures (Hull, 2002). Every three-nucleotide segment of the RNA strand fits into a groove in each protein s ubunit (Lewandowski and Dawson, 1999), lending it great resi stance to chemical and physical degradation (compared to several other plant viruses). Vi rions of TMV, for example, rema ined viable after at least 50 years in storage at room temp erature, and are unaffected by RNA-ase, although the naked RNA is unstable and will degrade (Hull, 2002). Associ ation between the CP and the RNA is so strong that they will self-assemble into active viral part icles in vitro, with aggr egations of CP forming first, that then encapsidate the RNA strand at a specific stem loop, the origin of assembly. In most tobamoviruses, the origin of assembly is located within the third open reading frame (ORF); when the CP assembles around subgenomic mRNA, the result is the formation of shorter virions. The virions will, however conveniently disassemble at the extremes of pH (Hull, 2002). The location of the origin of assembly w ithin the genome may be used to classify tobamoviruses into three subgroups (Fukuda et al., 1981). TMGMV is classified in Fukuda subgroup 1, along with TMV, ToMV, PMMoV, and ORSV. Subgroup 1 is composed of those species in which the origin of assembly is lo cated on ORF 3, and are, with the exception of ORSV, adapted to the Solanaceae In subgroup 2 tobamoviruses, such as CGMMV and SHMV, in which the origin of assembly lies with in ORF 4, most commonly infect species of Cucurbitaceae or Fabaceae Subgroup 3 is made up of tobam oviruses that primarily infect the Brassicaceae such as Turnip vein clearing viru s (TVCV) (Letschert et al., 2002).
28 The replication strategy and genome organi zation of TMGMV (Figure 1-3) and other tobamoviruses are comparable to those of the we ll-studied type species, TMV. The genome of TMV begins with a 5 7-methylguanosine cap (an inverted, methylated guanosine subunit). For mRNA to survive enzymatic degradation in a eukaryotic cell, it must carry this cap. Downstream from the cap, a 70 nucleotid e (nt) leader, known as the omega ( ) sequence, competes with host mRNA for ribosomes to great ly enhance translation in both prokaryotic and eukaryotic systems. Since its discovery, this pr operty has been utilized extensively, making it a useful tool in the hands of molecular biologists. The sequence, positioned before the first ORF, is required for tobamovirus replication (Gallie et al., 1987). ORF 1 codes for a 126 kiloDalton (kDa) protei n, with an amber stop codon which allows read-through 5-10% of the time, to create a longer 183 kDa product (Lewandowski and Dawson, 1999; Hori and Watanabe, 2003). A domain closest to the amino-terminal (N-terminal) of these proteins is believed to serve as a methyltransferase, used in forming the 7-methylguanosine cap. A domain near the carboxy-terminal (C-terminal) of the 126 kDa protein is believed to serve as the helicase, unwinding the dsRNA that forms du ring replication. Further downstream, within the 183 kDa protein is a domain believed to serv e as the polymerase for replication. The next ORF encodes an approximately 28 kDa movement protein (MP). Downstream from the MP, the next ORF encodes a 17.5 kDa CP. Downstream from these translated regions, the genome contains more replication machinery, and the R NA coils to form structures called pseudoknots. Finally, the 3 -end of the RNA sequence ends in a t-RNA like structure, which itself contains 2 pseudoknots (Lewandowski et al ., 1999; Yoon et al., 2006). Unlike the 126 kDa and 183 kDa replication-asso ciated proteins, the CP and MP are not translated directly, but from subgenomic RNAs. Subgenomic I1, which begins with a 5' start at
29 nt 3405, has been isolated in plantae In vitro I1 is capable of transl ating a 54 kDa protein, identical in sequence to the C-terminal end of the 183 kDa transcript (Hull, 2002). Subgenomic I2, serving as the template for translation of MP, spans both ORF 3 and ORF 4. The smallest subgenome produces the CP (Hull, 2002). It is possible that the inte raction of subgenomic promoters with viral replicase is responsible for the timing of indi vidual events in the infection process (Dawson, 1992). Translation of ORF 3 produces the ~28 kDa pr oduct in relatively small amounts, compared with the later accumulation of CP. The N-terminus of this MP binds single stranded nucleic acid, while the C-terminus binds to plasmodesmata (L ewandowski et al., 1999). Tobamovirus MP has been shown to associate w ith viral nucleic acid, and in plantae this protein is indeed localized in the plasmodesmata, where it will increase the pl asmodesmatal size exclusion limit (SEL) about 10-fold (Heinlein, 2002). Tobamovirus MP is required for efficient cell-to-cell movement through the plant. Differences in host range between closely re lated tobamoviruses may well be determined by mutations in this region. In the host Eryngium planum for example, TMGMV moves systemically, while TMV does not move out of the inoculated leaf. This feature enables E. planum to be used as a biological sieve to separate TMGMV from TMV in mixed infections (Johnson, 1947). The 17.5 kDa CP encoded by ORF 4 is translated generating subunits that encapsidate the viral mRNA, allowing long-distance movement through the vasculature of the plant. While CP itself is not necessary to establis h or maintain infection, mutation or deletion of ORF 4 results in genotypes less effective at movement th an the wild-type (Dawson et al., 1988). Other ORFs observable within the tobamovirus genome code mo stly for small proteins of undetermined activity. Morozov described a r eading frame (called ORF-X or ORF 6)
30 beginning at 5669 (5617 in TMGMV) found in some, but not all tobamoviruses, which encodes for a small protein of 33 amino acids (45 amino acids in TMGMV) in length that can be translated in vitro During in vitro translation, a larger protein of about 4.8 kDa is also produced (Morozov et al., 1993). The translation product of ORF 6 has been shown to have effects in plantae although its adaptive purpos e, if any, remains unknow n (Canto et al., 2004). The approximately 200-nt long 3 non-translated region (3 NTR), sometimes called the untranslated region (3 UTR), or the non-coding region (3 NCR), consists of three stemloops in the RNA known as pseudoknots and a tRNA-like 3 end terminal. Together, these structures help to stabilize the RNA and enha nce translation. Proximity to the pseudoknotcontaining region of the 3 NTR affects expression of ge nomic products, with genes (or transgenes) closest to the 3 NTR showing higher levels of e xpression (Shivprasad et al., 1999). The pseudoknot closest to the 3 end has been shown to be the starting point for synthesis of ( )RNA in TMV, and replication and accumulation of virus was reduced in certain pseudoknot deletion mutants of PMMoV. Both the amino acid (aa) sequence of the replicase protein and the 3-dimensional structure of the pseudoknots seem critical to translati on and synthesis of ( )RNA (Yoon et al., 2006). Wild populations of Nicotiana glauca in southern California ha rbor a large-type variant of TMGMV (historically classifi ed as TMV-U5), distinguish able from small-type TMGMV (TMV-U2, TMGMV-S) by a larger 3 NTR. Large-type TMGMV (TMGMV-L) may be identified by RT-PCR using primers specific to the 3 end of the virus, or by careful observation of symptom development on the indicator plants: N. benthamiana N. clevelandii and N. rustica TMGMV-L has a 3 NTR extended by a duplication resulting in six pseudoknots instead of three;
31 this difference appears to have consequences on disease phenotype and movement in some hosts (Bodaghi et al., 2000). While the ranges of these two TMGMV subtypes overlap in southern California, and both infect N. glauca mixed infections in this host remain rare (1.6% incidence; Bodaghi et al., 2004). TMGMV-L seems to have a competitive advantage over TMGMV-S, at least in the hosts N. glauca and N. tabacum ; whereas pre-inoculation by eith er subtype usually provides good cross-protection against the other, once establ ished (5-14 days), TMGMV-L usually displaces TMGMV-S from a host co-inoculated with it (B odaghi et al., 2004). Why duplication of the 3pseudoknot segment in the 3' NTR should confer this advantage is not currently understood. Tobacco mild green mosaic virus is unique in that it is the only tobamovirus found naturally associated with Satellit e tobacco mosaic virus (STMV). Satellite plant viruses are very small and completely dependent upon their associ ated helper virus fo r replication. STMV appears to have a parasitic relationship with TMGMV (the helper), reducing accumulation of the helper virus within the host. Discovered in association with TMGMV in wild populations of N. glauca STMV is characterized by 17 nm icosah edral particles containing a 1,059 nt ssRNA genome with two ORFs. ORF 1 codes for a 6.8 kDa protein of unknown function. ORF 2 codes for a CP, which, although about the same size as TMGMV CP (17.5 kDa), is serologically unrelated (reviewed by Hull, 2002). Detected only in samples of TMGMV-infected N. glauca from California, STMV has been experiment ally propagated with TMV, ToMV, and ORSV, which will serve as helper viruses. No artificiall y recruited helper virus, however, was nearly as effective as TMGMV in terms of the amount of STMV particles yielded. In contrast, coinfection with STMV significantly reduced the yiel d of helper virions reco vered from inoculated plants. However STMV did not alter the timing or severity of symptom development. Single
32 local lesion (SLL) isolation may be insufficient to separate this satellite virus from its helper detectable quantities of STMV reappeared in 3 out of 10 experimental isolates (Valverde et al., 1991). Because of the close relationship and great similarity between tobamoviruses, the wellstudied replication cycle of TMV probably also serves as a good model for that of TMGMV. Infection begins when plant tissu e is mechanically inoculated by contact between plants or some other contaminated animate or inanimate object. Although a lone viral part icle is theoretically capable of establishing infection, st atistics indicate that different regions of a host vary in their susceptibility to infection such that different quantities of virus are required (Roberts, 1964). Even the most susceptible hosts seem to require 50,000 or more vi rions to initiate a local lesion (LL) (Bawden, 1956). Upon gain ing entrance to a cell, viru s replication begins with cotranslational disassembly as the 5 end of the virus becomes uncoated within 2-3 minutes, exposing the first ORF for tran slation (Heinlein, 2002). Once th e 126 and 183 kDa proteins are being produced, disassembly proceeds to comple tion, and a () RNA template is synthesized. From these () RNA templates, replication of (+) RNA begins, produci ng both full-length, and shorter, ORF 3 / ORF 4 sub-genomic RNAs (Heinlein, 2002). Replication in TMV is associ ated with inclusion bodies in plantae ; X-bodies, characteristic of TMV infection, are localized to endoplasmic reticulum (ER) of the host, perhaps to coordinate the process, or to provide some protection from host defenses. X-bodies have been shown to be composed of the aggregated 126 kD a protein (Lapido et al ., 2003). Replication of negative strands ceases rela tively early, at least, in vitro with translation of MP and CP continuing for some hours afterwards. MP alte rs the size exclusion limit (SEL) of the hosts plasmodesmata, but the resulting slow ( m/h) movement from cell to cell is believed to take
33 place using a ribonucleoprotein co mplex rather than by complete virions (Heinlein, 2002). Virion assembly allows movement as fast as 8 cm/h through the phloem (Dawson, 1992; Palukaitus and Zaitlin, 1986). The mutation rate for RNA viruses is believed to be much higher than in other organisms. Mechanisms that may serve to generate change in such a virus include replication error and recombination (Schneider and Roosinck, 2000). A natural TMV/TMGMV recombinant, H7, was isolated from herbarium specimens preserved in New South Wales, Australia. Recombination appears to have occurred betw een nt 1250 and nt 3461 in the genome (Fraile et al., 1997). Studying an 804 nt region of ORF 3 in TMV, Malpica observed that while there was a high rate of mutation, most mu tations occurred at nucleotidebinding sites (NBS) and were lethal to the virus progeny (Mal pica et al., 2002). When replic ating in the permissive host, N. benthamiana TMV was found to have a mutation frequency of 4.3 x 10-4, which is intermediate between that of Cucumber mosaic virus CMV and Cowpea chlorotic mottle virus (CCMV). The percentage of mutants did not in crease or decrease significantly following passage from host to host. None of the mutations studied survived 10 rounds of serial passage suggesting that they contributed negatively to fitness, and it seems that in the abse nce of severe bottlenecks, the mutant population reaches a stable level of diversity that remain s constant with passage from plant to plant (Schneid er and Roosinck, 2000). Kearney (1999) obtained a mutation rate of 3.1 x 10-4 nt substitutions/base-year while passaging TMV through a variety of plant hosts. E xperimental plants from six families with a variety of responses were inoculated with TM V. Population shifts in TMV had been reported following physiological changes in the host, or following transition from one host to another. Kearny (1999), however, found relativel y little host-shift, and that mutations that did occur were
34 predominantly silent (85%), indicating a st rong tendency towards conservation of the aa sequence. Of the genomic regions studied, the conserved region of the CP gene, the 3 region of the MP, and the replicase, the greatest number of nt substitutions was found in the CP sequence. Fewer mutations were found in the replicase dom ain, and the least number in the MP gene. In another experiment, no changes in a scruti nized region representi ng 12% of the TMGMV genome were observed after a field populat ion was serially passaged 23 times through N. tabacum cv. Samsun. Given the high error rate of viral RNA polymerases, it is surprising that more mutations did not occur. The progenys need to escape the inoculated leaf is a possible mechanism restricting the spread and proliferation of more divergent tobamovirus genomes within an infected plant (Kear ney et al., 1999). This need to conserve the properties of functional viral proteins may reflect a limit of change where back mutations and parallel mutations in the CP of a tobamoviruses are as common as divergen t ones (Hull, 2002). Competition between TMV and TMGMV in N. glauca observed in New South Wales, Australia, has been cited as support for a theo retical evolutionary process called Mullers ratchet (Muller, 1932), whereby repeated bo ttlenecks cause asexually reproducing populations to decline until the minimum diversity of fit genot ypes required to maintain the population is lost (Fraile et al. 1997). In the case of a tobamoviru s, a certain minimum diversity of viral genomes would need to be transferred with each inoculation event to ensure at least one highly competitive genotype gets through to establis h a strong beachhead: one well adapted to conditions in the new host environment. As Fr aile et al. have argued, TMV appears to have arrived in Australia first. As TMGMV reduced the proportion of TMV in co-infected plants, fewer and fewer TMV virions (and therefore fewer to tal genotypes), were av ailable to infect the next host. Eventually, passing through a series of bottlenecks and co-infected hosts, the fittest
35 genotypes were lost, and only less fit genotype s remained in the declining population. The observed number of mutations found in the TMV samples increased, while the population genetics of TMGMV remained unchanged. The end re sult of this process was the extinction of TMV-type strain in New South Wales N. glauca by about 1950 (Fraile et al., 1997; Gibbs, 1999). Tobacco mild green mosaic virus occurs worldwide, wherever N. glauca exists and where N. tabacum and the other host species grow. TMGMV can be expected to occur where such plants are cultivated, especially where pruning or vegetative propagation is involved. TMGMV has been found in several members of the Gesn eriaceae (Zettler and Nagel, 1983), for example, and the ornamentals Rhoeo spatheace (Baker and Zettler, 1988), and Tabernaemontana divaricata (Cohen et al., 2001). Tobamoviruses may be recovered from cigarettes and other smoking materials; the practice of smoking has probably contributed to their current widespread distribution (Agrios, 1997). Both TMV and ToMV cause problems for growers of tobacco, peppers, and tomatoes in Florida (Kucharek et al., 2003). Extensive experimentation with TMGMV, however, has shown that it is incapable of infecting tomato in plantae or in vitro (protoplasts) (Morishima et al., 2003). In Florida, cultivati on of the TMGMV host N. tabacum is of historical importance. As of 2004, tobacco was cultivated commercially in Alachua, Baker, Columbia, Gilchrist, Hamilton, Jefferson, Le vy, Madison, Sumter, Suwannee, and Union counties, in Florida (Mr. William Brown, Direct or of the Alachua Co. Extension Service, personal correspondence). Despite the historical significance of tobacco in the state, Florida ranks as one of the smallest producers of tob acco, and tobacco production in the state has been declining rapidly. In 1996, 7,500 acres of tobacco were harvested in Florida, with a cash value of $36.3 million dollars. In 2006, only 1,100 acres were harvested, bringing in $4.3 million dollars (USDA-NASS, 2007.) The accelerated declin e in production is larg ely the result of the
36 Fair and Equitable Tobacco Reform Act of 2004 (H.R. 4520-105 Title VI Fair and Equitable Tobacco Reform), which ended the quota and pr ice supports for tobacco and created the $10 billion Tobacco Transition Payment Progr am, also known as the Tobacco Buyout (Anonymous, 2007). Worldwide, TMGMV shows little overall gene tic variation, although differences occur in specific gene sequences between geographically isolated populati ons. The greater diversity of genotypes found within a geographically isol ated population of TMGMV compared with populations worldwide led Fraile (1996) to c onclude that there may be an upper limit for TMGMV diversity, which may have been reache d in some populations (Fraile et al., 1996, 1997). Examination of century-old herbarium specimens from NSW, Australia has revealed that genetic diversity in TMGMV has not increased ove r that time (Fraile et al. 1997; Gibbs, 1999). Early researchers did not dist inguish between species of Tobamovirus classifying them, with the best information then available, as strains of TMV. Many synonyms for TMGMV exist, including: Mild dark-green tobacco mosaic virus, "para-tobacco mosaic virus, mild strain of tobacco mosaic virus, South Carolina mild mottli ng strain of tobacco mosaic virus, strains U2 and U5 of tobacco mosaic virus (Siege l and Wildman, 1954), and green-tomato atypical mosaic virus (Wetter, 1989). The misleading gr een-tomato atypical mosaic virus refers to the observation that TMGMV (then considered to be a strain of TMV) would not infect tomatoes, an atypical characteristic when compared to TMV or ToMV. Subsequent experimentation has confirmed that TMGMV is incapable of infecting Lycopersicon esculentum either in plantae or protoplasts (Morishima et al., 2003; my observations). Wetter (1984) compared various isolates of what was then considered to be TMV from around the world using serological techniques and host-range tests and was able to distinguish
37 them from TMV-type strain and group them as the species TMGMV. Furthermore, Wetter was able to prove that TMGMV was a species in it s own right, and not merely a mutation caused by repeated passage of TMV through various hosts (Wetter, 1984). A good immunogen, TMGMV can be distinguished from other tobamoviru s species by immunodiffusion. The 308 nm x 18 nm rod-shaped particles have a sedi mentation coefficient of 186 S a nd an isoelectric point of pH 4.17. TMGMV virions are thermally inactivated in 10 minutes at 85C. Virus preparations may be evaluated using spectrophotometry: the extinction coefficient for TMGMV at A260 = 3.16; a nearly pure preparation should have a ratio: A260/A280 = 1.22 (Wetter et al., 1989). The genome of TMGMV-S is 6355 nt long, as opposed to TMV, which is 6395 nt long. The four essential TMGMV genes range from 480 nt to 3336 nt in size (Table 1-2). Genomic ssRNA is encapsidated in ~2130 subunits, (TMGMV-S) (Wetter, 2004). The specific composition of TMGMV RNA has been determ ined to be: A=30.9%, C=17.2%, G= 23.8%, U = 28.1% (Knight et al., 1962). The mass of a single tobamovirus virion may be determined by adding the mass of the RNA genome (5%) to the mass of the sum of the CP subunits (95%). For TMGMV-S: mw ssRNA = (329.2 An + 305.2 Cn + 345.2 Gn + 306.2 Un + 159), where n represents the ratio of each nucleotide to the total, and 159 is the mass of the 5' triphosphate group; mw genome = (6355 x 0.309 x 329.2) + (6355 x 0.172 x 305.2) + (6355 x 0.238 x 345.2) + (6355 x 0.281 x 306.2) + 159 = 2.05 x 106; mw CP subunits = 17.5 kDa x n Subunits = 17.5 x 103 (2130) = 3.72 x 107 amu; mw TMGMV virion = 2.05 x 106 + 3.72 x 107 amu = 3.93 x 107 amu; mass of TMGMV virion in g = 3.93 x 107 amu x 1.66053873 x10-24 amu/g; mass of TMGMV virion = 6.53 x 10-17 g.
38 Using this information, the number of TMGM V-S virions in a given volume of solution can be determined. For example, 5.0 x 10-7 g / 6.53 x 10-17 g = 7.66 x 109 TMGMV virions would be expected per milliliter of a 0.5 g/mL purified virus solution. More recently, the technique of electrospray ioniza tion (ESI) mass spectrometry has been used to determine that the mass of preparations of intact TMV virions is between 39 x 106 and 42 x 106 Da, which is consistent with the calculated mass of 40.5 x 106 Da (Fuerstenau et al., 2001). Host-Pathogen Interaction While systemic infections by plant viru ses are a very common; albeit largely inconspicuous phenomenon, lethal infections app ear to be uncommon, and have rarely been reported. Research by Pettersen et al. (2000) established that TMGMV elicits a lethal host response from TSA. Visual obser vations of the disease process, particularly the development of necrotic local lesions early on, suggested a hyp ersensitive-response (HR) type mechanism may be involved (Pettersen et al., 2000, 2001). Lacking the immune system that animals po ssess, plants are nevertheless capable of defending against invading organi sms by a variety of means, both physical and biochemical. One common defense mechanism is the hypersensitiv e response (HR). Usually localized, HR is a form of programmed cell death (PCD) and occurs in response to specific pathogen-associated molecular patterns (PAMPs). This response is contingent upon rec ognition of a pathogenassociated molecule, coded for by an avirulen ce (Avr) gene, by the pr oduct of a corresponding host resistance (R) gene. HR is associated with other biochemical plant defenses including the rapid production of reactive oxi dative antimicrobial compounds, upregulation of pathogenesisrelated (PR) protein genes, and reinforcement of the cell walls (A grios, 1997). Transcription of PR genes, in particular, is a useful molecular in dicator that HR is taking place (Hajimorad et al.,
39 2005). Other chemical processes associated with PCD include activation of certain proteases, Ca++ flux, and membrane exposure of phospha tidylserine (Xu and Roossinck, 2000). The exact way in which some plant viruses overcome HR and spread systemically is not known. Despite the increasing facility in finding and identifying viral sequences involved in virus-host interactions, the exact mechanisms a nd processes by which thes e sequences exert their influence are more difficult to observe and unde rstand. Virus-host intera ctions may be more complex than simple gene-for-gene interaction. For example, it is known that several variables may allow a virus otherwise confined by HR to spread systemically, including the age of the tissue infected, alteration of the Av r or R gene or genes, inhibiti on of salicylic acid, inoculation by grafting, and/or extremes of temperature (C ulver et al., 1991; Hajimorad et al., 2005). For convenience, plant hypersensitive respons es may be categorized as localized HR, nonlethal systemic HR (SHR), and lethal system ic HR (LSHR). SHR and LSHR can further be qualitatively described in some instances as trailing HR, a disease process whereby the necrotic areas of the localized HR continue to expand and spread th roughout the host tissue progressively from the point of inoculation (Hajimorad et al., 2005). The trailing HR phenotype may be seen when certain cultivars of N. tabacum are inoculated with TMGMV. Several hypotheses related to the HR or HR-like disease phenot ype observed in the TMGMV-TSA interaction have been proposed. Dr. William Dawson (Eminent scholar, Plant Pathology Department, University of Florida, personal communication) likens the disease process to a horse race, in which the viru s escapes confinement within the LL by rapid replication and/or movement. The progressive necr osis is a HR following the systemic spread of the virus. This hypothesis fits well with the tr ailing HR observed in cert ain tobacco species and
40 cultivars, such as TMGMV on N. benthamiana and may also explain the effects seen on certain cultivars of C. annuum following inoculation with TMGMV (Culver et al., 1991). Culver has proposed a weak elicitor hypothesi s: that systemic necrosis is at least sometimes the result of a weak interaction between the viral elicitor and a host detection mechanism (Culver and Dawson, 1991; Culver et al., 1991). Mutagenesis of TMV produced several CP mutants capable of producing LL on N. sylvestris The mutants provoking a rapid HR generated small LL successful in confining the viru s. Mutants slow to evoke a HR would often escape from the initial lesion to spread systemically, causing necrosis throughout the host. Further experiments with transgenic tobacco plan ts designed to express CP from these mutants revealed that low concentrations of CP from fast-acting mutants (st rong elicitors) produce a more severe necrotic response than high concen trations of CP from slowly-acting mutants (weak elicitors). Many common genes conferring resi stance to viruses including the N gene, have been determined to code for proteins of the To ll-interleukin-1 recept or/nucleotide-binding site/leucine-rich repeat (TIR-NBS-LRR) variety, where the leucine-rich repeat region (LRR) is thought to recognize PAMPs in a wa y perhaps analogous to the antibod ies of animals, that is, by binding directly to the pathogen-associated molecu les. Although a direct in teraction has, at the time of this writing, not yet (to th e authors knowledge) been obser ved, it has been established in some pathosystems that variation observed in LRRs is directly related to specificity of pathogen recognition. Based on this knowledg e, the weak elicitor hypothesis predicts that the binding of the elicitor virus protein to the LRR sticky ta pe of the host R-gene product is weak or inadequate to provoke a strong resp onse (Dinesh-Kumar et al., 2000.)
41 Another way in which the elicitor may fail to provoke an adequate response from the host, leading to LSHR, is when chemicals needed to trigger the defense-related signal cascade accumulate too slowly to confine the virus, analo gous perhaps to water flowing from a faucet to fill a sink with a partially closed drain. Accu mulation of proteins necessary for the HR signal cascade (the flow from the faucet) within the cell (the sink) is balanced against destruction of these proteins by the ubiquitin-protea some system (the drain). If th e viral elicitor fails to cause a rapid enough accumulation of the factor(s) that initi ate the signal cascade, th en the level of that signal fails to rise to the threshol d level in the sink fa st enough to allow HR to occur in time to prevent the virus from escaping (Dr. Wen-Yu an Song, Associate Professor, Plant Pathology Department, University of Florida, Gainesville, personal communication.) Similar to the horse race hypothesis, the delayed localized HR hypothesis posits SHR is the result of delayed biochemical events asso ciated with HR, and represents a failure of Rmediated defense response (Dinesh-Ku mar et al., 2000). Working with the N gene, DineshKumar observed that mutations in the TIR and NBS, as well as the LRR could produce plants showing a delayed HR phenotype in response to i noculation with TMV, and that some deletion and point mutations to the TIR or NBS would pr oduce plants that respon ded to TMV inoculation with necrosis, but allowed TMV to move, resulting in SHR. These hosts not capable of mounting an effective defense are described by Dinesh-Kumar as partial loss-of-function mutants (Dinesh-Kumar et al., 2000). Some evidence exists that host-resistance res ponse is actually two separate phenomena: extreme resistance (ER), believed to be an HR-independent response where the virus is halted at the point of infection before visible sympto ms or accumulation; and hypersensitive response (HR), where specific biochemical changes take place (e.g., production of PR proteins, etc.),
42 leading to a necrotic response which usually lim its virus spread from the general area of inoculation. In the two-tiered hypothesis, ER is believed to operate independently of HR, except inasmuch as it is controlled by the same elicitor and R-gene. Supposedly, HR develops only when the ER function is overwhelmed or evaded by the pathogen (Hajimorad et al., 2005). In one hypothesis, ER operates through PCD of th e inoculated cells, disr upting viral replication and even degradation of pathogen RNA (Xu and R oossinck, 2000). This seems in contrast with the micro HR hypothesis, whereby it is believed that the absence of visible symptoms in highly resistant plants is due to a micro HR which limits the infection to a region of 1-3 cells from the point of entry. Cases of virus-induced lethal system ic hypersensitive response (LSHR) Although uncommon, examples of plant virus-host interactions resulting in LSHR have been reported, and may serve as models usef ul in understanding the TMGMV-TSA interaction. In soybean ( Glycine max ), host to Soybean mosaic virus (SMV ), the (dominant) resistance gene Rsv1 provides immunity against virt ually all strains of SMV, except for strain G7, which causes a LSHR, and the artificial variant G7d, which ca uses non-necrotic systemic symptoms. This variation in virulence was mapped to mutations causing 3 significant amino acid substitutions in the P3 region of the virus. Notably, silent mu tations had no effect on vi rulence, supporting the hypothesis that HR or the lack ther eof is determined by translated pathogen gene products, rather than nucleic material itself. Curiously, LHSR in the soybean-SMV interaction appears to be delayed by up to 6-8 weeks post infection/inocul ation (Hajimorad et al ., 2005); a delay similar to, but much longer than, the delay ob served in the TSATMGMV interaction. Recently, two sub-isolates of a Japanese stra in of Plantago asiatica mosaic virus (PlAMV) obtained from lily were found to have very different disease phenotypes in N. benthamiana plants. When inoculated onto N. benthamiana sub-isolate Li1 produced LSHR, while sub-
43 isolate Li6 produced a latent/asymptomatic infec tion in this host. When sequenced, these two sub-isolates (PlAMV-Li1 and PlAMV-Li6) were surprisingly similar, sharing a 99.7% identity. Analysis of the genome using chimeras of the two isolates revealed mutation in the viral polymerase to be the source of the varia tion in pathogenicity (Ozeki et al., 2006). Certain cultivars of bean ( Phaseolus vulgaris ), develop HR, which is sometimes lethal, in response to inoculation with B ean common mosaic virus (BCMV). As with tobamovirus/host systems, the HR that develops in this pathos ystem is variable and influenced by genotype and environment. Depending on the presence or abse nce of the incompletely dominant allele, I and environmental conditions (i.e., temperature), res ponse to inoculation varies from immunity, to LL, to systemic phloem necrosis (Collmer et al., 2000). Cucumber mosaic virus is a (+) RNA virus w ith a broad host range and a tripartite genome that produces two subgenomic RNAs. CMV causes systemic disease in tomato ( L. esculentum ), usually characterized by systemic mosaic. In so me infections, CMV is accompanied by satellite RNAs (satRNAs), ranging in size from 332405 nt in length, capable of modifying the symptomatology. In particular, one satellite, D4-sat RNA, (335 nt), causes an often lethal systemic necrosis in tomato, which develops la te in the infection period and does not coincide with systemic movement of the virus. This n ecrotic symptom appears to be separate from chlorosis induced or enhanced by satellite RNA, and not all satRNAs will cause this symptom. Furthermore, this symptom is host specific; in tobacco D4-satRNA acts to attenuate CMV symptoms (Xu and Roosinck, 2000). Using a P VX vector system, Taliansky (1998) introduced satRNA into Lycopersicon cells without the CMV helper and observed that only negative-sense D4-sat RNA produced the delayed necrotic symp toms. Normally, the satRNA would enter the cell packaged in the CMV virions as the positive-sense form. CMV genomic RNA was not
44 required. Taliansky was able to identify a specific stem and hairpin loop in the 5 region of the (-) satRNA (the 3 region of the plus sense), required fo r necrogenic activity. Conversion of the stem-loop portion of the 5 RNA changed satRNA (-) B5-sat from a non-necrotic virulence to a necrotic one. It is worth noting th at very high levels of ds satRNA are recovered from extracts of plants infected with CMV and satRNA, and that unlike many other plant satellite viruses, the ratio of (+):(-) CMV satRNA may be as low as 2:1 (Taliansky et al., 1998; Xu and Roosinck, 2000). Xu and Roosinck (2000) established that the cause of the necrosis, observed first in the vascular cells around the area of the second node below the meristem, is indeed the result of PCD. The PCD begins in phloem or cambium cells and spreads to adjacent infected cells. The authors suggest that timing i ndicates the involvement of cell developmental processes in initiating tomato cell death. At 9 10 days postinoculation (DPI), the first visible signs of necrosis appeared in the prevascular cells clos e to the meristem, and occasionally in phloem or cambium cells on one side of the stem. Necrosis th en spread to nearby cells, including those in the leaf petiole and adjacen t tissues below the node. S upporting earlier observations, accumulation of (+)satRNA in the apices of infect ed plants was observed to increase rapidly after inoculation, but rapid accumulation of ( ) satRNA did not occur until such time as necrosis was observed. Another satRNA of CMV causes fruit necrosis Necrosis produced by the CMV-tomato interaction is not well understood, but thought to be the result of incomplete vascularization in the fruit stalk. Fruit necrosis also sometimes occurs when watermelon ( Citrullus lanatus ) is infected by CGMMV. First repor ted from Macedonia and Thessa ly (Greece), starting in 1999 and 2000, symptoms of the CGMMV-induced disease range from chlorotic leaf spots to pedicel
45 necrosis and degeneration of the pulp of the fruit. Symptoms we re observed on grafted watermellons only (Bourbourakas et al., 2004), sugge sting that mode of transmission may have an effect on HR. Cases of gene-for-gene interactions between tobamoviruses and Solanaceae Several gene-for-gene interactions between Solanaceae and tobamoviruses have been studied and characterized. Examin ation of these interactions may lead to insight as to how the TMGMV-TSA interaction occurs. It was discovered that N. glutinosa plants displayed a resistance to tobamovirus infection, marked by the development of LL. The gene res ponsible for this phenotype, N, was introgressed into varieties of N. tabacum by F.O. Holmes, who demonstrated that the inheritance pattern of this resistance was Mendelian (Holmes, 1938). Plants heterozygous, as well as homozygous, for N will exhibit cell death at the site of infection, but the virus part icles are confined to the region of the cell death (Hull, 2002). It is known that the 126 kDa product of the TMV ORF 1 will produce HR necrosis in tobacco homozygous for the N resistance gene. C-terminal fr agments (HEL & MOREHEL) also induced this response, though e xpression of the complete 183 kDa sequence did not. Of particular interest is the findi ng that transient expression of th e shorter fragment, HEL, caused symptoms that were delayed in deve lopment by 11 days (Abbink et al., 1998). Nicotiana sylvestris plants exhibit a response similar to that observed in tobacco plants which have the N-gene when challenged with most tobamoviruses, but not TMV. The elicitor in this interaction is the viral CP, and it has been shown that certain aa substitutions in the CP will abolish the response. The N gene has been successfully (re) incorporated into at least one cultivar of tobacco Bright Yellow. It has been established that the HR is elicited by the CP of tobamoviruses such as ToMV, but not TMV pr oper, although mutants of TMV capable of
46 eliciting a response from this gene have been created (Culver et al ., 1991; Saito et al., 1987; Pfitzner and Pfitzner, 1992). Interestingly, the pedigree of N. tabacum which includes N. sylvestris and N. tomentosiformis indicates N. tabacum should contain N in its genome. Expression of N however, for some reason, is usually s uppressed, leading to susceptibility in most tobacco cultivars. A spontaneous mutant of N. tabacum cv. Samsun ( nn ), Samsun EN does in fact appear to express the N phenotype, as does the N. tabacum cv. Java these are presumably back-mutations (Pfitzner and Pfitzner, 1992). Three genes: Tm1, Tm2, and Tm22, conferring resistance to tobamoviruses have been discovered in tomatoes (genus Lycopersicon ). They are designated Tm, and tomato breeders organize the tobamoviruses they are challenge d with (mostly TMV and ToMV) into Pelham groups, based on their ability to overcome these a lleles. Pelham 0 strains are unable to infect plants with any of these genes, P1 strains ar e capable of overcoming th e resistance conferred by Tm1, P2 strains are capable of overcoming the re sistance conferred by Tm2, and P1.2 strains can overcome Tm1, Tm2, and Tm22 resistance (Hollings and Huttinga, 1976). Tm1 probably originated in L. hirsutum and the resistance it provides has been shown to involve the 126 kDa and/or 183 kDa proteins. Tm1 appears to cause inhibition of the virus to replicate, rather than HR (Hull, 2002). It is relati vely easily overcome, with just two amino acid substitutions being sufficient (Meshi et al., 1988; Pelham, 1966). Tm2, from L. peruvianum and Tm22, from L. chilense (Mueller, 2004) are elicited by t obamovirus movement proteins. Tm22 is particularly durable: the viral mutants capable of overcoming it seem to have low virulence. Furthermore, LL development in resistant plants is microscopic, suggesting rapid recognition of the parasite (Lanfermeijer et al., 2003). Over th e years, resistance from these plants has been painstakingly introgressed into lines of L. esculentum
47 Resistance genes in peppers (genus Capsicum ) are designated L, followed by a number in superscript, in order of increasing effectiveness. All L genes are elicited by tobamovirus CP, and initiate HR, characterized by LL and leaf abscission that usually confines the virus to the site of infection (Holmes, 1937). Amino acid sequencing of the L genes reveal them to be approximately 60% identical to one another (Gil ardi et al., 2004). Breeders often designate (advertise) tobamovirus resistance in peppers by pathotype, based on the viruss ability to overcome resistance conferred by the L genes. Pathotype 0 (P0) viruses (i.e., TMV, ToMV) are able to systemically infect only those peppers lacking resistance (hos t genotype = ll), P1 viruses (i.e., TMGMV) overcome L1 resistance, P1,2 viruses (e.g., PMMoV-J, PMMoV-S) overcome L1 and L2 resistance, P1,2,3 viruses (e.g., PMMoV-I, PMMoV-Ij) overcome L1, L2, and L3 resistance (Gilardi et al., 2004; Tsuda et al., 1998). In studying the natural history of any given gene for disease resistance, it is important to remember that a resistance gene isolated from one member or accession of a species may not be present in all members of that species, or even a common allele within the population. The study of the L genes in the genus Capsicum is complicated not only by th e early lack of differentiation between tobamoviruses (many common tobamoviruse s were initially desc ribed as strains of TMV), but by the various controversial app lication of species epithets within the Capsicum genus. Although Linnaeus distinguished between Capsicum annuum and C. frutescens forms, the name C. frutescens was later adopted to desc ribe all cultivated peppers within the United States, while C. annuum was used to describe all cultivated peppers by the rest of the world. Later research involving plant breeding ultim ately justified Linnaeus separation, with hybridization between C. annuum and C. frutescens possible, but difficult. The bell peppers of
48 commerce, and virtually all other peppers cu ltivated in the United States are properly Capsicum annuum except Tabasco, which is C. frutescens (Smith and Heiser, 1951). In the early part of the 20th Century, Holmes (1937) observe d that certain varieties of garden pepper ( Capsicum sp.) responded to tobamovirus in oculation with either systemic symptoms (mosaic/mottling) or with confining or non-confining necrotic resistance responses, Although at the time, the taxonomies of both genera, Capsicum and Tobamovirus were poorly understood, Holmes work established the presence of the re sistance allele L and laid the foundation for other researchers to better define it. Holmes reported a qu alitative difference in the response between plan ts carrying two allelic L genes (Table 1-3): l No LL, systemic mottling/mosaic, no recovery. Li yellowish or largish LL, abscission, usually followed by recovery. L distinct necrotic LL, ab scission, followed by recovery. Heterozygotes between Li and l responded to inoculation with LL, followed by death. Through cross-breeding experiment s, Holmes determined that L and Li are allelic, (or very nearly so; Holmes, 1937). Holmes called the a llele from Long Red Cayenne the "imperfect localization" gene, or Li, because the lesions formed are large and indistinct, and because occasionally systemic infection results without obvious symptoms. This is most likely the Rgene now represented as L1, since he notes distinct lesions in Tabasco ( C. frutescens ). Holmes showed that Li is partially dominant ove r the recessive condition, l and L is dominant over Li. Holmes reports Li to be present in at least some members of Long Red Cayenne (homozygous) Anaheim Chili, Magnum Dulce, Red Cluster, Ruby King, Sweet Meat Glory, and Sweet Mountain (Holmes, 1937). Holmes reports L to be present in at least some specimens of Tabasco. He reports the heterozygous recessive condition, ll (no resistance), in the following varieties/cultivars: California Wonder, Cele stial Chinese Giant, Coral Gem Bouquet, Early
49 Giant, Giant Crimson, Golden Dawn, Hungaria n, Large Bell, Baby Bell, Oshkosh, Pimiento, Red Cherry, Spanish Monstrous, Upright Sweet Salad, and World Beater. It would be tempting to speculate that a pattern of inheritance similar to the Capsicum -TMV interaction might be responsible for the TSA-tobamovirus interaction; however, since all of th e TSA tested thus far have consistently exhibited LSHR in response to inoculation with TMGMV, the probability that all of these plants (tested in the field or grown from field-collected seeds), should be heterozygous seems highly unlikely. R-gene L1 is derived from Capsicum annuum the common pepper (Hull, 2002), and is the least effective of these resistances. TMGMV can escape localization in these plants, leave the inoculated leaf, and cause severe systemic symptoms, including necrosis. This particular interaction is of special inte rest, because the HR shown by C. annuum appears similar to that shown by TSA, and may eventually yield clues as to the nature and mechanics of the response. L2 resistance occurs in C. frutescens commonly known as Tabasco, and was one of the first viral resistance genes described. It c onfers resistance to most toba moviruses, except PMMoV (de la Cruz et al., 1997). L3 resistance is found in C. chinense a species which includes the Habanero and scotch bonnet types. L3 provides resistance against most tobamoviruses except for certain strains of PMMoV (Gilardi et al., 1998). More recently, L4 resistance, discovered in a wild species of pepper, C. chacoense (USDA GRIN accession PI 260435), has been shown to be elicited by the CP of all tobamoviruses for wh ich peppers are a natural host, including every strain of PMMoV. While the exact mechanis m for this interaction remains unknown, it appears that the gene product of L4 perhaps recognizes a broader struct ural configuration within the CP elicitor, relative to the othe r alleles (Gilardi et al., 2004).
50 In considering the L -gene resistances in pepper, it should be remembered that Capsicum annuum C. frutescens and C. chinense bear many resemblances to each other and are assumed by some to be re-domestications of the same ances tral wild species. In addition, the plants can all be hybridized with some degr ee of difficulty, so gene transfer is possible, and even probable as development and improvement continues by breeder s. Thus, when evaluating for resistance, it cannot be assumed that just because a Capsicum has been classified a part icular species, exhibits resistance, that the resistance allele is the one originally reported from that population (Pickersgill, 1988). Nor can it be assumed that all members of a species will carry a resistance allele as evidenced by the presence of L1 in only some cultivars of C. annuum (Holmes, 1937). Many varieties of eggplant, Solanum melongena exhibit HR when exposed to tobamovirus infection. Differences between varieties may be the result of the diverse genetic background, ( aethiopicum line, anguivi line, melongena line). The HR response was investigated by Dardick and Culver and discovered to be elicited by th e CPs of several tobamoviruses, including TMGMV (U2) (Dardick and Culver, 1997). It has been observed th at various cultivars of Phaseolus vulgaris, especially Pinto, will produce LL when inoculated with TMV, but not TMGMV. This phenomenon is significantly influenced by the growing conditions, but th e speed and convenien ce of generating bean seedlings has made it a useful tool in plant virology (Zaitlin and Israel, 1975). Petunia X hybrida a commercially important hybrid ornamental from S. America develops HR in response to inoculati on with certain tobamoviruses. TMGMV will elicit HR on at least some petunia species and hybrids unde r some conditions. PMMoV strain J produces LL on Petunia X hybrida then mosaic, while PMMoV-Ij produces mosaic only (Tsuda et al., 1998). ToMV also causes the formation of LL (Alexandre et al., 2000).
51 Table 1-1. Strains/isolates of TMGMV Isolate Locality Reference Accession # DSMZ PV0110 France, Corsica Wetter DSMZ PV112 Bald, J.G. DSMZ PV0113, (TMGMV-NZ) USA, Ohio Zettler and Nagel, 1983 EF469769 DSMZ PV0115 Wetter DSMZ PV0116 Wetter DSMZ PV0117 Wetter DSMZ PV0118 Wetter DSMZ PV0119 Wetter DSMZ PV0120 Italy Wetter DSMZ PV0122 Wetter DSMZ PV0124 Italy Marte DSMZ PV0228 Japan Nejidat, et al 1991 ATCC PV226 McKinney, H.H., 1929 ATCC PV228 McKinney, H.H., 1952 ATCC PV585 Valverde, R.A. et al., 1991 ATCC PV586 Valverde, R.A. et al., 1991 ATCC PV635 Solis and Garca-Arenal, 1990 TMGMV-J Japan Morishima et al., 2003. AB078435 TMGMV-U2* Solis and Garca-Arenal, 1990. M34077 TMGMV-U5* USA, California Dodds, J.A., 1998. TMV U Japan Nagai et al. 1987 ATCC = American Type Culture Collection, DSMZ = Deutsche Sammlung von Mikroorganismen und Zellkulturen (German Collec tion of Microorganisms and Cell Cultures)
52 Table 1-2. Size of TMGMV genes and gene products Reading frame Gene (nt) Product (kDa) ORF1 3336 126 ORF2 4830 183 ORF 3 771 28.5 ORF 4 480 17.5 Table 1-3. Relationship between ge notype and disease phenotype in Capsicum sp. peppers. Variety Genotype (Holmes) Genotype (modern) Phenotype TMV (Holmes) Phenotype TMGMV (Elliott) Anaheim Chili LiLi L1L1 LL (imperfect), SYS LL, SYS Hungarian ll ll SYS (mosaic/mottling) SYS only Tabasco LL L2L2 LL (distinct) only LL only Response of Capsicum sp. cultivars to inoculation with TMV (Holmes, F.O. 1937) and TMGMV (Mark S. Elliott, unpublished data)
53 Figure 1-1. TSA colonizing pastureland [Cred it: Dr. Raghavan Charudattan, University of Florida]. A B Figure 1-2. TSA infestations A) TSA infes ting pastureland [Credit: FL Division of Plant Industry] B) TSA infesting wooded area [C redit: Dr. Julio Medal, University of Florida]. Credit: Dr. Raghavan Charudattan, University of Florida
54 Figure 1-3. Genome organizati on map of Tobacco mild green mosaic virus (TMGMV; not to scale). Genome is capped 5 with 7-methylguanosine (shaded pentagon). Omega ( ) leader sequence is followed by open readi ng frames (ORFs) 1,2, 3 and 4 that are directly translated (arrows) into replicat ion-associated (126 kD a, 183 kDa) proteins, and subgenomically translated into movement (MP) and co at (CP) proteins. Circles represent RNA pseudoknots (this region is duplicated in TMGMV-L), and are followed by the 3 tRNA-like structure.
55 CHAPTER 2 IDENTIFICATION OF COLU MNEA ISOLATE PV-0113 Introduction Viruses, nearly impossible to isolate as discrete entities, may be described as quasispecies, with multiple, related genomes serving as a source of genetic variability allowing the population to adapt to changing environmental conditions (Hull, 2002). Techniques such as single local lesion (SLL) isolati on have been used to restrict the variability of the virus population in the hopes of obtaining single-genotype isolates. Tobacco mild green mosaic virus (TMGMV) exists as at least two distinct sub-types, dist inguished by duplication of the pseudoknot-containing region of the genomic 3 non-translated region (3 NTR) (Bodaghi et al., 2000). Small-type TMGMV (TMGMV-S), referred to in older literature as the U2 strain of TMV, appears to co-exist in nature alongs ide large-type TMGMV (TMGMV-L) or U5, although mixed infections of the two are rare. Th ese two subtypes appear to compete within the hosts Nicotiana glauca and N. tabacum cv. Xanthi, usually driving one subtype or the other to extinction. Mixed infections sometimes develop in plants initially inoc ulated with TMGMV-S, especially when challenged only a short time afterwards (Bodaghi et al., 2004). The effect of a mixed infection of TMGMV-S and TMGMV-L on TS A has not been reported. Furthermore, three species of tobacco were identified which di splay different symptoms that are distinct and diagnostic following inoculation with TMGMV-L or TMGMV-S. Finally, wild populations of TMGMV in N. glauca sometimes are associated with Satellite tobacco mosaic virus (STMV). Satellite viruses, including STMV, may attenuate or modify expression of the helper virus disease phenotype, at least in ce rtain hosts (Dodds, 1998). The pr esence or absence of STMV in PV-0113 isolate is a concern, sin ce SLL isolation may be inadequate to remove STMV from a TMGMV culture (Valverde et al., 19 91). It is possible that any STMV present could affect the
56 symptomatology of the TMGMV-TS A plant-pathogen interaction. Techniques to detect and identify STMV include serology and double-st randed RNA (dsRNA) extraction and analysis. In the case of the Columnea isolate of TMGMV, PV-0113, there are three sorts of contamination that might have an influence on the disease process or otherwise influence the outcome of experiments using that culture. First, a culture might contain multiple virus species, a possibility that may be assessed by inoculating a range of differential ho st plants, serology, or reverse transcriptase-polymerase chain reaction (R T-PCR). Next, there is the possibility that multiple genotypes within a species might interact to cause, enhance, or inhibit the tropical soda apple (TSA)-killing effect of TMGMV PV-0113. Th en, there is the possibi lity that satellite RNA or a satellite virus may be a ssociated with the isolate. The objective of this chapter is to confirm the identification a nd purity of the TSA-killing Columnea isolate DSMZ PV-0113 as TMGMV-S (TMV U2) by host range determin ation, ELISA, immunodiffusion, and diagnostic RT-PCR methods. Methods used to determine plant viral infection include physical, in plantae serological and nucleic acid-based technique s. Physical techniques, su ch as observing stained viral inclusions using light microscopy and direct observation of virus par ticles under the electron microscope, may reveal virus morphology and provide clues as to the virus family or genus. Other methods are used for identification to species. In plantae techniques, such as the use of indicator plants and host-range studies, involve inoculation of living plants with the viral pathogen. Techniques based on serology use antibodi es to detect viruses, and include enzyme linked immunosorbent assay (ELISA), immunodiffusion, im munoblotting, and fluorescent antibody microscopy. Molecular biology techniques ba sed on analysis of ex tracted nucleic acids,
57 such as RT-PCR, dsRNA analysis, Northern blo tting and Southern blotti ng are also useful in accurately and precisely identifying pl ant virus infections (Agrios, 1997). Disease phenotype and host suscepti bility to a virus infection can be highly variable under diverse growing conditions. Furthermore, mixe d infections of viruses are not uncommon in nature, and sometimes the virulence of one viru s can be enhanced or inhibited by another. Tobacco leaves infected with Potato virus X (PVX), for example, develop more local lesions (LL) when inoculated with Tobacco mosaic vi rus (TMV) than when infected alone. This increase is associated with a severe veinal necr osis not seen in singly infected plants. (Hull, 2002). Tomato mosaic virus (ToMV) also interact s synergistically with PVX, causing a severe necrotic streak symptom in tomatoes that often leads to the death of th e plant. Infection by Potato virus Y (PVY) in tobacco plants can in crease the concentration of PVX in the leaves tenfold, and in potato the two viruses synerg ize, producing a destructive rugose mosaic symptom. Likewise, infection by a less virulent virus, or strain of virus, may cross-protect a host plant against a more virulent one, greatly reducing disease seve rity. In addition to modifying the visible symptoms, infection by one virus may have other signi ficant effects on the biology of a co-infecting species; for example, a virus normally incapable of movement in a particular host may spread systemically (Hull, 2002). Brome mosaic virus (BMV), for example, gains the ability to systemically infect tomato when i noculated along with TMV (Carrington et al., 1996). One of the most venerable approaches adopted by plant pathologists, the use of differential host range and indicator plants, requires time and greenhouse space, but is relatively simple and straightforward. In addition to the time and sp ace requirements, drawbacks include the need for careful vigilance to avoid and identify potential mixed infections. Desp ite these limitations, host range testing may be more di scriminating than certain othe r techniques, allowing one to
58 distinguish strains of a virus th at serological methods cannot (Hu ll, 2002). Plants selected for host range should respond di stinctively to inf ection by different strains a nd/or species of virus, allowing the creation of a logica l table to distinguish between them. One criterion for host selection is the use of a Plant Disease Index (Alfieri et al., 199 4) or another reference for the locality from which the infected source material originated. There is li ttle sense in beginning a study by testing for a species not kno wn to occur within the geographi c area, or in a host it is not known to infect. Tobamoviruses, as a group, are said to ha ve a wide host range (Gibbs, 1977; 1986). TMGMV is known to infect species in the Apiaceae, Apocynaceae, Boraginaceae, Chenopodiaceae, Commelinaceae, Gesneriaceae, and Solanaceae (Wetter, 1989; Randles, 1986; Cohen, et al., 2001). Chenopodium quinoa Datura stramonium Eryngium planum Lycopersicon esculentum and various species and cultivars of Nicotiana have been used for diagnosis of various tobamovi ruses (Wetter, 1989). Unlike TMV and ToMV, TMGMV will not infect tomato (Morishima et al., 2003). TMGMV can be separated from mixed infections with Pepper mild mottle virus (PMMoV) or TMV by inoculating E. planum which the TMGMV will infect systemically, though the infection may become symptomless (Johnson, 1947; Wetter et al., 1984). Conversely, TMV may be separated from TMGMV using Nicotiana sylvestris as a host, and various species of Capsicum ( C. chinense C. chacoense ) may be used to isolate PMMoV (Gilardi et al., 1998, 2004). Serological techniques involve the use of antibodies produced by laboratory animals immunized with purified virus. Immunodiffusion detects viruses and hints at relationships between viruses using the property of immunop recipitation which occu rs when antibodies specific to a plant virus encounter and bind to sites on viral coat proteins (antigens), forming a
59 visible matrix within the gel me dium (Purcifull and Batchelor, 1977). Anim al-derived antibodies may also be used for ELISA, where antibodies chem ically conjugated to an indicator are used to highlight or reveal antigens bound (either directly or indirectly) to the surface of wells in a polystyrene microtiter plate. The plates can then be analyzed using a spectrophotometer, specially adapted for this purpose. In PCR, a process popularized by Kary Mullis the enzyme Taq polymerase is used in conjuction with multiple heating and cooling cycles to multiply a template DNA strand into as many as a billion copies, enabling highly sensitive detection and identification of DNA sequences. With a properly designed primer set, PCR results are usually quite specific (Metzker and Caskey, 2001). To detect RNA viruses, su ch as tobamoviruses, the enzyme reverse transcriptase (RT) may be employed to create a cDNA template from which the PCR reaction can proceed. This may either be performed sepa rately beforehand, or in the same tube as the PCR reaction (Sambrook and Russell, 2001). A seri es of primers has been designed that could be used to selectively amplify 750-800 bp fragme nts of different tobamovirus species, which, when subjected to restriction en zyme analysis provide identification of the specific strain or pathotype (Letsche rt et al., 2002). Materials and Methods Plants of Arachis hypogaea cv. Virginia Jumbo, Capsicum annuum X Camelot, Capsicum annuum cv. Jalepeo M, Capsicum frutescens cv. Tabasco, Cucumis sativus cv. Poinsett, Cucurbita pepo cv. Early Straightneck, Datura stramonium Eryngium planum, Lycopersicon esculentum cvs. Beefsteak and Rutgers, Nicotiana benthamiana N. glutinosa N. langsdorfii, N. rustica N. sylvestris, N. tabacum cv. Samsun(both nn and NN genotypes), Phaseolus vulgaris cv. Pinto, Vigna unguiculata cv. Cream, and Zinnia elegans were grown from seed sown in Metromix potting medium in 4 (10.16 cm) clay pots. Upon attaining a
60 convenient size (4-6 weeks), plants were inoculat ed with extracted tobacco sap containing virus provided by F.W. Zettler as TMGM V, and hereinafter referred to as TMGMV PV-0113 (Zettler and Nagel, 1983). Sap extract from TMGMV PV-0113 infected toba cco, previously maintained frozen at 20C, was diluted 1:10 in 20 mM NaHPO4/ Na2PO4 inoculation buffer (pH 7.2). Diluted sap was applied to both surfaces of each of three carborundum-dusted leav es of each host-range plant by rubbing with sterile cotton cheesecloth pads. For ease of comparison, one specimen of each species of indicator host was inoculated with buffer only and allowed to mature alongside the treated specimen. Host-range plants were observed daily fo r development of symptoms. Four common, Solanceae -infecting tobamoviruses, TM V, TMGMV, ToMV (kindly provided by Mark S. Elliott, Plant Pathology Department, University of Florida/IFAS, Gainesville), and PMMoV (kindly provided by Dr Carlye Baker, Florida Division of Plant Industry, Gainesville), were prop agated in their Commonwealth Mycological Institute (CMI)recommended propagation hosts (Zaitlin, 1975; Wetter, 1989; Hollings and Huttinga, 1976; Wetter and Conti, 1988), and inoculated onto various host-range plants, including Datura stramonium Lycopersicon esculentum cv. Rutgers, and Nicotiana sylvestris A good propagation host accumulates vi rus rapidly and allows for easy purification of the virus under study. A good maintenance host is long-lived, compact in habit, and tends to resist infection by virus species other than the one under study. Polyclonal antibodies to sodium dodecyl sulf ate (SDS)-dissociated PV-0113, derived from a rabbit antiserum prepared by Zettler and Nage l (1983), and kindly provided by M. S. Elliott were used to detect TMGMV PV-0113 and explor e its serological relati onship with other known tobamoviruses. Immunodiffusion was carried ou t in the manner of Purcifull and Batchelor,
61 1977. Briefly, 3.0 g of tissue, collected 3 week s post-inoculation (WPI) from tobacco plants inoculated with TMGMV PV-0113, were triturate d in a sterilized mortar under 3.0 mL of distilled H2O (dH2O) to create a 1:1 homogenate, whic h was filtered through a double layer of sterilized, dH2O-saturated cheesecloth. Two 0.75-mL porti ons of the filtrate were placed in separate plastic Falcon culture tubes. To the first portion, 0.75 mL of dH2O was added, and to the second, 0.75 mL of a 3% solution of SDS in dH2O was added. The samples were mixed by inversion and used to load the immunodiffusion pl ates. A control for the serological tests was tissue extracts from uninoculated tobacco plants. Wells punched in chilled immunodiffusion medi a plates containing 0.8% noble agar, 1.0% NaN3, and 0.5% SDS, were loaded with sample a nd incubated in a moist plastic chamber at 27 oC for 24-48 hours or until precipitate bands were clearly developed and discernable to the naked eye. Samples were also loaded into non-SD S plates containing 0.7% noble agar, 0.85% NaCl, 0.03% NaN3, and 50 mM Tris, pH 7.5. When necessar y, plates were cleared using activated charcoal (15 g/100 ml H2O) and photographed using a li ght box and a Nikon Coolpix 5700 digital camera. Polyclonal antibodies from rabbit to SDS-disso ciated PV-0113 were also used in ELISA, which gave strong positive result s for most symptomatic (and so me asymptomatic) host-range plants tested. Although the rabbi t anti-TMGMV serum used in these experiments was prepared to purified, SDS-dissociated PV-0113 itself, reciprocal homologous reactions between PV-0113 and TMV U2 and antisera prepared to them ha ve been reported (Zettler and Nagel, 1983). Briefly, tissue samples were collected from bot h inoculated and mock-inoculated control hostrange plants. Approximately 0.3 g of frozen-t hawed tissue from each sample was processed through a rotary sap-extractor using 3 mL of extraction buffe r containing 1x phosphate buffered
62 saline + Tween-20 (PBST), 0.2% bovine serum albumen (BSA), and 10 mM Na2SO3, adjusted to pH 9.6. One hundred (100) l of 1:10 extract from each sample was loaded directly into 3 wells of a 96-well ELISA plate and incubated at 37o C for 1 hour. Plates were washed 3 times with 1x PBST with 3 minute soaks in PB ST between washes. The sample and control wells were then loaded with rabbit anti-TMGMV se rum cross-absorbed to healthy N. tabacum tissue and incubated at 37 oC for 1 hour. Rabbit anti-TMV and rabbi t anti-ToMV sera were also used on some plates for comparison (data not shown). Rabbit antisera were diluted in conjugate buffer containing 1x PBST, 0.2 % BSA, 2 % PVP-40, a nd pH adjusted to 7.4. Following incubation, wells were again washed 3 times for 3 min with 1x PBST. The washed wells were loaded with a solution of goat anti-rabbit IgG conjugated to alkaline phosphatase dilu ted 1:30,000 in conjugate buffer and again incubated at 37 oC for 1 hour and washed 3 times with PBST. After incubation and washing, plates were loaded with a freshly-prepared 1 mg/ml solution of pnitrophenylphosphate (PNP) in a substrate bu ffer containing 0.92 mM diethanolamine, 0.02 % sodium azide, pH 9.8., and incubated in the dark for 15 minutes, or until a yellow reaction color was visible to the naked eye. Absorbance at 405 nm was read at 5, 15, and 30 minutes. Samples with average absorbance read ings 3 times the background (neg ative control readings) were assessed positive. Presence of TMGMV RNA was confirmed using RT-PCR with the diagnostic primers (Fraile et al., 1996) 5 -ATGCAGCTTCCATTTTGGCAG-3 and 5 GGTAAGTTAACGCTTTGGCTTG-3 (kindly provided by Kris Beckham, Plant Pathology Department, University of Florida/IFAS, Gainesv ille). Briefly, total RNA was extracted from a SLL isolate using Trizol extracti on reagent as follows: a leaf disk (~0.2 g) was excised using the lid of a sterile 1.5 mL micro-ce ntrifuge tube and homogenized using a sterile, RNA-ase-free
63 Kontes micropestle and Trizol reagent. The homogenate was extracted with 0.2 mL CHCl3 and centrifuged at 12,000 x g for 10 minutes. The aqueous (upper) layer was removed, and RNA precipitated using volume icecold isopropanol. The RNA pellet was washed once with 75% ethanol, air dried under vacuum (aspirator), a nd suspended in 25 L diethylpyrocarbonate (DEPC)-treated water. RT was carried out using the Titan One-Tube kit (R oche). Each reaction contained 250 M deoxynucleoside triphosph ate (dNTP), 1x RT-PCR buffer, 1.5 mM MgCl2, 0.4 M of each primer, 1 L AMV RT/Taq polymerase solution, and 2-5 L of diluted (1:10, 1:100) or undiluted template RNA, prepared as described above, for a total volume of 50 L per reaction. RT-PCR was carried out in a Biometra ther mocycler using sterile mineral oil to prevent evaporation. The program used consisted of : 45C for 30 minutes, 94C for 3 minutes, and 35 repetitions of 94C for 30 seconds, 60C for 1 minute, and 72C for 1 minute. After 35 repetitions, PCR-generated fragment s were allowed to extend for 5 minutes at 75C, then cooled to 4C and held for electrophoretic analysis. Samples of PMMoV, TMV, ToMV, TMGMV PV-0113, American Type Culture Collection (ATCC) TMGMV accession PV-586, and tw o strains of TMGMV from Lake Alfred, FL, provided as U2 (TMGMV-LAU2), a nd U5 (TMGMV-LAU5), along with two unidentified virus infections in TSA were inoculated onto the CMI-recommended propagation hosts. At 3 WPI apical tissue was harvested, and total RNA was extracted from the tissue samples using Trizol extraction reagent as described for PV-0113. Extracted template RNA was adjusted to approximately 1-5g/L in molecular biology grade (MBG) H2O, and diluted further 1:10 and 1:100. To 2-5 L of each template RN A were added 1 L reverse primer (Tob-Uni_1), 1 L 10mM dNTP, and 12 L MBG H2O. This mixture was heated to 65C for 5 minutes and then chilled to 4C briefly before adding 4 L 5X first-strand buffer, 2 L 0.1 M DTT and 1 L
64 RN-ase OUT enzyme (40U). Samples were then heated to 37 C for 2 minutes before addition of the reverse transcriptase (RT). First-strand synthesis was car ried out at 37C using 0.5 L Moloney murine leukemia virus reverse transcriptase (Invitrogen) for 1 hour, after which the enzyme was inactivated by heating for 15 minutes at 70C. PCR amplificatio n was carried out using an Invitrogen kit, with a mixture (for each reaction) of 16 L MBG H2O, 2.5 L 10X reaction buffer, 0.9 L 50 mM MgCl2, 2.5 L 2 mM dNTP, 0.125 L of Taq polymera se, and 1 L of each specific primer (Table 2-1). Programmed PCR was carried out as: 94C for 5 minutes, and 25 repetitions of 94C for 1 minute, 55C for 45 seconds, and 72C fo r 1 minute. After 25 repetitions, the PCRgenerated fragments were allowed to extend for 5 minutes at 72C, then cooled to 4C and held at that temperature until el ectrophoresis on 0.9% agarose. Results Symptoms caused by this is olate on the propagation host N. tabacum cv. Samsun ( nn ) were generally mild and included a greenish mosaic symptom frequently coupled with leaf pleating, and flower color-break (Figure 21). On the maintenance host, E. planum TMGMV PV-0113 often produced a yellowish spot ting or flecking (Figure 2-2). On several plant species used in the host range (Table 2-2), TMGMV PV-0113, produced symptoms including but not limited to chlorosis, mosaic, necrosis confined to LL, necrosis not confined to LL, stunting, leaf and fruit abscis sion, and fruit lesions. When compared to host range data reported in CMI/AAB bulletins (Z aitlin, 1975; Hollings and Huttinga, 1976; Wetter and Conti, 1988; Delgado-Sanchez and Grogan, 1970; Purcifull and Hiebert, 1982; Francki et al., 1979), and Plant Viruses Online (Brunt et al., 199 6), results obtained from the host-range study were consistent with the previous identific ation of TMGMV, and ruled out other common tobamovirus (TMV, ToMV, PMMoV) and non-tobamovirus (CMV, TEV, PVY) contaminants.
65 Results of SDS/non-SDS immunodiffusion us ing the common tobamoviruses TMV and ToMV for comparison were consistent with the host-range diagnosis of TMGMV for the Columnea (PV-0113) isolate. Rabbit antiserum pr epared to SDS-dissociated TMGMV PV-0113 (Zettler and Nagel, 1983) detected the virus in both SDS (Figur e 2-3) and non-SDS (Figure 2-4) immunodiffusion media. TMGMV PV-0113 did not react with ra bbit anti-ToMV serum (Figure 2-5), and reacted to rabbit anti-T MV only weakly (Figure 2-6). ELISA using rabbit antiserum to PV-0113 yielded data consistent with the im munodiffusion data and an identification of TMGMV (Table 2-3). Antisera to TMV and ToMV produced a weak reaction but did not consistently produce a reac tion exceeding the cutoff. The diagnostic primers for TMGMV consistent ly amplified RNA extract from infected tissue samples, producing single bands of th e expected size for TMGMV (Figure 2-7). In the initial run of the experiment using the tobamovirus analysis prim ers (Letschert et al., 2002) every tobamovirus isolate produced a positive PCR product using the universal tobamovirus primer. In addition, every TMGMV isolate except the culture obtained from the Citrus Research and Education Ce nter (CREC), Lake Al fred, as U2 tested positive using the primers designed to be specific to TMGMV. The U2 culture obtained from Lake Alfred tested positive using the universal tobamovirus primer and the primer specific to TMV. The TMV isolate behaved as expected, with clear posi tives from only the universal and TMV-specific primers. The rest of the cu ltures processed under this syst em produced nonspecific product bands when analyzed on agarose. Isol ates TMGMV PV-586, ToMV, and PMMoV also developed bands in the reactions using the T MV-specific primer set. There was a positive result for the TMGMV PV-0113 culture using th e ToMV-specific primer, and also the PMMoV primer, although this latter reacti on resulted in a product of incorre ct size. The culture obtained
66 from Lake Alfred as U5 tested positive using the ToMV primer set, as did the PMMoV culture. The ToMV virus culture itself behaved more or less as expected although there were bands generated by the TMGMV and TMV primer se ts, these were not within the size range acceptable as positive. The PMMoV culture obtain ed from Dr. Carlye Baker (Florida Division of Plant Industry, Gainesville) tested positiv e using all primer sets EXCEPT the PMMoVspecific primer set. Raising the annealing temper ature to 65C resulted in no product. In an attempt to adapt or optimize this system to work with the virus cultures under study, a two-factor experiment using cDNA prepared from TMV wa s devised using cDNA dilution and temperature as the independent variables. The primers will amplify product at 50C, but may produce non-specific products. It would be important to check the size of the PCR products when annealing at this temperature. Lower annealing temperatures appe ar to increase sensitivity, but also produce more non-specific products. Products of sizes outside the range of 600-1000 bp should be ignored. First-strand DNA may be diluted 100 fold at lower annealing temperatures, but not at higher temperatures. The optimum temperature for this system will be somewhere between 50C and 60C. No reaction was seen; even with the universal tobamovirus prim er set and undiluted cDNA above 60C, indicating that the system is not useful at these temperatures. Discussion Collectively, the results of these tests indicate that the TMGMV PV-0113 culture contained TMGMV-S and was unlikely to contain any other virus reported to infect TSA. Compared with tobamoviruses liste d in the Florida Plant Disease Index (Alfieri et al., 1994), and viruses recovered locally from TSA (McGovern et al., 1994), or artificially transmitted to it (Chagas et al., 1978), symptoms observed on inoculated host-range plants fit best with the symptoms on these plants caused by TMGMV, but not well with other candidates. In the course
67 of this study it was determined that the co mmon tobamoviruses, TMV, TMGMV, ToMV, and PMMoV could be distinguished from one another by symptoms elicited in just three kinds of differential indicator plants: D. stramonium L. esculentum and N. sylvestris (Table 2-2). The serological tests ELISA and immunodiffusi on supported the host-range determination. Rabbit antiserum prepared to SDS-dissociated PV -0113 reacted with virus recovered from tissue, indicating that the coat protein (CP) is of the isolate TMGMV and not one of the other common tobamoviruses, ToMV or TMV type-strain. Reverse transcriptase-PCR using various prim ers amplified a product of size and quantity consistent with TMGMV. A series of species-s pecific primers (Letschert et al., 2002), proved useful for sensitive tobamovirus detection, but not for discrimination/identification under the conditions used. The optimal temperature for the TMV-specific primer set was determined to be 50C < y < 60C.
68 Table 2-1. Species-specific primers used in RT-PCR. Name 5 3 Sequence Nt position Tob-Uni_1 ATT TAA GTG GA S GGA AAA VCA CT 6283 -6260 Tob-Uni_2 GTY GTT GAT G AG TTC RTG GA 5479 -5498 TMV CGG TCA GTG CCG AAC AAG AA 5609 -5589 ToMV CGG AAG GCC TAA ACC AAA AAG 5618 -5597 PMMoV GGG TTT GAA TAA GGA AGG GAA GC 5617 -5595 TMGMV AAR TAA ATA AYA GT G GTA AGA AGG G 5590 -5565 Species-specific tobamovirus primers proposed by Letschert et al., 2002.
69 Table 2-2. Experimental host range and symptoms caused by TMGMV PV-0113. Family Species Common name Symptoms Apiaceae Eryngium planum NS/YF Asteraceae Zinnia elegans NS/NS Cucurbitaceae Cucumis sativus cv. Poinsett Cucumber NS/NS Cucurbita pepo cv. Early straightneck Squash NS/NS Fabaceae Arachis hypogaea cv. Viginia jumbo Peanut NS/NS Phaseolus vulgaris cv. Pinto Pinto Bean NS/NS Vigna unguiculata cv. Cream Cowpea NS/NS Solanaceae Capsicum annuum X Camelot Bell pepper Ab, Ch, LL/N, PD Capsicum annuum cv. Jalepeo M Jalepeo pepper LL/N, PD Capsicum frutescens cv. Tabasco Tabasco pepper Ab, LL/NS Datura stramonium Jimsonweed LL/NS Lycopersicon esculentum cv. Beefsteak Tomato NS/NS L. esculentum cv. Rutgers Tomato NS/NS Nicotiana benthamiana Ch, N/N, W, PD N. clevelandii Ch/Ch, M, SP N. glutinosa LL/NS N. langsdorfii LL/NS N. rustica Wild tobacco LL/Ch, N, PD N. tabacum cv. Samsun ( nn ) Tobacco NS/FM N. tabacum cv. Samsun ( NN ) Tobacco LL/NS N. sylvestris LL/NS Symptoms on inoculated leaves and upper uninoculated leaves i ndicated to left and right of slash, respectively: Ab = abscission, Ch = chloro sis, FM = faint mosaic, LL = local lesions, M = mosaic, Mo = mottling, N = necrosis, NS = no symptoms, PD = plant death, SP = shoot proliferation, W = wilting, YF = yellow fleck ing or speckling (not always visible). Table 2-3. ELISA using cr oss-absorbed antiserum. Antibody dilution Healthy-sap ( N. tabacum ) TMGMV PV-0113 % difference A405 A405 1:2500 0.058 0.282 490 1:2500 X-abs 0.053 0.218 410 1:5000 0.044 0.183 420 1:5000 X-abs 0.037 0.182 490
70 A. B C D Figure 2-1. Sympto ms of TMGMV on N. tabacum cv. Samsun ( nn ) A) Mild green mosaic and pleating on leaves B) Close-up of symptoms on a l eaf C) Flowers of an uninoculated plant D) Flower break symptoms on a TMGMV PV-0113 inoculated plant. A B Figure 2-2. Eryngium planum A) Symptomless (mock-inoculated) E. planum B) Yellow fleck symptoms on TMGMV PV-0113inoculated E. planum.
71 Figure 2-3. TMGMV PV-0113 reacts with rabbi t anti-TMGMV serum in SDS media. 1,2 = TMGMV PV-0113; 3,4 = ToMV; 5,6 = TMV; A rabbit anti-TMGMV serum. A 2 3 4 5 6 1
72 Figure 2-4. TMGMV-PV-0113 reacts with rabbi t anti-TMGMV serum in non-SDS media. 1,2 = TMV, 3,4 = TMGMV PV-0113, 5,6 = healthy tobacco sap (no antigen); B = rabbit anti-TMGMV serum. 1 2 3 4 5 6 B
73 Figure 2-5. TMGMV PV-0113 does not react with rabbit anti-ToMV serum in SDS media. 1,2 = TMGMV PV-0113; 3,4 = ToMV; 5,6 = TMV; B = rabbit anti-ToMV serum. B2 3 4 5 6 1
74 Figure 2-6. TMGMV PV-0113 reacts weakly with rabbit anti-TMV seru m in SDS media. 1,2 = TMGMV PV-0113; 3,4 = ToMV; 5,6 = TMV; B = rabbit anti-TMV serum. C2 3 4 5 6 1
75 Figure 2-7. RT-PCR products of TMGMV PV-0113 and TMV, amplified with diagnostic primers. 1 = H2O only, 2 = negative control ( no template), 3,4,5 = TMGMV PV-0113 from production house, 6,7,8 = SLL inoculat ed plants (asymptomatic), 9 = TMV (using Manjunath TMV primer set), L = 1 kB Invitrogen ladder, E = empty lane TMV Primer 1 (423): 5 ATG TCT TAC AGT ATC ACT AC 3 TMV Primer 2 (424): 5 TCA AGT TGC AGG ACC AGA G 3 Source: Chandrika Manjunath TMGMV Primer 1 (400): 5 ATG CAG CTT CCA TTT TGG CAG 3 TMGMV Primer 2 (397): 5 GGT AAG TTA ACG CTT TGG CTT G 3 Source: Virology (1996) 223: 148-155. 1016bp 517 bp 2 1 34 56789L E E
76 CHAPTER 3 COLUMNEA ISOLATE PV-0113: SUBTYP E, SATELLITE, AND DISEASE PROGRESS Introduction Tobacco mild green mosaic virus (TMGMV) has been found naturally associated with Satellite tobacco mosaic virus (STMV) in Califor nia. Satellite viruses, including STMV have been known to attenuate or otherwise modify the expression of the helper virus disease phenotype in some hosts (Dodds, 1998). Single local lesion (SLL) passage ma y be inadequate to remove STMV from a TMGMV culture (Valverd e et al., 1991); the presence or absence of STMV in the Columnea isolate, TMGMV PV-0113, is therefore a concern. It is possible that the presence of STMV could affect symptomato logy of the TMGMV-TSA interaction. Direct methods for detecting/observi ng STMV infection include cent rifugation and dsRNA extraction, where the dsRNA extract will reveal a band of 1059 bp when subjected to electrophoresis on agarose or polyacrylamide. The task of detecting contaminants is co mplicated by the tendency for the heritable material of viruses to change during replication, giving rise to new genotype s or strains. The countless multitudes of virions in every host constitu te a quasi-species a collection of related genomes undergoing constant pressures of mutation, competition, and selection. The quasispecies status of a virus populat ion helps ensure survival by maintaining a pool of genotypes such that at least some members may survive a nd proliferate should conditions change. Should a favorable mutation arise, the distribution of ge notypes within the quasi-species will shift to accommodate it (Hull, 2002). One method for reducing genotypic variability in a plant virus culture is isolation from a single local lesion (SLL). Although at least 50,000 vi rus particles are said to be required to initiate infection (Roberts, 1964), in dividual local lesions (LL) may be excised in order to restrict
77 the culture to the genotype of the founder populations at the infection foci. Subsequent repeated inoculations onto hypersensitive hosts cause the gene pool to b ecome increasingly restricted. The two subtypes of Tobacco mild green mosaic virus, TMGMV-S and TMGMV-L, are believed to compete with each other in Nicotiana glauca and N. tabacum cv. Xanthi, usually resulting in the extinction of one subtype or the other within th e host. Mixed infections in plants first inoculated with TMGMV-S are possible, at least for a limited time period after the primary infection (Bodaghi et al., 2004). The effect of a mixed infection of TMGMV-S and TMGMV-L on tropical soda apple (TSA) has not been reported, nor attempted. Some tobacco species have been reported to express differential symptoms following inoculation with TMGMV-L or TMGMV-S (Bodaghi et al., 2000). Nicotiana benthamiana developed wilting 5 days after inoculation wi th TMGMV-L, but not until 7 days following inoculation with TMGMV-S. Death occurred in TMGMV-L inoculated plants by day 10, but not until 14 days post-inoculation (DPI) in TMGMVS inoculated plants. Small-type TMGMVinoculated N. clevelandii plants were systemically necrot ic by 15 DPI, whereas large-type TMGMV-inoculated plants survived with shoot proliferation and rugosity. Nicotiana rustica plants developed LL in response to inoculation with either subtype, but only those inoculated with TMGMV-L developed systemic infection. Nicotiana rustica inoculated with TMGMV-L also developed larger LL (Bodaghi et al., 2000). The objective of this study/chapter was to first determine to which subtype TMGMV PV0113 belongs, based on the reaction of inoculated differential indicator plants and by RT-PCR. The second objective is to assess the presence/a bsence of STMV in this isolate by subjecting extracted dsRNA was subjected to electropho resis in agarose and polyacrylamide. Immunodiffusion using rabbit antiserum prepared to STMV was also used to detect STMV. To
78 better characterize and quantify the disease process produced in TSA by TMGMV PV-0113, a rating scale based on relatively una mbiguous symptoms was devised. Materials and Methods Three-week old tobacco plants ( N. tabacum ; cv. Samsun; genotype nn ), were inoculated with TMGMV PV-0113, which is known to produce lethal disease in TSA. After symptoms (leaf pleating, mild green mosaic ) developed (2 to 4 weeks postinoculation; WPI), leaf tissue was harvested and used to inoculate Eryngium planum (Apiaceae). Symptoms of viral infection (yellow streaks and flecks) became apparent afte r several weeks, and newly emergent leaves were harvested. Eryngium tissue was triturated in 20 mM NaHPO4/Na2PO4 inoculation buffer (pH 7.2) and applied to the hypersensitive host, N. tabacum cv. Samsun ( NN ). LL appeared within 72 to 96 hours; select indi vidual lesions were harvested us ing a narrow-blade scalpel, with care being exercised to avoid including excess leaf tissue. Each piece of ex cised leaf tissue (1-2 mm2) was homogenized in 0.3 mL of 20 mM NaHPO4/Na2PO4 inoculation buffer (pH 7.2) in a 1.5 mL micro-centrifuge tube usi ng a Kontes micro-pestle. Additionally, 2-3-week-old tobacco plants (both nn and NN genotype) were dusted with carborund um and inoculated by pestle (e.g. simply rubbing the upper surface of the leaf with th e pestle.) After formation, LL were harvested by the aforementioned method and inoculated onto N. tabacum cv. Samsun ( nn ). Symptoms developed after a time, and apical tissue fr om these isolated plants was harvested. All isolates were tested simultaneously under cond itions as identical as possible. Threeto 4week-old TSA and 3to 4-week-old tobacco ( N. tabacum cv. Samsun; genotype nn ) plants were inoculated with the SLL isolates and main tained in a greenhouse, at temperatures between 16C and 28C. The day length was approximate ly10 hours, supplemented by artificial lighting (fluorescent bulbs) for a 15-hour to tal photoperiod. Plants in this study were fertilized weekly
79 with a 280 ppm solution of Peters 20-20-20 so luble plant food. The plants were observed carefully for potential differen ces in symptom development. To determine the subtype of TMGMV PV -0113, the differential indicator species N. benthamiana N. clevelandii and N. rustica were grown and inoculated as previously described in Chapter 2. Primers to the 3 NTR were designed to clone the regi on for analysis and further research. The forward primer: 5 GGGGACAAGTTTGTACAAAAAAGCAGGCTTCATGCCTTATACAATCAACTC 3 was designed to begin annealing to the template from the start of ORF 4, (nt 5666-5687), and had a Tm of 66.2C. The reverse primer: 5 GGGGACCACTTTGTACAAGAAAGCTGGGTTGGGCCGCTACCSGCGGTTA 3 was designed to anneal nt 6355, to nt 6336, with a Tm of 73.3C. This primer included the degenerate base, S to allow for G/C dimorphism in the 3 NTR occurring in some strains of TMGMV (Bodaghi et al., 2000). To determine whether or not STMV was pr esent in the TMGMV PV-0113 culture three cultures were propagated in tobacco and dsRNA ex tracts prepared from the infected tissue: TMGMV PV-586 (a culture of TMGMV-L c ontaining STMV), TMGMV PV-0113 (TMGMV PV0113, the culture being evaluated), and CMV (Cucumber mosaic vi rus, a tripartite nontobamovirus). CMV was propagated in N. glutinosa to prevent tobamovirus contamination, and the two TMGMV cultures were propagated in N. tabacum cv. Samsun ( nn ). Tissue harvested from apical regions 3 WPI was used in a modified Morris and Dodds dsRNA extraction procedure (Morris and Dodds, 1979), as follows:
80 Seven grams of each infected tissue sample was quick-frozen under liquid N2, triturated in a chilled sterile mortar, and added to beakers containing a cold (4C) mixture of 14 mL 2x STE buffer (pH 6.8), 20 mL equilibrated phenol, 2 mL 10% SDS, and 14 mg of bentonite. The mixture was held at 4 C while stirring for 30 mi nutes, after which the mixtures were centrifuged 15 minutes at 8,000 rpm in phenol-resistant, NaOHtreated sterile (Oakridge) centrifuge tubes. The aqueous phase was then collected, the volum e adjusted to 20 mL with 1x STE buffer (pH 6.8), and 3.8 mL 95% ethanol was added, along with 1.0 to 1.5 g Whatman CF-11 charged cellulose powder, under con tinued stirring at 4C. One hour later, the mixtures were poured into columns prepared from 12 mL disposable syringes and baked glass wool, using a solution of 15.0 to 16.5 % ethanol in 1x STE to wash the cellulose until colorless. Samples were then pressed dry using the sy ringe pistons and dsRNA eluted with 14 mL of 1x STE. To the eluent 1/10 vol. 3M NaAc (pH = 5.5) and 2.5 vol. 95% ethanol were added, and the samples incubated overnight at 20C. The samples were then centrifuged at 9,000 rpm (9,750 x g ) for 30 minutes at 4C to r ecover the precipitated RNA, which was washed once with 70% ethanol, dried under vacuum, and dissolved in 50 to 100 L of sterile, distilled, deionized water. This procedure was repeated using different c oncentrations of ethanol in the wash step until samples were obtained that produced relatively distinct dsRNA bands on non-denaturing agarose and polyacrylamide gels. To reduce background contamination, samples were then digested at 37C for 1 hour using 4 to 6 units of DNAase I per 100 g of nucleic acid extracted, after which samples were heated to 75 C for 10 min in order to inactivate the enzyme. The presence/absence of STMV was also assessed using immunodiffusion on non-SDS media, (containing 0.7% noble agar, 0.85% NaCl, 0.03% NaN3, and 5 mM Tris, pH 7.5). Sixty
81 (60) L of undiluted rabbit anti-STMV serum (kindly provided by Deborah M. Mathews; University of California, Riverside) was loaded into the central well of each system. Sixty (60) L of the antigens TMGMV PV-586 (ATCC accession deposited as TMGMV-L + STMV), TMGMV PV-0113, and sap from uninoculated tob acco (Hsap) were loaded into the surrounding wells. Results There was little variability in the disease symptoms produced by the SLL isolates in tobacco, with all isolates produc ing symptoms typical of TMGMV infection: leaf pleating, faint green mosaic, and flower-breaking. Likewise, all is olates retained the abi lity to infect and kill TSA, and the rate of disease progress was much the same for each isolate (data not shown). The timing of disease onset produced by each is olate inoculated onto TSA was consistent; all plants inoculated in the first and second re petitions developed symptoms 4 DPI. Symptom development in two isolates used in the third repetition was delayed 1 and 3 days, respectively. Progress of the disease in TSA produced by TMGM V PV-0113 was also rather consistent. The first observable symptom, epinasty, was not as unambiguous as other symptoms, being easily confused with wilting from natura l leaf movement in response to time of day, temperature stress, and wilting from other causes). However, data colle cted indicate that plan ts developing epinasty did so within 5 DPI. Two types of hypersensitive res ponse (HR) were observed: di screte LL, and patchy, more diffuse necrosis. Discrete LL often took the form of star-like spots, with a central lesion having minute necrotic veins extending outward from the focus. Usually beginning at 5 DPI, local lesions/necrotic patches were n early always present by 9 DPI, although sometimes plant collapse made observation difficult. Epinasty and LL/necr otic patches were near ly always immediately followed by permanent apical wilting, resulting in collapse of the plant. The expression of
82 wilting was modified by the age, size, and condi tion of the plant. Young, small, and rapidlygrowing plants (usually 3 weeks) would often completely lo se turgor and collapse, while in older, larger, and slowly-growing plants (usually > 3 weeks), th e wilting was confined to the young tissue around the apical regions. There was so me variability in the onset of this symptom between the sets of single local lesion (SLL) is olates; however, this st age of the disease had usually occurred by 12 DPI Perhaps the most variable symptom observed in the greenhouse was leaf abscission. This symptom was not always present and time of on set was variable, occu rring between 8 and 21 DPI. When it occurred, especially in more mature specimens, the leaf abscission symptom would often progress to comple te defoliation of the plant. Disease progress in TMGMV-infected greenhou se-grown TSA inevitably culminated in death. Despite the consistent mortality, some plants, though completely defoliated and mostly necrotized, retained living (green) stem tissu e for more than a month. Based on general observations of TMGMV-infected TSA, a scale (Table 3-1) was devised to measure the disease progress, and indexed to photograp hs (Figure 3-1) for comparison. Symptoms produced by TMGMV PV-0113 on the experimental indicator hosts N. benthamiana and N. clevelandii were generally consistent with the results reported for small-type TMGMV (TMGMV-S, U2) (Bodaghi et al., 2000), while symptoms produced by TMGMV PV-0113 on N. rustica were variable, but similar to results reported for large-type TMGMV (TMGMV-L, U5) (Bodaghi et al., 2000) Timing of death in the host N. benthamiana was variable, but always later than 10 da ys post-inoculation. Infection in N. rustica by the culture provided as U2 was not confined to LL, but became systemic.
83 Symptoms produced on N. clevelandii displayed characteris tics of both TMGMV-L and TMGMV-S infection as reported in literature (Bodaghi et al., 2000). Although the plants often developed systemic necrosis, this was preceded in some cases by leaf deformation and shoot proliferation, a symptom reported for TMGMV-L-infected N. clevelandii Symptoms of TMGMV PV-0113 on N. rustica varied between experimental se ts, but LL were always present. Reverse transcriptase-PCR using primers specific to the 3 NTR of TMGMV generated a product of different size dependi ng upon the culture analyzed (Fi gure 3-2). A culture obtained from CREC, Lake Alfred, as TMGMV U2 did not generate any product. A culture obtained from CREC, Lake Alfred, as TMGMV U5, and TMGMV PV-586, known to contain TMGMV-L, both produced products visibly larg er than that genera ted by TMGMV PV-0113. Relatively impure dsRNA was obtained from TM GMV-infected tobacco leaf tissue using the extraction method employed. C ontent of dsRNA relative to othe r nucleic acid materials was highly enriched by the procedure, but contam inants were not completely eliminated. Electrophoresis on agarose and polyacrylamide allowed visualization of subgenomic RNA bands. Digestion using DNA-ase I helped to reduce background created by contaminating host DNA molecules. Nucleic acid bands of a size cons istent with STMV could be visualized when dsRNA extract from TMGMV+STMV-infected ti ssue was analyzed on both polyacrylamide (Figure 3-3) and agarose (Figure 3-4) gels. Satellite tobacco mosaic virus dsRNA was present in quantity and easy to observe in TMGMV culture PV-586 (which is reported to contain it; Anonymous, 2007). It was not obs erved in any of the extracts from TMGMV PV-0113. Rabbit anti-STMV serum reacted (forming precipiti n) to the STMV-containing ATCC accession PV586, but not to TMGMV PV-0113 or healthy tobacco (Figure 3-5).
84 Discussion The disease phenotypes of isolates within each series appeared similar. All SLL isolates tested demonstrated the ability to kill TSA; symptoms on both tobacco and TSA appeared identical. Based on these observations it seems unlikely that TMGMV PV-0113 is a mixture of viruses, but is instead, a quasi-sp ecies, with all contributing genotype s retaining the ability to kill TSA. While data from the use of inoculated indica tor hosts was not conclusive, the results were usually consistent with symptoms reported in literature. Greenhouse-grown, uninoculated N. clevelandii plants have been observed with thickene d, waxy, and distorted leaves similar to the effect seen on some TMGMV-inoc ulated ones. Perhaps, also, s ubtle differences in the genotype of virus, host, or growing conditions used re sulted in these anomalies. Nevertheless, gel electrophoresis of RT-PCR pr oduct corresponding to the 3 end of the virus indicated that the smaller fragment produced by TMGMV PV-0113 corresponds with TMGMV-S and not the larger fragment produced by TMGMV-L. While all isolates used in the subtype hos t-range determination by Bodaghi, Yassi, and Dodds (2000) were subcultured using SLL isolation, nearly all of the cultur es initially contained STMV, which has been shown to modify disease phenotype in some hosts (Bodaghi et al., 2000). Since TMGMV PV-0113 does not contain STMV, it is possible that the reaction of these hosts to TMGMV-S and TMGMV-L cultures lacking STMV will be different from those reported in the study (Bodaghi, et al., 2000). The figures may be misleading, because the criteria for rating changed over time, upon realization th at plants could recover. At the time data was recorded, plants were rated dead when leaves were necrotic or the plant completely, or nearly completely defoliated, or after collapse had occurred.
85 Rabbit anti-STMV serum reacted (forming precipitin) to the STMV-containing ATCC accession PV-586, but not to TMGMV PV-0113 or h ealthy tobacco. These results, combined with the absence of visible STMV dsRNA in ex tracts, support the conc lusion that TMGMV PV0113 is a pure culture of TMGMV-S and does not contain STMV.
86 Table 3-1. Scale for rating disease progress in TMGMV-infected TSA. Mature Plant Immature Plant 0 Healthy no disease symptoms Healthy no disease symptoms 1 Local lesions and/or necrotic spots Local lesions and/or necrotic spots 2 Permanent apical wilting leaf abscission Permanent wilting 3 Complete defoliation green stem Collapse complete loss of turgor 4 Dying stem 50% necrotic Dying plant 50% necrotic 5 Dead stem 100% necrotic Dead plant 100% necrotic
87 A B C D E F
88 G H Figure 3-1. Disease progress in TSA plants inoc ulated with TMGMV PV-0113. A) 3 week old TSA immediately prior to inoculation. B) TSA developi ng local lesions (LL) and epinasty 5 DPI. C) Close-up of LL s een through abaxial surf ace of leaf, 5 DPI D) permanent wilting beginning at s hoot apices on TSA plant 7 DPI E) TSA plant 11 DPI F) green stem fully defo liated/leaves necrotic on TSA plant 12 DPI G) necrosis spreading along stem on TSA plant 28 DPI H) Dead (all visible tissue completely necrotic) TSA pl ant 37 DPI (Photographs, except for F and G, are of the same plant.)
89 Figure 3-2. Agarose gel analysis of RT-P CR products. 1 = blank, 2 = TMGMV-LAU5, 3 = TMGMV-PV-586 (U5+STMV), 4 = TMGMV-LAU2 (No. 1), 5 = TMGMV-LAU2 (No. 2), 6 = TMGMV PV-0113, 7 = 1 kB ladder, 8 = empty. 1 2345678 2, 000 b p 1,650 b p 3 fra g ment, TMGMV-L 3 fra g ment, TMGMV PV-0113
90 Figure 3-3. Comparison of ds RNA extracted from TMGMV (P V-0113) with that of TMGMV (PV-0586) demonstrating the absence of satellite TMV in isolate TMGMV (PV0113) resolved in a polyacrylamide ge l. Lane: 1 = CMV, 2 = blank (H2O + EtBr), 3 = TMGMV PV-0113, 4 = TMGMV-PV-586 (TMGMV-L + STMV). 1 2 3 4 STMV 3.3 kb 3.0 kb 6.4 kb genomic RNA I2 subgenomic RNA
91 Figure 3-4. Comparison of dsRNA extracted from CMV, TMGMV (PV-0113), and TMGMV(PV-586) demonstrating the absence of satellite TMV in isolate TMGMV (PV-0113) resolved in an agarose gel. 1 = 1 kB I nvitrogen plus ladder, 2 = CMV, 3 = TMGMV PV-0113, 4 = blank, 5 = TMGMV-PV-586. STMV 1 000 12 000 5 000 Genomic RNA Subgenomic I2 Sub g enomic I1 1 2 3 4 5
92 Figure 3-5. Imuunodiffusion analysis with rabb it anti-STMV serum showing no detectable reaction with TMGMV PV-0113 culture. 1,2 = ATCC PV0586; 3,4 = TMGMV PV0113; 5,6 = Healthy tobacco sap. A 1 2 3 4 5 6
93 CHAPTER 4 DISEASE PHENOTYPE AND COMPARISON OF PV-0113 WITH OTHER CULTURES Introduction Since the 1927 collection of TMGMV described as mild dark-green mosaic strain of TMV from Gran Canaria in the Canary Is lands, (McKinney, 1929), the species has been recovered from several locations including Australia, France, Germany, Israel, Italy, Japan, Taiwan, and the United States (Bodaghi et al., 2000; Cohen et al., 2001;. Conti and Marte 1983; Li and Chang, 2005; McKinney, 1952; Morishima et al., 2003; Parrella et al., 2006; Wetter, 1989). Within the United States, samples of TMGMV have been obtained from nurseries or outdoor sources in Florida, Geor gia, Ohio, South Carolina, Wisc onsin, and elsewhere (Baker and Zettler, 1988; Bodaghi et al., 2000; McKinney, 1952; Wetter, 1989; Zettler and Nagel, 1983). Zettler and Nagel (1983) sampled TMGMV (then cal led TMV U-2) from cultivated gesneriads obtained from growers in California, Connecticut, Florida, and Ohio. The strain isolated from an Ohio sample of Columnea was archived as DSMZ accession PV 0113. Reference cultures of TMGMV exist in the American Type Culture Collection (ATCC) and the Deutsche Sammlung von Mikroorganismen und Zellkulturen, the German Resource Center for Biological Material (DSMZ). Despite functional limitations on change for tobamovirus genes (Fraile et al., 1996.) enough genotypic variation exists in the virus to influence expression of disease phenotype in different hosts (Bodaghi et al., 2000.). For exam ple, some isolates of TMGMV cause lethal systemic hypersensitive response (LSHR) in Nicotiana benthamiana (Bodaghi et al., 2000), while an isolate from Taiwan identified as TMGMV caused systemic mosaic symptoms on this host (Li and Chang, 2005). As mentioned previous ly (chapters 1 and 2; Introduction) TMGMV
94 exists in both large type (TMG MV-L) and small type (TMGMV-S) varieties. Infections also sometimes contain Satellite tobacco mosaic virus (STMV). That different strains, or, at least, subtype s of TMGMV are capable of producing different symptoms on different hosts was established by Bodaghi et al. (2000.), who identified three Nicotiana species in which disease pr ogress and/or character varied enough that these species could serve as indicators capable of di stinguishing TMGMV-L from TMGMV-S. Nicotiana benthamiana expressed delayed symptom developmen t when inoculated with TMGMV-S, relative to TMGMV-L; N. clevelandii was killed by TMGMV-S, but not TMGMV-L, and N. rustica was systemically infected by TMGMVL but not TMGMV-S (B odaghi et al., 2000). Although overall yield of TMGMV-S from infected plan ts was higher (~2x) than from those infected with TMGMV-L (Bodaghi et al., 2000), experime nts have shown that TMGMV-L competes strongly with TMGMVS preventing it from becoming es tablished, suggesting that the duplication in the 3 pseudoknot-containing region may confer some competitive advantage, such as faster replication, ge ne expression, or movement (Bodaghi et al., 2004). It is known that the pseudoknot-c ontaining region of the 3 NTR binds cellular proteins, and seems to interact with the leader sequence to enhance tran slation (Leathers et al., 1993). Duplication of this region in TMGMV-L might improve competition for host resources. Based on observations that proximity to the 3 NTR affects levels of gene expression, it was discovered that transgene expression may be enhanced by placing a copy of the pseudoknot-containing region immediately downstream (Shivprasad et al., 1999). STMV does not modify visible symptoms in the hosts N. tabacum or N. glauca but in Capsicum annuum symptoms produced by the helper vi rus are modified, and in this host accumulation of the helper virus is severely re duced (Dodds, 1998). Furthermore, the genome of
95 STMV is itself variable, with at least 14 unique clones; one infectious isolat e survived despite a 71-nt deletion (Mat hews and Dodds, 1998). An example of the effect a satellite virus might have can be found in the systemic necrosis induced in tomato ( Lycopersicon esculentum ) by the negative sense (-) strand of the CMVassociated D satellite RNA (D-sat RNA). Differe nt satellite RNAs associated with CMV cause different effects within a partic ular host, and may add to or de tract from the virulence of the helper virus. Thus, while D-sat RNA induces a lethal systemic necrosis, WL1-sat RNA attenuates CMV disease symptoms, and B-sat R NA induces chlorosis. Furthermore, these modifications are host-dependent: the tomato-lethal D-sat RNA attenuates CMV symptoms in tobacco (Taliansky et al., 1998; Xu and Roosinck, 2000). When this study was undertaken, it was not known whether TMGMV PV-0113 is the only strain of TMGMV that kills TSA or whether all or some isolates ar e equally lethal. In order to determine this, and the possible differences in disease phenotype, the tw o cultures of TMGMV, TMGMV-S and TMGMV-L were compared on TSA and other species. Materials and Methods To characterize TMGMV disease on TSA, inoc ulation trials were conducted using plants propagated under various temperature and ligh ting conditions (greenhous e, outdoor, indoor, and growth chamber). The following plant specie s were tested in various experiments: Capsicum annuum Eryngium planum Nicotiana benthamiana N. clevelandii N. glutinosa N. rustica N. sylvestris and Solanum viarum (TSA). All were grown from s eeds and inoculated at 2-4-weeks post-emergence. These trials were repeated with other available is olates of TMGMV-S and TMGMV-L, including an ATCC accession, PV-5 86 a culture of TMGMV-L reported to contain STMV. In addition to TSA, indicato r species used by Bodaghi et al. (2000) were
96 included in some experiments, as well as C. annuum in the hope that one of the isolates might prove virulent to TSA, but not to the commercially important C. annuum Two cultures of TMGMV, one each of U2 (TMGMV-S) and U5 (TMGMV-L) were obtained from the Plant Pathology De partment, University of Florid a, Gainesville and the Citrus Research and Education Center (CREC) in Lake Alfred, Florida. The presence or absence of STMV in these cultures was uncertain. Plants of TSA, N. benthamiana and N. sylvestris were inoculated by Dr. Dennis J. Lewandowski in Lake Alfred, with in vitro transcripts of the chimeric tobamoviruses 30-B (Figure 4-1), and 30-B GFPc3, as well as sap from fluorescing tissue of 30-B GFPc3-infected plants (Shivprasad et al., 1999). Half of the trea ted plants were maintained at the CREC in Lake Alfred, while half were returned to Gainesville and maintained in a growth chamber for observation, both in visible light and under a long-wave UV lamp ( = 366nm). Growth chambers were set to maintain a maximum daytime temperature of 25C, a minimum nighttime temperature of 20C, and a photoperiod of 16 hours. Under these conditions, 3-week-old TSA seedlings were germinat ed in peat pellets and inoculated with U5 and U2 cultures of TMGMV obtained from the CREC, as well as the TMV/TMGMV chimera 30-B. Sets of 3-week-old TSA and N. sylvestris plants maintained under 12 hour lighting in an Apopka Room in Gainesville were inoculated with infectious sap containing the chimeric construct 30B as a follow-up to the e xperiments conducted in Lake Alfred. To evaluate the effect of STMV on TSA, 1-month-old TSA plants maintained in a quarantine greenhouse at the Florida Division of Plant Industry (DPI) in Gainesville were inoculated using sterile cotton gauze applicator pads saturated with reconstituted, dried tissue
97 obtained from the ATCC as PV-586 (the STMV-containing culture of TMGMV-L), and maintained in a cool, sealed quarantine greenhouse at the DPI in Gainesv ille, Florida, under ~14 hours of daylight and observed daily for sy mptom development. Control plants, mockinoculated with buffer only, and plants inoculated with infec tious tobacco sap containing PV0113, were maintained in a different greenhouse. Plants were fertilized weekly with 280 ppm Peters 20-20-20 fertilizer and observed daily for symptom development. The experimental setup at the DPI was then expanded to include the cultures obtained from Lake Alfred, PV-0113, and an assortment of experimental hosts. Due to limited greenhouse space, plants were started in trays of peat pots, and watered and fe rtilized using an ebb and flow method. Seeds of Capsicum annuum X Camelot, N. benthamiana N. clevelandii N. rustica and TSA were started in Ferry Morse brand Quick and Easy Peat Pellets and germinated in a cool greenhouse. At 3 weeks of age, wh en the seedlings were of a si ze suitable for inoculation, they were moved to the quarantine green house at the DPI in Gainesvill e where three leaves per plant were inoculated with virus tissue homogenized in ice-cold 20 mM Na2PO4/NaHPO4 inoculation buffer and applied using a sterile cotton swab. Pl ants were fertilized by drenching (soaking and then draining) with approx. Treatments consis ted of inoculation with PV-0113, LA U2, LA U5, and PV-586. Plants were also mock-inoculated with Na2PO4/NaHPO4 buffer only as a control. Plants were maintained with biweek ly drenching of 280 ppm solution of Peters 20-2020 fertilizer and observed daily for symptom development. Day length was decreasing over the course of this experiment. Cultures U2 and U5 obtained from the CRE C in Lake Alfred were inoculated onto Eryngium planum in an attempt to remove potential co ntaminants, and tissue extract from these
98 Eryngium plants was used to inoculate th e diagnostic indi cator plants Nicotiana sylvestris N. rustica and N. glutinosa Results Transcripts of 30-B GFPc3 produced no symptoms on TS A distinguishable from wounding. Infectious sap containing 30-B G FPc3 produced systemic symptoms on TSA, including GFP expression visible under UV light starting 4 days post-inoculation. Transcripts were infectious to Nicotiana benthamiana producing systemic symptoms, and, in the 30-B GFPc treatment, fluorescent patches app eared as soon as 3 days post-inoc ulation (DPI). Similar results on N. benthamiana were obtained with the infectious sap. Nicotiana sylvestris inoculated with the 30-B transcript developed LL, but were otherwise asymptomatic. Plants maintained at the CREC in Lake Alfred responded in a similar, if not identical manner as those in Gainesville (Dr. Dennis J. Lewandowski, formerly at CREC, La ke Alfred, personal communication). In subsequent experiments conducted under A popka Room conditions, TSA responded with systemic symptoms to inoculation with 30B, and N. sylvestris responded with numerous LL to inoculation with 30B. In the growth chamber, the large type isolate from Lake Alfred (LA-U5) produced symptoms similar to PV-0113 (epinasty, wilting, n ecrosis), while the U2 culture from Lake Alfred (LA-U2) gave mixed results, and no sympto ms were observed in the second repetition of this experiment. The TMV/TMGMV chimera, 30B, caused systemic symptoms on TSA without necrosis (Figure 4-2). Eryngium planum responded to inoculation with both small type (PV0113) and large type (LA-U5) isolates with yellow flecking and/or speckling. A variety of symptoms were produced on TSA by the STMV-containing culture TMGMV PV-586, including abscission, chlorosis, epinasty, mo saic, necrosis, veinal necrosis, and wilting.
99 Except for the veinal necrosis symptom, diseas e progress began normally, starting 6-8 DPI, and proceeded to about stage 3 or 4, with complete defoliation of the plant, after which there was new growth, mostly from the axial buds, which disp layed a distinct chloro sis/mosaic symptom. Results on the different ial indicator hosts N. clevelandii and N. rustica were generally consistent with those reported for TMGMV-L. Nicotiana benthamiana inoculated with PV-586 did not die before 18 DPI, although necrosis was obser ved as early as day nine (Table 4-1). Isolates PV-586 and PV-0113 produced epinasty on TSA. Leaf abscission was seen with the PV-0113 culture, but not w ith the STMV-containing isolat e. Leaf abscission was uncommon. Perhaps the earliest, most striki ng difference in symptom development on TSA plants inoculated with TMGMV-L-STMV was the appearance of a networ k of small necrotic threads or veins across the leaf. This symptom wa s concurrent with, but a ppeared different from, the development of LL. Generalized necrotic symptoms developed on both PV-0113and PV-586inoculated TSA plants at about the same time. No obvious diffe rences were observed at this point. Chlorosis developed in the new growth emerging from the PV-586 inoculated plants only. Wilting produced by both STMV (+) and STMV (-) cultures of TMGMV occurred simultaneously, and tended to affect the apical portions of the pl ant first. The mosaic symptom was observed on TMGMV-L-STMV inoculated plan ts only, and was distinct in appearance from the mosaic produced by TMV, ToMV, SHMV, or 30B. The plan t inoculated with PV-0113 eventually died. TSA plants inoculated with PV-586 initially a ppeared to be dying, but recovered with the aforementioned mosaic symptoms (Figure 4-3).
100 In addition to cultures of TMGMV, TSA pl ants were inoculated with TMV, ToMV, PMMoV, and SHMV (Figure 4-4). All of thes e tobamoviruses were able to infect TSA systemically, but none produced LL or necrosis on TSA. Plants maintained in the DPI quarantine greenhouse developed symptoms, on average, between 5 and 9 DPI. While PV-0113 reliably killed TSA in the DPI greenhouse environment, the other cultures of TMGMV did not. The ATCC culture PV-586 caused disease symptoms resembling those of PV-0113, but inoculated plants did not ultimately die. Epinasty was seen in several inoculated plants in all treatment gr oups in the DPI greenhouse, but not in the mockinoculated, buffer-only control gr oup. LL and other necrotic symp toms developed reliably only on the PV-0113 inoculated plants. Some LL and necrosis were seen in the PV-586 and on the TMGMV-L culture from CREC, Lake Alfred treatment groups, but not on the TMGMV-S culture from Lake Alfred, or the negative control group. Permanent wilting was observed only in those TSA plants inoculated with PV-0113, and in some plants inoculated with PV-586. The othe r treatments did not produce this symptom on TSA. Mosaic symptoms were eventually obser ved in plants inoculat ed with cultures of TMGMV except PV-0113. While this was expect ed in the PV-586 inoculated plants, the occurrence of mosaic in th e other groups was a cause for concern, possibly indicating contamination with another virus. Ultimately, al l plants inoculated with PV-0113 died, but not plants inoculated with other vi ruses, except some plants inocul ated with PV-586. Other plants survived in either symptomless or sympto matic condition. The presence of other unusual symptoms such as mosaic, leaf deformation, ch lorosis, etc. was limited to treatment groups containing cultures other than PV-0113.
101 Tissue from E. planum inoculated with U2 obtained from the CREC, Lake Alfred produced no immediate symptoms on N. sylvestris N. rustica or N. glutinosa plants. Inoculated N. sylvestris plants eventually developed systemic symptoms. Tissue from E. planum previously inoculated with U5 obt ained from the CREC, Lake Alfred caused LL on N. sylvestris N. rustica and N. glutinosa Nicotiana rustica plants inoculated with this culture went on to develop a more severe necrosis and other syst emic symptoms, consistent with the results reported for TMGMV-L. Discussion In trials conducted in a grow th chamber, isolates of TMGMV from CREC, Lake Alfred identified as U5 produced local lesions (LL) in the majority of the inoculated TSA. The duplicated sequences in 3 region of this subtype, which disti nguishes this type from the U2 type, do not apparently prevent the HR. As r ecovery was unanticipated, these plants were discarded before the disease proc ess had run its course. Failur e of the Lake Alfred culture provided as U2 to produce consistent sympto ms may be due to contamination, inadequate inoculation, or nonconductive plant growth conditi ons (i.e. peat pots, limited space, lack of fertilizer, etc). Large-type TMGMV that appeared to kill immature TSA plants under growth chamber conditions were discarded soon after symptoms (collapse/wilting/defoliation) occurred because previously, necrotic disease ha d invariably proved lethal. It remains uncertain whether the duplicated pseudoknot region or the presence of STMV is responsib le for survival/recovery of these inoculated plants. One possible experiment would be one in which the response of TSA to isolates of TMGMV-S/TMGMV-L not containing STMV were compared with response to isolates that do contain STMV.
102 None of the plants inoculated with 30B viri ons or 30B transcripts developed LL, but most developed systemic mosaic instead. Since bot h TMGMV-L and 30B contain the open reading frame ORF 4 expressing TMGMVL coat protein (CP), it stands to reason that TMGMV-L CP accumulates in both the LL and non-LL displaying pl ants, suggesting that some elicitor other than CP was responsible for the HR component of the lethal response seen in the TSA. Nicotiana sylvestris inoculated with transcri pts developed very few (0-2 ) local lesions per leaf, suggesting that in vitro transc ripts are not very infectious compared with intact virus. The recovery of plants infected with PV-586 was unexpected, which underscores the importance of preserving inoculated plants until necrosis has progressed to the full extent. While the CREC, Lake Alfred isolate U2 produced sy mptoms in tobacco, indicating that virus is present, most TSA plants inoculated with U 2 remained symptomless, suggesting possible contamination or replacement by some other vi rus. The indicator plant data, along with difficulties encountered in PCR-amplifying RNA ex tract from this culture, further raise the possibility that this culture is not TMGMV. Systemic symptoms observed on N. sylvestris suggest replacement by TMV type-strain. Other possibi lities are that it is a genetically distinct strain of TMGMV, or that it contains STMV or some satellite RNA that modifies its effects on the indicator hosts. Early symptoms observed in TSA inoculated with PV-586 such as LL, wilting, necrosis followed by partial recovery, introduce the possibility that the TSA-killing effect is independent of HR, or at least incompletely dependent upon it. If STMV is responsible for the survival and recovery of the plants, the mech anism of this action is unclear. Ultimately, necrosis progressed to death of all plants inoculated with TMGMV PV-0113, but not in plants inoculated with other isol ates of TMGMV or the TMV/TMGMV chimera 30B,
103 with the exception of a few plan ts inoculated with PV-586. Th ese results were unexpected; at this stage of the research, however, a rating scale had not yet been devised. In the first trial, it was not known that defoliated TSA could recover, so plants may have been incorrectly recorded as dead and discarded. It is also possible that the mortalit y in the PV-586 group may be the result of using immature plants. Perhaps immatu re plants die before the STMV has a chance to counteract the effect. The abse nce of symptoms in the mock-i noculated, control group suggests that cross-contamination, if it had ha ppened, occurred prior to inoculation. Peat pots are not the best medium for growing plants for use w ith this pathosystem. They were used in this case, because limited space was available in the greenhouse and the growth chamber, and because such a method is often us ed with success for other applications. In particular, N. benthamiana is not suited for use in peat pots. Rate of growth has a big influence on diseas e phenotype. Even when available space is limited, it is better to give experimental plants plenty of room to grow and expand during the research period, than it is to allow plants to become pot-bound or cease active development, even if this means working with a reduced number of replicates per trial. Epinasty appears to develop frequently when TSA plants are inoculated with tobamoviruses in general (my observations). Th e involvement of epinasty in the TSA-killing response may be coincidental. Although some variability in host response was observed, the culture labeled U5 obtained from the CREC produced effects consis tent with those reported in literature for TMGMV-L. The hosts Eryngium planum and N. rustica provide a method for identifying and purifying TMGMV-L. One future experiment should be to determine if STMV can be eliminated from an STMV-containing culture using E. planum or some other biological sieve.
104 Cultures of TMGMV-S and TMGMV-L used in subsequent research should come from genotypically and/or geographi cally defined populations, ( not PV-0113), and of known STMV (+)/ (-) status.
105 Table 4-1. Symptoms on vari ous differential/indicator hos ts by TMGMV PV-0113 (TMGMV-S) and TMGMV PV-586 (TMGMV-L + STMV). Host TMGMV PV-0113 TMGMV PV-586 Nicotiana benthamiana Death Death Nicotiana clevelandii Chlorosis, Mosaic, Shoot prol iferation Chlorosis, Mosaic Nicotiana glutinosa Local lesions Local lesions Nicotiana rustica Local lesions, Death Local lesions Nicotiana tabacum ( nn ) Mosaic Mosaic
106 Figure 4-1. Genetic map of TMV/TMGMV chimera 30 B.
107 Figure 4-2. Typical symptoms produced in TSA seedling by 30-B TMV/TMGMV chimera.
108 Figure 4-3. TMGMV PV-586 i noculated TSA recovering w ith mosaic after dieback.
109 A. B C D Figure 4-4. Photographs of sy mptoms by various tobamoviruse s on TSA. A) TMV, B) ToMV, C) SHMV, D) PMMoV.
110 CHAPTER 5 CROSS-PROTECTION EFFECTS Introduction McKinney (1929) observed that the symptoms of Tobacco mosaic virus (TMV) infection could often be moderated by pre-infection with a le ss virulent strain. Sin ce that time, researchers have tried to ameliorate the e ffects of tobamovirus infection by inoculating susceptible crops with mild strains of the virus (Rast, 1979). Cross-protection has been used as a strategy for managing tobamovirus and non-tobamovirus diseases that reduce yield in tomato, pepper, and citrus crops (Yoon, et al., 2006). Conjecture about cross-protec tion can be segregated into hypotheses that (1) attribute cross-protection to direct competition for host resources, such as replication sites or translation machinery; (2) propose inhibition of one virus by the coat protein (CP) of the other, perhaps disr upting the timing of the infecti on cycle; and (3) relate crossprotection to nucleic acid interf erence, perhaps through genetic r ecombination or activation of gene silencing. These hypotheses are not mutually exclusive: indeed, more than one mechanism may be involved in cross-pr otection (Aguilar et al., 2000). In studies of cross-protection between muta nts of Pepper mild mottle virus (PMMoV), it was observed that cross-protection resulted fr om infection by some mutants, despite low accumulation of virus in plantae (Yoon et al., 2006). The author s speculated, host response to cross-protection may prevail throughout the plan t without propagation of viral components (Yoon et al., 2006). Aguilars studies of cross protection between TMGMV and Oilseed rape mosaic virus (ORMV) in Nicotiana tabacum and Arabidopsis thaliana provide evidence for host involvement in cross-protection and support for th e idea that different mechanisms operate in different situations. When used as the prim ary inoculum, both TMGMV and ORMV appeared able to prevent systemic infection by the challeng ing virus. Results in tobacco seemed consistent
111 with a direct competition hypothesis, but interpreting the results in Arabidopsis proved more difficult. Interestingly, TMGMV prevented sy stemic spread and accumulation of ORMV in A. thaliana even though TMGMV symptoms were latent, and the virus very slow to move and accumulate in this host (Aguilar et al., 2000). Th is may suggest gene silencing or some other active process within the host. Tropical soda apple (TSA) seeds collected from fields in the area around Immokalee (Hendry Co.), Florida (2000), developed into pl ants displaying mosaic-type symptoms when raised in a greenhouse. Plants infected with th e contaminant virus failed to die when challenged by inoculation with TMGMV. Subsequent inves tigation proved the contaminating virus to be TMV. The objective of this study was to exam ine the effects of pot ential cross-protection against TMGMV in TSA by an unknown virus, a tobamovirus and a non-tobamovirus. Materials and Methods Seeds collected from TSA fruit harvested in the Immokalee area in Hendry Co., Florida in 2000, were planted in Metromix potting medium, and allowed to germinate, and develop in a greenhouse. A portion of the collected seed was surface-sterilized prior to planting, by soaking the seeds for 30-minutes in 1 M HCl, followed by rinsing and a 30-minute soak in a concentrated trisodium phosphate (TSP) solution (400g TSP/ 3.78 L H2O). The seeds were then rinsed and dried. To verify that the inf ection was seed-borne, and to dete rmine if the virus was carried internally, or limited to the seed coat, several treated seeds were planted at the same time as the untreated seeds; the rest were sa ved for use in future studies. Tissue from TSA symptomatic of a virus c ontaminant, designated C-1, was used to inoculate healthy TSA plants. When these pl ants developed mosaic symptoms, they were challenged by inoculation with TMGMV PV-0113. Experimental trials consisted of nine systemically infected plants inoculated w ith TMGMV PV-0113 and one positive control plant
112 inoculated with TMGMV PV-0113 only and one negative control plant inoculated with buffer only. The identical experiment was conducted excep t only eight plants were used per treatment and C-1 and TMGMV PV-0113 were inoculated on these plants simultaneously. The plants were maintained and observed under greenhouse or screen-house (outdoor) conditions. The differential indicator, ( Phaseolus vulgaris cv. Pinto), which devel ops local lesions (LL) in response to inoculation with TMV, but not TMGMV, was also inoculated with C-1. One-month-old TSA plants were inoculated with TMV and observed until symptoms developed uniformly on all inoculated plants (approx 3 weeks). Plan ts were then inoculated with infected plant tissue triturated in 20 mM Na2PO4/ NaHPO4 buffer, pH 7.2. Three leaves per plant were inoculated manually, by rubbing carborun dum-dusted leaves with inoculum-soaked cheesecloth pads. Plants were inoculated c oncurrently and sequentially with TMGMV PV-0113 and TMV. TSA plants were also mock-inoculated with buffer as a negative control, TMV only, and TMGMV PV-0113 only. Plants were maintained in a greenhouse, fertilized weekly with 280 ppm Peters 20-20-20 soluble fertilizer, a nd observed daily for symptom development. Examining the possibility that a non-tobam ovirus might also cross-protect TSA from TMGMV PV-0113 induced lethality, 3-week-old TSA plants were inoculated with Cucumber mosaic virus (CMV). Once symptoms develope d uniformly on all inoculated plants (approx 5 weeks), the plants were then challenge inoculat ed with PV-0113-infected plant tissue triturated in 20 mM Na2PO4/ NaHPO4 buffer, pH 7.2 (except for a negative control set, which was mockinoculated, using buffer only). Three leaves per plant were i noculated manually, by rubbing the carborundum-dusted leaves with chees ecloth pads. Treatments consis ted of inoculation with sap extract from a common strain of CMV obtained cour tesy of Dr. Carlye Baker (Division of Plant Industry, DPI) in Gainesville, Florida and propagated in Nicotiana glutinosa Plants were mock-
113 inoculated with buffer as a nega tive control. Plants were ma intained in a cool greenhouse, fertilized weekly with 280 ppm Peters 2020-20 soluble fertilizer, and observed daily for symptom development. Results Symptomatic tissue harvested from TSA and t obacco infected with C-1 produced results consistent with TMV infection on the differential indicator ( Phaseolus vulgaris cv. Pinto) and other differential host-range plants. Immunodi fffusion, ELISA, and RT-PCR also identified C-1 as TMV-type strain (data not shown). No symp tomatic or infected pl ants were found among the seedlings grown from HCl/TSP-treated seeds. TS A plants inoculated with C-1 became chlorotic and developed epinasty when challenged with TM GMV, but did not die. Instead, a distinct mosaic pattern continued to appear on the ne wly emergent leaves, al ong with chlorosis that followed the leaf veins to create a branching ap pearance. In most trials, TSA symptomatically infected with type-strain TMV did not die when challenge-inocula ted with TMGMV PV-0113, except in one study where the treated plants die d, after lagging behind the positive control plant in symptom development by about 72 hours. TSA plants inoculated with CMV developed symptoms including vein-clearing and mosaic with frequent chlorotic islands. All TSA pl ants inoculated with CMV and subsequently challenged with TMGMV PV-0113 developed local lesions, wilting, and necrosis, followed by death (Table 5-1). Discussion Seed transmission of tobamoviruses, when it occurs (e.g. PMMoV in Capsicum sp.) is believed to be due to the presence or persistence of virus on the seed coat, from which it infects the seedling (auto-inoculation as it emerges during germination). Multiple diagnostic techniques
114 (host-range, ELISA, RT-PCR) confirmed the contam inating virus associated with the Immokalee seeds to be Tobacco mosaic virus (TMV). TMV appears to cross-protect against TMGM V, while the non-tobamovirus CMV does not appear to cross-protect. Studies conducted with Tomato mosaic virus (ToMV), and TSA mosaic virus (TSAMV) indicate that these viruses also cross-protect TSA from TMGMV PV-0113 lethality (Mark S. Elliott, Se nior biological scientist, Univ ersity of Florida, personal communication). Although the exact reason s for cross-protection remain unknown, the relationship between the viruses that do crossprotect, sharing similar genomic nucleotide and CP amino acid sequences, seem consistent with both CP interference and gene silencing hypotheses. Since TMV appears to cross-protect against TMGMV-induced lethality, additional studies of cross protection should be conducted under cont rolled temperature and lighting conditions. The anomalous decline and death of TSA plan ts previously inoculated with TMV after inoculation with TMGMV in one study is interes ting, and consistent with the multiple mode of action hypothesis.
115 Table 5-1. Crossprotection study Inoculum Symptoms Buffer only NS CMV E, M, VC TMV M TMGMV (PV-0113) E, LL, W, N, D TMGMV (PV-0113) + CMV E, LL, W, N, D/M, VC, E, LL, W, N, D TMGMV (PV-0113) + TMV E, Chl/M CMV = Cucumber mosaic virus, TMGMV = Tobacco mild green mosaic virus, TMV = Tobacco mosaic virus. Symptoms: Chl = chlorosis, D = death, E = epinasty, LL = local lesions, M = mosaic, N = necrosis, VC = vein clearing, W = wilting. For challenge-inoculations, data presented to the left of the slash is the result of concurrent co-inoc ulation; data to the right of the slash is the result of sequential inoculation.
116 CHAPTER 6 TMGMV PV-0113: PURIFICATION, LY OPHILIZATION, AND RNA EXTRACTION Introduction Although sap extract from plants infected wi th the TMGMV, PV-0113 isolate is reliably lethal to tropical soda apple, for in-depth characteriz ation, the virus should be purified. A nonenveloped virus with the stability of TM GMV may be extracted, purified, freeze-dried (lyophilized) even crystallized while retaining pathogenicity (Yordanova, et al., 2002). Although tobamoviruses in herbarium specimens re main infectious for several decades, (Gibbs, 1999; Hull, 2002), possibly centuries, extracted viru s will begin to degrade over time, even if kept in a purified and lyophilized state. Various protectants such as so rbitol and dext ran may be used to extend the life of tobamovirus preparations (Yordanova, et al., 2002). For molecular characterization of TMGMV, geno mic RNA must be extracted. As part of the replication cycle, the fragile genomic RNA of (+)ssRNA viruses forms stable doublestranded RNA structures (dsRNAs). These more stable dsRNAs are generally indicative of infection, although dsRNAs of an unknown origin (cryptic RNA) are also known to occur in plants (Valverde, 1990). Various methods exis t for viral RNA extraction from plant tissue, although it is generally easier to extract total RNA than dsRNA onl y. One approach to isolating dsRNA utilizes selective binding of nucleic acid to charged cellulose followed by washing and elution at different ethanol concentrations (Morris and D odds, 1979). Total RNA may be adsorbed to cellulose in a column at a particular ethanol concen tration, the column washed with ethanol, and then the different ssRNA types released as the etha nol concentration is lowered. The dsRNA may be eluted (Morris and Dodds, 1979). The concentration of ethanol is critical, and must be adapted to each individual batc h of cellulose (Dr. Ja ne Polston, Professor, University of Florida, Plant Pathology Depa rtment, Gainesville, pe rsonal communication).
117 Materials and Methods To purify TMGMV PV-0113 for use in furt her research, two methodologies were followed: a typical preparation (Virus Prep-D ), and a simplified, ( quickie) purification (TMGMV prep-Q). Healthy, 3-week-old tobacco plants ( Nicotiana tabacum cv. Samsun, genotype nn ) were inoculated manually with a 1:10 suspension of TMGMV PV-0113 in 20 mM Na2PO4/ NaHPO4 buffer (pH 7.4) and maintained in a gree nhouse. The plants were fertilized weekly with 800 ppm Peters 20-20-20 soluble fert ilizer. By 3 weeks postinoculation, systemic symptoms (green mosaic, leaf pleating) had de veloped, and leaves of various ages from the plants were harvested. One hundred grams (100 g) of symptomatic leaf tissue was homogenized in a Waring blender at 4 C with 200 mL cold (4C) 200 mM K2PO4/ KHPO4 buffer (pH 7.5) and 0.125 M Na2SO3 (15.76 g). One hundred milliliters (100 mL) of cold chloroform (CHCl3) was then added, and the mixture further homoge nized under a fume hood for 2-3 minutes. The homogenate was centrifuged at 2500 rpm (1,000 x g ) in a Sorvall centrifuge for 10 minutes. The aqueous supernatant was filtered through moist cheesecloth and re-centrifuged at 12,500 rpm (18,000 x g ) for 10 minutes. The clarified supern atant was again f iltered though moist cheesecloth and to this was added 8% w/v pol yethylene glycol (PEG), 0.5 M NaCl, and 1% Triton X-100. The liquid was stirred for 1 hour at 4 C and then centrifuged (under refrigeration) at 10,000 rpm (12,000 x g ) for 10 minutes. The supernatant was discarded and the pellet resuspended in 50 mL of phosphate buffer. The sample was again centrifuged at 10,000 rpm (12,000 x g ) for 10 minutes; this time the supernatant wa s retained and combined with 8% (w/v) PEG and 500 mM NaCl. The soluti on was stirred for 1 hour at 4 C and centrifuged at 10,000 rpm (12,000 x g ) for 10 minutes. The white pellet wa s re-suspended in 8.0 mL of 20 mM HEPES buffer (pH 7.5) to yield a virus suspension that was faintly bluish, visibly opalescent, and
118 birefringent when viewed across polarized light. Purification was halted at this step. For the quickie purification, the step of filtration through cheesecloth and the chloroform treatment was omitted. Purified virus adjusted to 1 optical density (1 OD) at A260 by dilution in Na2PO4/ NaHPO4 inoculation buffer was aliquotted into 500-L un its before drying under vacuum using a DuraDry condenser module (manufactured by FTS Syst ems, Inc.). Some lyophilized virus was resuspended in sterile, distilled, de-ionized wate r and used to inoculate plants. The rest was stored on the shelf at room temperature. Viral preparations D and Q were evaluated using a Bio-Rad SmartSpec 3000 spectrophotometer and a ssayed for infectivity by inoculation onto the local lesion host, N. tabacum cv. Samsun ( NN ) and the systemic host, TSA. Infectivity was determined by counting the number of LL produced by different preparations in comparison with a freshly purified virus preparation. Total RNA for RT-PCR from newly emergent leav es was prepared using the Trizol reagent (Invitrogen). Leaf disks of sy mptomatic tissue of tobacco ( N. tabacum cv. Samsun nn ) inoculated 3 weeks prior were removed usi ng the lid of the micr ocentrifuge tube and homogenized with 200 L Trizol with a DEPC or NaOH treated Kontes micro-pestle. An additional 550 L of Trizol was added to th e homogenate which was then incubated for 5 minutes at room temperature before adding 200 L of chloroform. The mixture was held at room temperature for an additional 8-10 mi nutes, after which it was centrifuged at 12,000 x g for 10 minutes at 4C. The aqueous (top) phase wa s transferred to a new microcentrifuge tube, 0.375 mL of isopropanol was added to precipitate the RNA, the c ontents mixed by inversion, and held at room temperature for 10 minutes before another centrifugation at 12,000 x g for 10 minutes at 4C. The supernatant was then carefully removed and the RNA pellet washed with 1
119 mL of 75% RN-ase free ethanol. The supern atant was removed and a 15-second pulse spin, followed by a 6-minute spin-down was applied to remove any remaining supernatant before allowing samples to dry under air-flow (hood) for 10 minutes. Prepared RNA was dissolved in 30 L molecular biology grade (MBG) water. Re suspended samples were incubated for 10 minutes at 55-60C before us e or storage at -80C. Total RNA was extracted from 100 L purified virus by addition of 100 L RNAdissociating solution containing 20 mM Tris, 2% SDS, and 2 mM Na2EDTA (pH 9.0), along with 10 L protease K (Invitrogen). Af ter incubation for 10 minutes at room temperature, the solution was combined with 400 L ice-cold solvent mixture (200 L adjusted phenol, 200 L 24:1 chloroform-isoamyl alcohol) and agitated gently for 2 minutes This virus-solvent mixture was centrifuged at 14,000 rpm (16,000 x g ) for 5 minutes and the aqueous layer recovered. Three (3) volumes of 100% ethanol and 0.1 volumes (20 L) of 3M NaAc (pH 5.3) were added to the aqueous layer and held overnig ht at -80C. The mixture was then centrifuged 14,000 rpm (16,000 x g ) for 30 minutes and the supernatant discar ded. Precipitated RNA was then washed with 500 L 70% EtOH and re-precipitated at 14,000 rpm (16,000 x g ) for 5 minutes. The supernatant was carefully removed and the R NA dried under a low vacuum (aspirator) and resuspended in 50 L MBG H2O. Unused preparation was stored immediately at -80C for later use. Cultures of Tobacco mild green mosaic virus (TMGMV PV-0113, PV-586, LA-U5) were propagated in Nicotiana tabacum cv. Samsun ( nn ) as described above. Cucumber mosaic virus was propagated in N. glutinosa Tissues from both plants were harvested 3 weeks postinoculation, quick-frozen in liquid N2, and stored at -80C until use.
120 Seven grams of each infected tissue sample were frozen under liquid N2 and triturated in a sterile mortar and added to beakers containing a cold (4C) mixture of 14 mL 2X STE buffer (pH 6.8), 20 mL equilibrated phenol, 2 mL 10% SDS, and 14 mg of bent onite. The mixture was held at 4C while stirring for 30 minut es, after which the mixtures were centrifuged 15 minutes at 8,000 rpm in phenol-resistant, NaOH-treated sterile (Oakridge) centrifuge tubes. The aqueous phase was collected, the volume adjusted to 20 mL with 1X STE buffer (pH 6.8), and 3.8 mL 95% ethanol added, along with 1.0 to 1.5 g Whatman CF-11 ch arged cellulose powder, under continuous stirring at 4C. One hour later, the mixtures were pour ed into columns prepared from 12 mL disposable syringes and baked glass wool, using a soluti on of 15.0 to 16.5% ethanol in 1X STE to wash the cellulose until it became colorl ess. Samples were then pressed dry using the syringe pistons and the dsRNA eluted with 14 mL of 1X STE. To the eluent, 3M sodium acetate (pH 5.5; 0.1% v/v) and 95% ethanol (2.5 % v/ v) were added, and the samples incubated overnight at -20C. The samples we re then centrifuged at 9,000 rpm (9,750 x g ) for 30 minutes at 4C to recover the precipitated RNA, whic h was washed once with 70% ethanol, dried under vacuum, and dissolved in 50 to 100 L of sterile, distilled, deionized H2O. This procedure was repeated using different concentr ations of ethanol in the wash step until samples were obtained that produced relatively distinct dsRNA bands when separated on agarose and polyacrylamide gels. All nondisposable autocl avable materials used in extr action were made DN-ase and RNase free either by baking (glass) at 200C for 8 h or soaking (pla stic) in 100 mM NaOH for 1 h, then rinsing with RN-ase free H2O, and autoclaving to inactivate the enzymes. Results The A260/280 was 2.85/2.40 (= 1.19) for preparation D and 2.79/2.36 (= 1.18) for preparation Q; the reported A260/A280 ratio for TMGMV is 1.22 (Wetter, 1986). Significant yield of virus for both preparations was approxi mately 0.9 mg per gram of fresh leaf tissue
121 processed or 6.9 mg per gram of dry tissue, based on the extinction coefficient for TMGMV reported in literature of 3.16 at A260. Fresh tobacco tissue is 87% water (Mark S. Elliott, University of Florida, Plant Pathology Depart ment, Gainesville, unpublished). Preparation D was slightly bluish, while preparation Q was yellowish or cream-colored and milky. Preparation Q produced many more local lesions (LL) on Samsun NN tobacco at a high concentration (~320 g/mL), but at lower concentrations ( 0.3 g/mL) it gave approximately the same number of LL as prepar ation. D (Figure 6-1). Freshl y prepared virus had a higher biological activity than either reconstituted lyop hilized virus or a suspension in inoculation buffer stored in the refrigerator for 14 months. Virus reconstituted from a lyophilized state wa s sometimes transparent or turbid and/or had aggregates. A transparent reconstituted sample of preparation D had the A260/280 value of 2.87/2.34, for a ratio of 1.23, which is very close to the reported value for pure TMGMV (Wetter, 1986). Other virus preparations had A260/280 ratios of 1.39, 1.37, and 1.31. TSA plants inoculated with PV-0113 purifie d and stored for 14 months at 4C developed symptoms and succumbed to infection. Discussion Both types of virus purification methods produc ed similar yields of TMGMV per gram of leaf tissue. Of the two prepara tions, visually preparation D was the most pure. However, the absorbance values (2.85/2.40) and the rati o (A260/280 = 1.19) for preparation D and preparation Q (2.79/2.36; A260/ 280 = 1.18) indicate a yield of 0.9 mg virus per g of leaf tissue, which is relatively low for a tobamoviru s. The values (2.87/2.34) indicate a successful resuspension of the virus to the former concentr ation, so effects from d ilution or concentration are probably not a factor. Inte restingly, the less-purified qui ckie preparation produced more LL than prep D on Samsun NN tobacco at 1:10 dilution from 1 OD in buffer.
122 Virus stored refrigerated in Na2PO4/ NaHPO4 inoculation buffer remained infective even after 1 year (data not shown). Also surprisingly, the lyophiliza tion procedure used appears to have reduced the viability of the virus inocul um. Perhaps the virus used in any future lyophilization should be suspended in sterile distilled water instead of buffer solution to preclude possible denaturation by the buffer salts as th e buffer concentrates. The reasons for the cloudiness/turbidity observed in some lyophiliz ed samples upon reconstitution remain unknown. The reconstituted preparations proved inf ective despite their abnormal appearance. The ribonucleic acid (RNA) obtaine d from leaf tissue using Triz ol was as useful as RNA prepared from purified virus for producing c DNA for use in RT-PCR. The appearance of dsRNA extracted using the modified Morris and Dodds protocol (Morris and Dodds, 1979), following electrophoresis on agarose and polyacryl amide gels, was consistent with patterns observed from TMGMV extracts r ecorded in the literature. An ethanol concentration of 15.5% was critical to successful dsRNA extracti on using the available charged cellulose. 0.00 50.00 100.00 150.00 200.00 250.00 300.00 350.00 1:101:1,0001:10,0001:50,0001:100,000 DilutionAverage Local Lesions Textbook Preparation "Quickie" Preparation Figure 6-1. Comparison of the activity of two preparations of TMGMV PV-0113.
123 CHAPTER 7 DILUTION ENDPOINT AND BIOLOGICAL ACTIVITY OF PV-0113 Introduction It is necessary to establish an optimum inoc ulum concentration of TMGMV that will elicit the lethal response in tropical s oda apple (TSA). The point at wh ich a serial dilution by a virus fails to elicit a particular bi ological activity, which in the case of the TMGMV-TSA system is local lesions and plant death, is called the dilution endpoint. To determine the dilution endpoint it is necessary to start with a known amount or concentrati on of the virus. To quantify a virus present in a viral prep aration, several physical, chemical serological, and biological methods could be used. Physical methods incl ude counting individual vi rions under an electron microscope or weighing dried virus preparations Recently, a physical technique involving mass spectrometry of intact virus pa rticles has become available (F uerstenau et al., 2001). These techniques are often highly laborious and time consuming, limiting their application beyond a modest scale. Chemical microa nalysis of viral components, such as nitrogen, may be used, but suffers from the same drawback. Serological methods, such as antigen-antibody precipitation, are relatively sensitive and have been used, but require pure virus prep arations to generate antiserum. Also, animal care requirements make production of polyclonal antiserum laborious and/or expensive (Ball, 1964). Bi ological assays may be qualitativ e as well as quantitative, and because they depend upon the intrinsic property of vi ruses to infect and replicate, can provide the most sensitive means available for detecting viruse s. Mere infection of a host plant, however, tells nothing more than that there was sufficient virus present to establish infection; inoculation with that amount or a hundred thousand tim es more might have the same result. In 1929 F.O. Holmes discovered that Tobacco mosaic virus (TMV) produces countable necrotic lesions when inoculated onto Nicotiana glutinosa plants, and that the number of lesions
124 depended on the virus content of the inoculum. This discover y allowed plant viruses to be assayed by counting the number of local lesions (LL) in a manne r similar to Kochs method of counting bacterial colonies on ag ar plates. However, the LL a ssay, like all infectivity tests cannot provide an absolute measure of the amount of virus, but may, if carefully planned and conducted, indicate the relative amounts of virus contained in inocula co mpared simultaneously, in the same study (Bawden, 1956). LL assays suffe r from the high number of difficult-to-control variables. Different LL-producing hosts could vary in their sensitivity to inoculum from species to species, cultivar to cultivar, even from plan t to plant. Adjacent leaves on a host plant could vary more than different halves of the same leaf. The age of the plant, as well as the age of the leaf on the plant influence the number of LL produced (Bawden, 1956). Various environmental conditions have been shown to influence the forma tion of LL on a host plant. Holding plants in darkness or low light intensity for 24 hours prior to inoculation tends to increase the number of LL that develop. Variation in temperature ma y affect LL development upon a host, although the number of LL produced may increase or decreas e relative to the variation in temperature depending on the specific plant-pathogen inte raction involved. Ot her factors affecting susceptibility include the availabil ity of water and nutrients. Sin ce any inhibition of plant growth will affect LL development, any substances added during inoculation (Bawden, 1956), such as phosphate-based inoculation buffer and particles of carborundum or diatomaceous earth could influence LL formation. Phosphate and abrasive s will increase the number of LL that develop, (Roberts, 1964), while any ribonuclease presen t will inhibit LL formation (Bawden, 1956). Bentonite, which interferes with ribonuclease, will in crease the number of LL produced by naked TMV RNA, but not by intact viri ons (Bawden, 1956). Even the type of mechanical inoculation the amount of pressure applied and the number of passes with the app licator will affect the
125 development of LL, probably because new infecti on sites are opened and others destroyed. In any experiment attempting to quantify a virus pr eparation using LL formation the inoculations should be conducted by the same individual (Bawde n, 1956). Finally, some kinds of plants, such as Capsicum sp., produce substances which inhibit mech anical virus transmission (Paliwal and Narianai, 1965). When LL development is plotted against vi rus concentration, curves generated for tobamoviruses tend to have a flat portion at high concentration, a steeper portion where the slope reflects a more-or-less proportiona l relationship between concen tration and LL number, and a flat portion at low concentrati on. While many viruses produce results like this when diluted, others such as TSWV and TRV produce curves with different characteristics (Bawden, 1956). Comparisons are most sensitive when samples to be tested are in concentrations producing an average 15-30 LL per half-leaf. For LL produced on tobacco by TMV, this more linear portion of the curve usually occurs after purified virus preparations have been diluted to around 1 part virus to 1,000 parts buffer (1:1,000), or about 10-6 g purified virus per milliliter. The best discrimination of virus (TMV) content reporte d was 25-50% using the half-leaf method and 50 N. glutinosa plants, and 10% usi ng bean (Bawden, 1956). The lesion numbers obtained in the LL assays do not translate directly into relative virus content. For each new virus preparation, a new dilution curve needs to be derived, because the slope of the dilution curve is not constant, indeed, it is unpredictab le (Bawden 1956). The minimum number of virions required to initiate an infection in any given host remains unknown (Roberts, 1964). Randomization is important, not just between leaves, but between plants in a treatment group. Data obtained from randomized bl ock designs may be subjected to analysis of variance (ANOVA) to determine the significance of differences between the number of lesions
126 produced by different inocula, provided the inoc ula do not differ greatly from one another. To plot the data obtained, a methods of transforming data independent of means depend upon the number of lesions obtained. For 10 or more le sions, the equation z = log (x+c) may be used, where x = lesion number and c is a constant obtained by plotting st andard deviations against the means and extrapolating the regression line to the abscissa. For less than 10 lesions, cx x c x z 2 2 1 log2 may be applied. To get the most out of a LL assa y involving serial dilutions, vari ation must be controlled as much as possible. It is known that virtually the same number of LL from a given inoculum will form on each half of an opposite leaf, so a random ized half-leaf method is probably the best to use in most circumstances. With the half -leaf method, the finest discrimination between inoculum concentrations possible was 10% on Phaseolus vulgaris (bean), and 20% on N. glutinosa using a 10-6 g/mL solution of TMV (Roberts 1964). One method of producing randomized treatment groups is the Latin square, a gr id with a series of le tters arranged such that no letter is repeated twice within the same row or column. Each treatment with a virus dilution is assigned a letter, and the arra ngement of half-leaves is such that no treatment falls upon both halves of the same leaf. The same experimenter must inoculate every ha lf-leaf in the assay. Uniformity may also be increased by removing all leaves not to be inocul ated and putting plants in darkness a day before they are inoculated; this will also increase the number of LL formed (Bawden, 1956). To determine dilution endpoint of TMGMV PV-0113 and to see if the host response to the virus is consistant with that of TMV (flat at lo w dilution, steeply-sloping in the middle, and flat at high-dilution), dilution experiments were conducted. Purified PV-0113 was also assayed
127 using LL counts after lyophili zation and storage to determine if the virus retains infectivity when subjected to these treatments. Materials and Methods Two experimental designs were used to obtain LL assay-dilution endpoint data. In the first design, two separate viral prep arations, D and Q, (chapter 6). After dilution with Na2PO4/NaHPO4 inoculation buffer to one opt ical density (1 OD) at A260, the preparations were diluted 1:10, 1:1,000, 1:10,000, 1:50,000, and 1: 100,000 from the 1 OD reference point (approximately 320 g/mL). Leaves on several tobacco ( N. tabacum cv. Samsun, NN a local lesion host) plants were labeled such that three entire tobacco leaves were inoculated with each virus preparation; three leaves were mock-inoculated with a buffe r solution only. The leaves to be inoculated were randomized in such a ma nner that one leaf from each of three general positions on the tobacco plant was in each treatme nt. In addition, each leaf in the set was on a separate tobacco plant. One TSA plant was i noculated concurrently with each treatment. In the second design, half-leaves on tobacco pl ants were assigned random treatments using the Latin square method (Table 7-1) and inoc ulated with various dilutions of PV-0113. Results In the first experiment, initially the LL did not appear simultaneously, but continued to appear over time (Figure 7-1, Figur e 7-2) until about 5 days post-i noculation (DPI), after which spots counted on the leaves were likely not LLs, but attributable to other causes. To account for this effect, the data were divi ded into initial LL, that is, those LL counted on the first day LL were observed (usually by 3 DPI), and maximum LL, the total amount of LL that developed on the inoculated leaf over a 9day period (Figure 7-3, Figure 7-4). Even highly diluted preparations (1:50,000; 6.3 x 10-6 mg/mL) produced LL on tobacco and disease symptoms (epinasty, LL, wilting) on TSA. An average of local lesions produced at each dilution from a
128 few different experiments produced a dilution end point curve characteris tic of a tobamovirus when plotted (Figure 7-5). Plants inocul ated with inoculum diluted 1:100,000 (3.2 x 10-6 mg/mL) did not develop symptoms and resembled the mock-inoculated plants in appearance. Both D and Q preparations of TMGMV demonstrated an ability to kill TSA when diluted 1:50,000 (6.3 x 10-6 mg/mL; Figure 7-6, Figure 7-7). The LL-forming ability on tobacco correlated with symptom development (epinasty, lo cal lesions, wilting, and systemic necrosis) on TSA at concentrations up to 1:50,000 (6.3 x 10-6 mg/mL), although in at least one trial TSA was not killed at this dilution. In one experiment, the reconsti tuted lyophilized preparation Q, at dilutions of 1:10 (3.2 x 10-2 mg/mL), 1:100 (3.2 x 10-3 mg/mL), 1:1,000 (3.2 x 10-4 mg/mL), 1: 5,000 (6.3 x 10-5), 1:10,000 (3.2 x 10-5 mg/mL) and 1:100,000 (3.2 x 10-6 mg/mL), failed to elicit LL except at the highest concentration. However, in a subseque nt experiment using reconstituted preparation Q, LL were observed at all concentrations except 1:100,000. A D preparation of PV-0113 stored under refrigeration (Figure 7-8) and reco nstituted from lyophilized virus (Figure 7-9) had a dilution endpoint on Nicotiana tabacum cv. Samsun ( NN ) at dilutions of 1:25,000 (1.3 x 10-5 mg/mL) and 1:50,000 (6.3 x 10-6 mg/mL), respectively. Discussion Purified TMGMV PV-0113 sometimes produced disease in TSA at the remarkably low dilution of 1 part in 50,000 (6.3 x 10-6 mg/mL). In agreement with results reported for TMV, the linear portion of the curv e appears at dilutions > 1:1,000; a ny future experiments should be carefully designed, probably with intervals 1:1,000 in order to narrow the dilution at which no more LL are formed. It may well be worth comparing the relativ e sensitivity of LL produced by different Rgene/elicitor interactions, because the data obtained might prove useful in establishing the
129 threshold of elicitor required for LL to occur. One problem with this experiment is that the Ngene and N -gene containing host species are superfic ially different and therefore may respond differently to the virus due to physical factors rather than biochemical ones (i.e., the HR signal cascade). However, this problem may be overcome using the Samsun EN tobacco (in which the N resistance gene is active) to minimize ef fects caused by differences in leaf morphology, topology, or other physiological or non-HR related biochemical differences between the species.
130 Table 7-1. Latin square design used in N. tabacum experiments. Plant #1 Plant #2 Plant #3 Plant #4 L R L R L R L R I E G H F B C D A II D F G E A B C H III H B C A E F G D IV B D E C G H A F V C E F D H A B G VI A C D B F G H E VII F H A G C D E B VIII G A B H D E F C Dilutions: A = 1:10, B = 1:100, C = 1:1,000, D = 1:5,000, E = 1:10,000, F = 1:25,000, G = 1: 100,000, H = (-) no virus. Leaf position: L = left half, R = right half.
131 0 50 100 150 200 250 0123456789 Days post inoculation (DPI)Average Lesions per Leaf 1:10 1:1,000 1:10,000 1:50,000 1:100,000 Figure 7-1. Local lesion developm ent over time, Preparation D. 0 50 100 150 200 250 300 350 0123456789 Days post inoculation (DPI)Lesions per leaf 1:10 1:1,000 1:10,000 1:50,000 1:100,000 Figure 7-2. Local lesion developm ent over time, Preparation Q.
132 0.0 50.0 100.0 150.0 200.0 250.0 "1:10""1:1,000""1:10,000""1:50,000""1:100,000" Dilution from 1 O.D. @ A260Average Local Lesions Initial 3 DPI Max. 9 DPI Average Figure 7-3. Local lesions produced on Samsun ( NN ) by TMGMV PV-0113 preparation D; as calculated by taking the average number of LL per half-leaf, divided by the number of half-leaves, x2. 0.0 50.0 100.0 150.0 200.0 250.0 "1:10""1:1,000""1:10,000""1:50,000""1:100,000" Dilution from 1 O.D. @ A260Average Local Lesions Initial 3 DPI Max. 9 DPI Average Figure 7-4. Local lesions produced on Samsun ( NN ) by TMGMV PV-0113 preparation Q; as calculated by taking the average number of LL per half-leaf, divided by the number of half-leaves, x2.
133 0 50 100 150 200 250 "1:10""1:1,000""1:10,000""1:50,000""1:100,000" Dilution from 1 O.D. @ A260Average Local Lesions Figure 7-5. Curve formed by average of all local lesions produced on Samsun ( NN ) by both D and Q preparations.
134 0 1 2 3 4 5 01234567891011121314 Days post inoculation (DPI)Disease Progress "1:10" "1:1,000" "1:10,000" "1:50,000" "1:100,000" Figure 7-6. Disease progress on TSA (1 plant per dilution) inoculated with TMGMV PV-0113 preparation D.
135 0 1 2 3 4 5 01234567891011121314 Days post inoculation (DPI) Di sease P rogress "1:10" "1:1,000" "1:10,000" "1:50,000" "1:100,000" Figure 7-7. Disease progress on TSA (1 plant per dilution) inoculated with TMGMV PV-0113 preparation Q.
136 pp(g) 54.4 14.0 10.1 6.8 2.6 0.9 0.50.5 0.0 10.0 20.0 30.0 40.0 50.0 60.0 1:101:1001:1,0001:5,0001:10,0001:25,0001:100,0000 DilutionAverage Local Lesions Figure 7-8. Average number of local lesions per half-leaf produced by (ref rigerated) preparation D; (storage time, four months: 10/03/07-2/04/07).
137 pp() 79.4 38.0 6.0 7.5 3.3 2.6 0.9 0.3 0.0 10.0 20.0 30.0 40.0 50.0 60.0 70.0 80.0 90.0 1:101:1001:1,0001:5,0001:10,0001:25,0001:100,0000 DilutionAverage Local Lesions Figure 7-9. Average number of local lesions per half-leaf produced by (l yophilized) preparation D; (storage time, four months: 10/03/07-2/04/07).
138 CHAPTER 8 CLONING OF PV-0113 COAT A ND MOVEMENT PROTEIN GENES Introduction Since 1982, when TMV became the first RNA virus to have its genome completely sequenced (Goelet et al., 1982; Robinson, 2004), nucleotide sequencing has become standard in characterization of plant viruses, and complete sequences exist for many tobamoviruses. At the time of this writing, complete sequences have been reported for TMV, (including strains crTMV, TMV-p, and CTMV-W) (Shimamoto et al ., 1998; Yamanaka et al., 1998), and many other tobamoviruses, including Bell pepper mottle vi rus (BPeMV), Cucumber green mottle mosaic virus (CGMMV), Kyuri green mottle mosaic virus (KGMMV), Maracuja mosaic virus (MarMV), Obuda pepper virus (ObPV/TMV-Ob ), Odontoglossum ringspot virus (ORSV), Oilseed rape mosaic virus (ORMV), PMMV, Sunn-hemp mosaic virus (SHMV), TMGMV, ToMV, Turnip vein clearing virus (TVCV), and Zucchini green mottle mosaic virus (ZGMMV). Partial sequences have been determined for many strains of TMV and other tobamoviruses, including Cactus mild mottle virus (CMMoV), Cu cumber fruit mottle mosaic virus (CFMMV), Hibiscus latent Fort Pierce virus (HLFPV), Hi biscus latent Singapore virus (HLSV), Nigerian tobacco latent virus (NTLV), Paprika mild mottle virus (PaMMV), Ribgrass mosaic virus (RMV), Streptocarpus flower break virus (SFBV) and even Tropical soda apple mosaic virus (TSAMV) (Rhie et al., 2007; Ryu et al., 2000; Boubourakas et al., 2004; Song et al., 2006; Heinze et al., 2006; Lapido et al ., 2003; Alexandre et al., 2000; Gibbs et al., 2004; Min, et al., 2006; Yoon et al., 2001; Yoon et al., 2002; Sr inivasan, 2004; Kamenova and Adkins, 2004; Adkins et al., 2007). At the time of this writing, there are tw o complete genomic sequences of TMGMV on record: M34077 (Solis and Ga rca-Arenal, 1990) and AB 078435 (Okuno et al., 2002), both
139 archived in the National Center for Biotechnology Information (NCBI) database, and 108 partial ones. Genomic RNA of TMGMV M34077 was 6355 nt in length, while that of TMGMV-J AB 078435 was 6356 nt in length, due to a 5 variation. TMGMV-J had an initial GT sequence while TMGMV had G. Alignment of the two se quences revealed 223 nucleotide variations or gaps; most notably a 3-nt gap in the TMGMV (M34077) sequence at TMGMV-J position 40694071, and a corresponding 3-nt gap in TM GMV-J (AB 078435) at TMGMV position 4770-4772. Search and alignment using the NCBI Basic Local Alignment Search Tool (BLAST) verified an overall sequence identity between the two sequences of 96 percent. In a 2004 publication, A. Gibbs reported that the phenotypically similar quasi-species such as tobamoviruses were better identified by short nucleotide motifs than traditional nucleototide signatures, because in tobamoviruses, traditio nal nucleotide signatures are very short and scattered throughout the genome. In a set of 10 sample sequences of TMV and 8 sample sequences of the extremely similar tobamovi rus, ToMV, 16.3% of the nucleotides in the sequences identified an isolat e as belonging to one group or another; however 1 out of 7 nucleotide positions were sufficient to distinguish each isolate of ToMV from TMV. Naming the distinguishing patterns in sequence variation nucleotide co mbination motifs, Gibbs chose the TMV 4404-4450 segment because 29 of the 47 nucleotides showed no variability between the many tobamovirus sequences analyzed and b ecause the region so far has proven to match well only within sequences reported for other t obamoviruses for which, so far as is known, it appears unique. Additionally, 109 combinations can be obtained from 18 variables, assuming independence and ignoring possibl e functional constraints on the RNA. This makes it unlikely that new tobamovirus species discovered will be misclassified using this motif. The universal
140 tobamovirus motif corresponding to Tobacco mo saic virus (TMV) genomic nt positions 44044450 reported is: 3 GG__A_GT_AC_AC_TT_AT_GG_AA_AC__T_AT_AT_GC__C_TG 5 In the genome of Tobacco mild green mosaic virus (TMGMV), the Gibbs motif begins at nt 4392 for isolate AB 078435 and at nt 4388 for isolate M34077, within the region bounded by the 3-nt deletions. The motif sequence itself however, is identical, for both sequences: 3 GGTGATGTGACTACTTTCATCGGCAATACTGTTATAATAGCAGCTTG 5 Since the production of a TMV-c ontaining plasmid (clone) capab le of producing infectious RNA transcripts (Dawson et al., 1986), several su ch infectious tobamovirus clones have been made, including TMV-Ob (Padgett and Beac hy, 1993), TMGMV (Morishima et al., 2003), ToMV (Hori and Watanabe, 2003; Weber et al., 1992), ORSV (Yu and Wong, 1998), MarMV (Song et al., 2006), KGMMV (Yoon et al., 2001), and ZGMMV (Yoon et al., 2002). In 2003, an infectious transcribable clone of a Japanese isolate of TMGMV was created. The virus, TMGMV-J, isolated from peppers in Kochi Prefecture, Japan, was sequenced, and the in-vitro transcribable clone, pTGKJ W created from it proved infectious. Infectivity was enhanced by encapsidation with purified TMGMV CP. The MP gene of this clone was then used to create a chimeric ToMV that would not infect tomato (Morishima et al., 2003). Chimeras prepared from tobamoviruses, especially the TMV/TMGMV-L hybrid 30-B, continue to be used as tools to investigate phenomena in mo lecular biology, and to express proteins of interest (Table 8-1). Peptides pr oduced using tobamovirus constructs include drugs (Takamatsu et al., 1990; Kumagai et al., 1993), vaccines (Yusibov et al., 2002; Szab et al., 2004; Fujiyama et al., 2006), antibodies (Giritch et al., 2006), and biodegradable pesticides
141 (Borovsky et al., 2006). The object ives of the work presented in this chapter was to obtain sequences of the TMGMV PV-0113 isolate to compare with other TMGMV and tobamovirus sequences on record, and to obtain and amplify the gene sequences to be used for transient expression work described later (chapter 9) in this thesis. Materials and Methods Tobacco mild green mosaic virus genomic RNA was reverse-transcribed and amplified using primers created to the 3rd and 4th open reading frames (ORF), as well as a segment containing the 4404-4450 tobamovirus motif (Gibbs et al., 2004), and purified for use in the Gateway BP recombination reaction in conjunc tion with the proprietary cloning vector, pDONR221 to transform chemically competent Esherichia coli. The bacteria were cultured on lysogeny broth agar (LBA) containing 50 g kanamycin per mL as a selective agent. Transformants were selected and cultured in lysogeny broth (LB) containing kanamycin (50 g/mL) and plasmids extracted from them using a Qiagen miniprep kit. The extracted plasmids were then sequenced, and the sequences compar ed to TMGMV in the NCBI BLAST database. Primers incorporating Gateway-specific m odified phage-derived insertion sequences att B1 and att B2 were designed with the help of pD RAW32 (Olesen, 2003), to each ORF of the TMGMV genome, using the sequence M34077 pub lished by Solis and Ga rca-Arenal (1990), and the sequence AB 078435 recovered by Okuno (N CBI BLAST database), as a guide. Oligonucleotides for ORFs 3 and 4, the movement protein and the coat protein, respectively, were synthesized by Integrated DNA Technologies, Inc. ORF 1/2 F: 5[GGG GAC AAG TTT GTA CAA AAA AGC AGG CTT C] ATG GCA CAC ATA CAA TCT A 3 ORF 1 R: 3[GGG GAC CAC TTT GTA CAA GAA AGC TGG GTC] CTA TCT AC(T/C) ACC TGC TTC 5
142 ORF 2 R: 3[GGG GAC CAC TTT GTA CAA GAA AGC TGG GTC] CTA ACA GCC ATT TAA AAA 5 ORF 3 F: 5[GGG GAC AAG TTT GTA CAA AAA AGC AGG CTT C] ATG GCT GTT AGT CTC 3 ORF 3 R: 3[GGG GAC CAC TTT GTA CAA GAA AGC TGG GTC] (T/C)TA AAA CGT ACT CGA TGA 5 ORF 4 F: 5[GGG GAC AAG TTT GTA CAA AAA AGC AGG CTT C] ATG CCT TAT ACA ATC AAC TC 3 ORF 4 R: 3[GGG GAC CAC TTT GTA CAA GAA AGC TGG GTC] CTA AGT AGC YGG AGT TGT 5 (Brackets indicate Gateway recombination sequences). Primers described as specific to TMGMV (Gi bbs et al., 2004) were supplied by Integrated DNA Technologies, Inc. for cloning and sequenc ing of a segment of ORF 2 containing the Gibbs tobamovirus identification motif Universal Tobamovirus Primer 2, 5 TTBGCYTCRAARTTCCA 3 used for first-strand synthesi s, anneals to nt positions 4572 in the TMV genome, and Tobamovirus Primer 3, 5 CARACXATWGTBTAYCA 3 anneals to nt positions 4034 in the TMV genome (Gibbs et al., 2004). First strand cDNA synthesis was accomplishe d using Invitrogen M-MLV RT or the improved version, SuperScriptIII. Briefly, genomi c RNA to be reverse-tr anscribed was diluted to <500 ng and 2 pmol (1-2 L) primers added, along with 1 L 10 mM deoxynucleoside triphosphates (dNTPs). The mixture was adjusted to a volume of 13 L with RN-ase-free water and incubated in a 65C water bath for 5 minutes. The mixture was then transferred to ice for 1 minute before pulse centrifugation to bring the contents to the bottom of the tube. To the incubated mixture was added 4 L 5x First-st rand buffer (250 mM Tr is-HCl pH 8.3, 375 mM KCl, 15 mM MgCl2), 1 L 0.1 M dithiotreitol DTT, 1 L RN-ase inhibitor (40 U/L), and 1-2 L SuperScript III (200 U/L). The reaction mi xture was then mixed by pipette and allowed to
143 incubate at 55C for 45 minutes. Samples were th en heated to 70C for 15 minutes to quench the reaction, after which 1 L RNase H was added a nd the reaction mixture incubated at 37C for 20 minutes. Template DNA thus prepared was used immediately for PCR or frozen at -20C for future use. Template DNA was obtained using ordinary M-MLV RT by following a similar protocol. PCR was carried out with the cDNA obtained fr om RT. For each sample, 3 reactions were conducted, using cDNA (undiluted, 1:10, 1: 100) in PCR reaction tubes with 24 L PCR mix (2.5 L 10x PCR buffer, 0.9 L 50 mM MgCl2, 2.5 L 2 mM dNTP, 16 L MBG H2O), 1 L forward primer, 1 L reverse primer, and 0.125 L Taq DNA polymerase. Reaction mixtures were overlaid with mineral oil (when required by the thermocycler), and put through a program including 35 cycles of heating and cooling. Products obtained from PCR were purified for use in Invitrogen's Gateway BP recombination reaction using polyethylene glycol (PEG) as follows. Fifty microliters (50 L) PCR product was combined with 150 L TE buffer (10 mM Tris-HCl, 1 mM EDTA, pH 8.0) and 100 L of PEG solution (30% PEG 8000 w/v, 30 mM MgCl2), vortexed briefly, and centrifuged at 10,000 g for 15 minutes. Supernatan t was removed and the pellet re-suspended in 50 L of TE buffer. Quality of the purified sample (free from dimers) was verified by electrophoresis on 0.9% agarose. Purified PCR-generated fragments were ligated into the commercially available cloning ve ctor pDONR221 (Figure 8-1). Concentration of DNA in each PCR sample was estimated using ag arose electrophoresis against a standard ladder and UV spectrophoto metry. To 100 femtomoles (100 fmol) of each PCR sample (19 L of ORF 3, 12 L of ORF 4), 2 L of pDONR (150 ng/ L), and 4 L of rxn buffer were added. Two microliters (2 L) of the plasmid pEXP7-tet (50 ng/ L) was used in one
144 reaction as the positive cont rol for recombination. To each prepared reaction, 2 L of freshlythawed BP Clonase enzyme was added, the r eaction mixture vortexed twice for exactly 2 seconds, and the samples allowed to incubate overn ight at 25C. Afterwards, recombination was quenched by addition of 2 L of Proteinase K (2 g/ L) to each sample and 10 minutes incubation at 37C. Plasmids obtained from this reaction were used to transform E. coli DH5 and stored frozen at -20C. The (standard) heat-shock method wa s utilized for transformation of E. coli Briefly, frozen competent (Invitrogen DH5 ) cells were thawed on wet ice and 50 L placed into sterile 1.5 mL polypropylene tubes for each tr ansformation. One microliter (1 L) of ligated pDONR221 was added and mixed very gently by inversion. The ampicillin-resistant plasmid, pUC19 was utilized as a tran sformation control. Tubes were then incubated on ice for 30 minutes. Heat shock was induced by placing the t ubes in a 42 C water bath for 30 seconds and immediately chilling them on ice with addition of 450 L room temperature super optimal catabolite-repression (SOC) medium Tubes were then capped a nd placed on a 37 C incubating shaker set for 200 rpm agitation and 37C for 1 hour, before spreading onto pre-warmed selective media plates. LBA plates laced with ka namycin (50 g/L) were used for selection of PCR/pDONR recombinants, while the recombin ation control, pEXP7-tet, was assayed on tetracycline (20 g/L) and carbenicillin ( 100 g/L) was used for selection of pUC19. Colonies were screened for inserts using co lony PCR. Colonies to be analyzed were sampled using a sterile toothpick and transferred to 50 L of extraction buffer (1% Triton X-100, 20 mM Tris pH 8.0, 2 mM EDTA) in a 1.5 mL micro centrifuge tube with lid lock and immersed in boiling water for 5 minutes. Upon cooling, 2-5 L of the bacterial extr act was amplified using PCR as described above.
145 Colonies identified as containing sequences of interest were cultured in 3.0 mL LB containing kanamycin (50 g/L) for selection. After 16-18 hours incubati on with agitation at 37C, cultures were centrifuged at setting 8.5 fo r 15 minutes in a TJ-6 Beckman centrifuge. Supernatant was discarded, and th e pellets resuspended in 250 L of P1 buffer (supplied in kit) before transfer to a micr ocentrifuge tube. Two hundred and fifty microliters (250 L) of P2 buffer (supplied in kit) were then added. After mixing by inversion, 350 L of N3 buffer (supplied in kit) was added, and the samples mixed again by inversion, and centrifuged for 10 minutes in a tabletop centrifuge at maximum speed. The supernatant was then removed by micropipette to the center of a QIAprep colu mn, which was centrifuged for 30-60 seconds at maximum speed, and washed with 750 L PE bu ffer (supplied in kit) before elution of DNA with 30 L molecular biology grade water. For applications in which plasmid quality wa s not critical, extraction was accomplished using chloroform/octanol. Bacteria containing plas mids to be extracted were inoculated into 4.5 mL LB containing appropriate antibiotic(s) fo r selection and incubated at 37C for 16-18 hours with agitation. Four (4) mL of culture was centrifuged in the TJ -6 centrifuge at setting 8.5 for 15 min, after which, supernatant was discar ded and the pellet re-suspended in 350 L P1 buffer (50 mM Tris HCl, pH 8.0; 10 mM EDTA; 100 g/mL RNA-ase A), and tran sferred to a 1.5 mL tube. Then, 400 L freshly-prepared P2 buffer (~200 mM NaOH [1 pellet in 9 mL distilled, deionized water]; 9 mL sterile distilled, deionized water; 1 mL 10% SDS) was added and the suspension mixed by inversion. The mixture was allowed to incubate at RT for 5 minutes before 400 L P3 buffer (3.0 M potassium acetate, pH 4.8) was a dded, and the volume mixed again by inversion, then placed on ice. After 15 minutes, the mixt ure was removed from ice and spun at maximum speed in a tabletop centrifuge for 15 mi nutes. The pellet was discarded and 900 L of
146 supernatant transferred into a 2.0 mL microcentr ifuge tube, to which 1 volume isopropanol was then added. After a 15-30 minute spin-down at the maximum setting, pellets were washed with 70% nuclease-free ethanol and vacuum (aspirator ) dried for 7 minutes. Then, the DNA pellet was resuspended in 200 L molecular biology grade water and ex tracted with octanol : chloroform (1:24). Following centrifugation, the aqueous phase was collected, a nd the DNA precipitated using 0.09 volumes of 3 M sodium acetate, a nd 2.5 volumes ethanol. Following a 20-minute centrifugation at 4C (maximum-speed), the pelle t was washed with 70% ethanol as described previously, vacuum dried, and resuspended in 25 L molecular biology grade water. Gateway-compatible gene constructs pKYW G2 and pKYWG2D, suitable for expression in plantae (Karimi et al., 2002), were obtained from the Interuniversity Institute for Biotechnology (VIB) in Ghent, Flanders, Belgium. Constructs were delivered in vivo in the form of bacterial stab cultures (i.e. a special strain of E. coli resistant to the effects of ccdB) and cultured in LB containing 50 g per mL spectinomycin as a selective agent. Plasmids harvested from the cultures, were recombined with the pDONR221 clones. Sequences of interest (PV-0113 ORF 3 and ORF 4) were ligated into VIB plant expression vectors pKYWG2D and pKYWG2 (F igure 8-2), designated E-1 a nd E-2, respectively, using Gateway LR Clonase catalyzed recombin ation. Briefly, 100 300 ng purified pDONR containing PV-0113 sequences (entry clone) we re combined with 300 ng of the expression vector and 4 L 5X LR Clonase reaction buffer in a 1.5mL tube, and the volume adjusted to 16 L with TE buffer. The plasmid pENTR-gus (50 ng/L) was used as a positive control. Four (4) L freshly-thawed LR Clonase was added to each reaction and briefly vortexed twice. Samples were then allowed to incubate for 1-18 hours at 25C. After incubation, the reaction was
147 quenched by 2 L Proteinase K added to each tube, and a 10 minute incubation at 37C. Reaction mixtures obtained were used to transform E. coli DH5 as described above. Transformed E. coli were selected on LB A plates containing 50 g per mL spectinomycin. Selected transformants were screened using P CR as described above, cultured in LB containing spectinomycin (50 g/mL), and the plasmids extracted by the Qiagen method. Subsequent transformations, screenings, extrac tions, and purifications were carried out as previously described. Sequencing was performed by the University of Florida Interdisciplinary Center for Biotechnology Re search (ICBR) sequencing lab using Perkin Elmer/Applied Biosystems automated sequencing. Nucleotide alignment was accomplished using ClustalW (Chenna et al., 2003; Thompson et al., 1997; Thom pson et al., 1994), using default parameters, and Multalin (Corpet et al., 1988), using the DNA sy mbol comparison table and default settings. Multalin processing typically used the DNA symbol comparison table. Sequences were translated using the Translate tool of the ExPASy proteomics server (Gasteiger et al., 2003), and pDRAW32 (Olesen, 2003). Amino acid sequen ces were aligned using Multalin with the Blosum symbol comparison table and default settings. Results Comparison of the ORF 3 nucleotide se quences of TMGMV PV-0113 with TMGMV M34077 (Solis and Garca-Arenal, 1990) and TMGMV AB078435 (Okuno et al., 2002), using Multalin (Figure 8-3) revealed 13 nt variations between PV -0113 and M34077, nine of which were translationally important (Figure 8-4). Between PV-0113 and AB078435 there were 17 nt substitutions, but only six of these were translati onally important. Overa ll, there was an ~98% identity between the ORF 3 of PV-0113, M34077, and AB078435. Translated, the relationship
148 between the sequences is more difficult to interpret. Interestingly, TMGMV PV-0113 and AB078435 share a 5-aa sequence spanning aa 20 9-213 different from that of M34077. Alignment of ORF 4, the coat protein doma in of PV-0113 with M34077 revealed 2 nt variations, neither of which was translationa lly important, while 10 nt substitutions were discovered relative to AB078435 (Figure 8-5), only 1 of which was translationally important (Figure 8-6). Overall, there was a >99% identity between the OR F 4 of PV-0113 and M34077, and a 98% identity between the ORF 4 of PV-0113 and AB078435. The nucleotide sequence from TMGMV PV0113 corresponding to nt 4404-4405 of typestrain TMV matched the predicted Gibbs mo tif for the species, TMGMV (Figure 8-7). Discussion Since the experiments described herein we re completed, a full-length clone of TMGMV PV-0113 has been obtained and sequenced (Dr. Ernest Hiebert, Professor, Un iversity of Florida, Gainesville, personal communicatio n). The nucleotide sequence of this clone (Genbank EF 469769) is highly similar to archived TMGMV seque nces. The quasi-species nature of viruses will allow variability between i ndividual clones; indeed, a certain amount of variability is expected within any species. Variability is, ad ditionally, a liability of the sequencing procedure itself. In parallel with nature, mistakes sometim es occur in the course of reading thousands of sequence reactions: the same genetic material will have an occasional nucleotide mis-read after multiple reactions have been performed (Dr. E. Hiebert, personal communication). Automated sequencing technology has improved considerably over the years; however potential errors in archived sequences may persist. The Gibbs 4404-4450 tobamovirus specific motif appears to be a good indicator for TMGMV. The TMGMV strain PV-0113, which s hows small differences between previously analyzed strains was 100% homol ogous for the entire 47 nt motif -containing segment. It would
149 be interesting to anal yze the tobamovirus-specific motif of tobamoviruses [e.g. Cactus mild mottle virus (CMMoV) and Tropical soda apple mosaic virus (TSAMV)] discovered since the publication of Gibbs work, and test the prediction that combinator ial probability ensures correct identification and classification using the motif. The extremely high degree of identity betw een TMGMV PV-0113 and the archived strains of TMGMV means that separation of particul ar TMGMV strains using serology may well be impossible. A better method would be to look for rather variable regions within the genome and to find a highly variable motif, perhaps based on Gibbs findings. In particular, variability at the 3 end of the virus between the TMGMV-L and TMGMV-S might be exploited. Used in combination with a TMGMV PV-0113-specific primer set, primers designed to a highly variable region could possibly dis tinguish closely-relate d strains of TMGMV.
150 Table 8-1. Useful products produced by using transgenic tobamoviruses. Reference Peptide Application Takamatsu et al., 1990. Leu-enkephalin Experimental painkiller Kumagai et al., 1993. -Trichosanthin HIV/AIDS treatment Yusibov et al., 2002. Rabies peptid e Inexpensive rabies vaccine Szab et al., 2004. Amyloidprotein Alzheimers research Borovsky et al., 2006. Aedes oostatic factor Biodegradable larvacide Fujiyama et al., 2006. Poliovirus pep tide Inexpensive polio vaccine Giritch et al., 2006. IgG Antibodies
151 Figure 8-1. Map of the proprie tary cloning vector, pDONR221 (Invi trogen Manual). attP1, attP2 = Gateway enzyme-mediate recombination s ites; ccdB = gyrase inhibitor for negative selection; CmR = chloramphenico l resistance gene for selection.
152 Figure 8-2. Constructs used in TMGMV PV-0113 research (Karimi et al., 2002).
153 Figure 8-3. Nucleotide alignment of TMGM V PV-0113 ORF 3 with those of GenBank TMGMV sequences M34077 and AB078435.
154 Figure 8-3. Nucleotide alignment of TMGM V PV-0113 ORF 3 with those of GenBank TMGMV sequences M34077 and AB078435 (continued).
155 Figure 8-4. Amino acid alignment of PV-0113 MP with those of (translated) GenBank TMGMV sequences M34077 and AB078435.
156 Figure 8-5. Nucleotide ali gnment of PV-0113 ORF 4 with those of GenBank TMGMV sequences M34077 and AB078435.
157 Figure 8-6. Amino acid alignment of TMGMV PV0113 CP with those of (translated) GenBank TMGMV sequences M34077 and AB078435. Figure 8-7. Isolate PV -0113 contains the Gibbs TMGMV specific motif.
158 CHAPTER 9 TRANSIENT EXPRESSION OF PV-0113 COAT PROTEIN Introduction To identify the gene or genes responsible fo r the lethal HR in the TSA-TMGMV system, transient expression systems were tested. Transient expression, in this case the expression of a single viral gene in planta without the intact virus, may be accomplished by various means, including inoculation with a chimeric virus or virus transcript and by plasmids (vectors) delivered using Agrobacterium tumefaciens or with the aid of proj ectiles. Sometimes, two methods may be combined: a plasmid capable of transcribing chimeric virus RNA in planta may be introduced into the host wher e proteins are then expressed. With a full-length clone in hand, one potentially appealing route is to create a chimeric virus by swapping heterologous regions of two or more closely related viruses. Successful chimeras may enable the iden tification of the function or phenotype of th e heterologous sequence gene. During transient expression of a foreign protein, one potential issue that arises is the way in which the protein interacts with host cellula r machinery. The protein might not, for example, accumulate to high concentrations within the cell. It might inhibit or interfere with cellular metabolism in some way. It might be outright to xic. Strategies employing hybrid vectors help to overcome these concerns, because the rapid expr ession of viral proteins theoretically would allow a large quantity to accumulate before trig gering toxicity to plants (Hori and Watanabe, 2003). In general, tobamovirus vectors have been de signed to work with certain well-studied and laboratory-friendly species such as tobacco th at are easily inoculated and infected and susceptible to a wide range of pathogens. One such example is the tender, highly permissive
159 host, Nicotiana benthamiana Often, the ability of the tobamovirus vectors to replicate and express foreign genes seems limited to th ese hosts (Hori and Watanabe, 2003). For tobamoviruses, it has been determined that more coat protein (CP) is translated than any other gene product. While short peptides have been obt ained by fusing the gene of interest directly to the CP/3 NTR of the tobamovirus, difficulties have been reported in attempting expression of larger proteins such as the marker gene green fluorescent protein (GFP) Success with virus transcripts may be enhanced by in vitro encapsidation of the transcri pts with viral CP, forming more durable and infectious part icles (Hori and Watanabe, 2003). Chimeras incorporating Tobacco mild green mosaic virus (TMGMV) (Table 9-1) are usually constructed by incorpor ating TMGMV genes into a modi fied TMV or ToMV genome. For example, chimeric tobamovirus constructs ca pable of multiplying a nd infecting solanaceous plants and protoplasts were created from ToMV and the Japanese isolate, TMGMV-J. The chimera, TocJ, which expressed TMGMV-J CP, wa s created with the ToMV CP start codon that was disabled using primer-forced nucleotide substitution, and by inserting the TMGMV CP sequence, and the GFP variant G3 (Hori and Watanabe, 2003). Another early TMV/TMGMV chimera, V23, wa s used to investigate the relationship between virus movement and the phenotype of local lesions. The product of a movement protein-defective TMV (V1), from which the 5 end of ORF 3 was removed and ORF 3 of TMGMV (ATCC PV-0228) was inse rted, it behaved much like TMV, moving systemically in the TMGMV non-host Lycopersicon esculentum (Nejidat et al., 1991). The TMV/TMGMV-L chimera 30B, which incorporates a duplicated 3 NTR to enhance transgene expression, suffers from recombination arising in planta which results in progeny that have lost the transgene but remain less vigor ous than wild-type (WT) tobamoviruses.
160 Interestingly, GFP-containing vect ors bearing TMGMV-L ORF 4 and 3 NTR sequences produced much more of the transgenic product, GFP, than heterol ogous sequences from any other tobamovirus species hybridiz ed (Shivprasad et al., 1999). A direct method for delivery of genes into a cell is the biolisitic or gene gun approach. Originally, devices were designed using 0.22 bl anks to propel a disk containing DNA-coated particle towards an abrupt st op, hurling the particles towards the plant target and embedding them in the tissue. This crude design was e ffective enough, but has been improved by the use of compressed helium and a vacuum, so that a finer degr ee of control over the ve locity of the pellets (and therefore the depth of their penetration) can be achieved. Gene rally, gold particles are preferred over tungsten for their uniformity a nd chemical inertness (Sawant et al., 2000). Gene delivery using particle bombardment has been carried out for some time. Transient expression (phytochrome, chloramphenico l acetyl transferas e) in cereals ( Hordeum and Oryza ) and other plants using the part icle gun delivery method was achieved by the end of the 1980s (Bruce et al. 1989.) Morikawa et al. (1989.) used projectiles to effect transient expression (luciferase, GUS) in cultured t obacco and eggplant tissues. The de livery device used pressurized N2 to drive a Teflon projectile coat ed with gold particles towards an abrupt stop which hurled the particles into the tissue. From these crude but effective gene guns a device called a particle inflow gun (PIG) was developed (Finer et al., 1992), using pre ssurized helium against a partial vacuum to fling DNA coated particles adhering to a screen grid towards the target, rather than to accelerate them using a carrier projectile, as prev ious models had. Later, Gray et al. (1994) described a simplified design using tungste n particles for delivery. Though relatively inexpensive, it was nonetheless effective, produc ing transient expression of GUS in cantaloupe cotyledons.
161 Using a Finer-type gun (Finer et al., 1992), PIG gun, and gold particles, Romano (Romano et al., 2001) describes transient expression ( -galacuronidase, lucifera se), and transformation (kanamycin resistance) in potato tissues. It was no ted that tissue and cell damage occurs from the process and so the researcher must balance dept h of penetration against the number of stable transformants developed. It was discovered that internode tissues were th e most tolerant of the mechanical insult from bombardment. The process was further improved when it was determined that calcium-spermidine was superior to calcium phosphate for use in the preparation of the DNA-coated projectiles (Aragao et al., 1993). Another improvement was the determination that baking the projectiles at a high temperature before precipitation would overcome agglomeration, which makes standardi zation difficult and damages the host tissue. Vortexing, and the use of glycerol and polyethyl ene glycol (PEG) are also employed to help reduce agglomeration (Sawant et al., 2000). The objective of the studies reported in this chapter is to determine whether the coat protein (CP) might elicit th e observed hypersensitive respons e (HR) of TMGMV PV-0113 in tropical soda apple (TSA), ect opic expression was attempted us ing a TMV/TMGMV chimera. Plasmid vectors for in planta transient expression of TM GMV PV-0113 CP and movement protein (MP) were constructed and attempts we re made to deliver these using biolistic and A. tumefaciens delivery systems. Materials and Methods The chimeric tobamoviruses 30B (Figure 91) and 30B-GFPc3 were kindly provided by Dr. Shailaja Shivprasad, formerly at the C itrus Research and Edu cation Center (CREC), University of Florida/IFAS, Lake Alfred, and the master sequence of 30B was kindly provided by Dr. Dennis Lewandowski, also formerly of th e CREC. 30B and 30B-GFPc3 were inoculated onto plants of Nicotiana benthamiana N. sylvestris N. tabacum and Solanum viarum (tropical
162 soda apple; TSA). TSA plan ts to be inoculated with TMGMV strain U2 (LAU2) and TMGMV strain U5 (LAU5) both obtained from Dr. Dennis Lewandowski, were grown in peat pellets and maintained in a growth chambe r, (25C day, 20C night; 16-hour photoperiod). Plants of TSA and N. benthamiana inoculated with 30B in frozen tobacco tissue, provided by Dr. Dennis Lewandowski, and TSA inoculated with TMGMV PV-0113 preser ved 1:1 in frozen tobacco sap extract, provided by M. S. Elliott (Senior Biologist, Plant Pathology Department, University of Florida), or with or with TMV pr eserved 1:1 in frozen t obacco sap extract, and provided by M. S. Elliott, were grown in clay pots and maintained in an Apopka growth room. After inoculation, plants were observed daily for symptom development under both visible (fluorescent) and UV light. UV illumination was initially provided by laboratory model UV illuminators ( = 356 nm, 366 nm), but further experimenta tion demonstrated that an ordinary blacklight fluorescent lamp illuminated areas in which GFPc3 was present just as vibrantly. After symptoms became visible, affected plan t tissues were harvested and extracted using Laemmli dissociation solution (LDS) (Laemmli, 1970). Separation was accomplished on a polyacrylamide gel composed of a 5.6% stacker and a 10% separato r. Samples loaded included 15L extract of symptomatic tissue from TSA inoculated with 30-B, LAU5, LAU2, PV-0113, and TMV. 15L of tissue extract from uninoculat ed TSA and 4L BioRad Precision Plus Allblue protein standard (10 kDa-250 kDa; cat. #1610373) were also loaded. Electrophoretic separation was done at 125 V and the proteins transferred to a nitrocellulose membrane under Laemmli running buffer (Laemmli, 1970), at 100 V, (current = 0.16 amp at start and 0.20 amp at finish). Afte r electroblotting, the membrane was blocked [to cover open nitrocellulose] with a 5% solution of evaporated milk, washed 3 times in Trisbuffered saline solution with Tween 20 (TBST), and incubated for 1 hour with primary antibody
163 (rabbit anti-TMGMV; rabbit #995). The membrane was then rebl ocked using sap from healthy tissue (1:100 in 0.1 M NaHPO4/Na2PO4; pH 7.0), washed 3 more times with TBST, and incubated for 1-2 hours with goa t anti-rabbit IgG conjugated to alkaline phosphatase. Both primary and secondary antibodies were diluted 1:1 with blocking solution before use. The final concentration of secondary, conjugated an tibody was 1:60,000. One (1) drop of sap from healthy TSA was added to 40 mL total primary anti body and blocker (20 mL total used per blot). Finally, the membrane was washed twice more in TBST and once with substrate buffer for 5 minutes with agitation. The protein blot was developed by exposure to Promega Western Blue stabilized substrate in darkne ss until good contrast was obtained (~5-10 minutes). Development was halted by a rinse with distilled water. The expression constructs used, pK7W G2 and pK7WG2D, are based on the spectinomycin-resistant pPZP 200 se ries of binary vectors (Karim i et al., 2002). They carry an origin of replication capa ble of functioning in both Escherichia (~200 copies/cell) and Agrobacterium (3-5 copies/cell). The pPZP vector s are small (~6.7 Kb) compared to other binary vectors, and incorporate a multiple cl oning site (MCS) for c onvenient engineering of larger constructs (Hajdukiewi cz et al., 1994). Binary vector s created from pK7WG2 and a control marker (pE1-gus), TMGMV PV-0113 ORF3 (pE1-A2) and TMGMV PV-0113 ORF4 (pE1-D2), as described previously, were incorporated into disarmed Agrobacterium strain ABI [Ti plasmid pMP90RK; GmR, KmR] (Koncz and Schell, 1986), (kindly provided by Dr. H. Klee, Eminent scholar, University of Florida/IFAS, Ho rticultural Sciences Department, Gainesville), using electroporation. Cells were cultured on D1 medium (Kado and He skett, 1970), before being transferred to yeast-extract peptone (YEP) medium (2 g Bactopeptone, 1 g yeast extract, 1 g NaCl; pH 7.2).
164 Five (5) mL of YEP broth, containing 50 g pe r mL kanamycin sulfate was inoculated with Agrobacterium tumefaciens disarmed strain ABI, and cultu red overnight at 28C with 200 rpm agitation. Two (2) mL of this culture was then transferred to 200 mL YEP and incubated at 28C, 250 rpm 4-5 hours, or until the cell concentration reached A600 = 0.3 OD. Cultures were centrifuged at 5,000 rpm for 10 minutes in a chilled rotor. The supernatant was discarded and the pellet resuspended in 20 mL of cold (4C), f ilter-sterilized 1 mM HEPES buffer (1mM HEPES, 1 mM KOH, pH to 7.0). Cells were recentrif uged and resuspended twice. After a final centrifugation, the cells were resu spended in 2 mL of ice-cold (0C) filter-sterilized 10% glycerol. Forty (40) L aliquots of cells were quick-frozen in liquid N2 and stored at -80C. The remaining electrocompetent cells were electroporated. Plasmid DNA (1 g/40 L cells) was introduced and the cultures incubated on wet ice for 2 min. The DNA-bacteria mix was then transfer red to a chilled electroporation cuvette and pulsed at 25 F, 2.5 kV, 400 using a BioRad Gene Pulser Xcell programmable electroporation system. To the electro-shocked bacteria, 1 mL super-optimal cat abolite-repression broth (SOC) was immediately added, and the culture incubate d at 28C for 4-6 hours. One hundred (100) L of each transformation mix were then cultured on YEP agar laced with 50 g per mL kanamycin and 100 g per mL spectinomycin for chromosomal and Ti plasmid selection. Colonies identified as carrying sequences of interest (as determined by colony PCR) were selected for agro-infiltration. Each colony sel ected was inoculated to 5 mL YEP containing 50 L per mL kanamycin and 100 g per mL spectin omycin and cultured for 24 hours at 28C. One half (0.5) mL of this culture were used to inoculate 200 mL of kanamycin and spectinomycin laced-YEP containing 10 mM MES (pH 5.85) to neutralize metabolites and 20 M acetosyringone which was dissolved in dimethyl sulfoxide (DMSO) and filter-sterilized to
165 induce vir gene expression. This culture was then incubated at 28C overnight with agitation before 40 mL aliquots were removed and centr ifuged at 5,000 rpm for 15 minutes in sterile Oakridge tubes. The malodorous pellet was su spended in 35 mL of ME S buffer (10 mM MES, pH 5.85 and 10 mM MgCl2). The cells were centrifuge d again 6,000 rpm for 15 min and resuspended in 35 mL induction buffer (10 mM MES, pH 5.85; 10 mM MgCl2, 150 M acetosyringone.) Resuspended cells were incuba ted at room temperature for 4 hours and then diluted with induction buffer to ~1 O.D. at A600 and infiltrated directly into the underside of the leaves. Treatments consisted of Agrobacterium carrying pE1_A2 and pE1_D2 alone, and in combination. Cotyledons and primary leaves were preferred for infiltration. Infiltrated plants were maintained in a warm greenhouse. For certain experiments, expression vector s were combined with AGL0 carrying gene silencing suppressors (various viral genes i nvolved in overcoming RNAi-based host cell defenses). These vectors, carrying either ge ne silencing suppressor p19 (Voinnet et al., 2003; Win and Kamoun, 2004), or 2b (Mayers et al., 20 00), in plasmid construct pCASS, were prepared for infiltration in the same manner, and mixed with the PV-0113 ORF3/ORF4-carrying vectors, 1:1 (v/v), immediately prior to infiltration. Another gene silencing suppressor used was already present in the transgenic host, Nicotiana tabacum cv. Xanthi [+HCPro] (kindly provided by Dr. Jane Polston, Professor, University of Fl orida, Plant Pathology Department, Gainesville). Plants infiltrated for each treatment consisted of TSA (5), Nicotiana tabacum cv. Xanthi [+HCPro] (3), Nicotiana sylvestris (1), and N. tabacum cv Samsun [ nn ] (1). Additionally, AGL0 carrying the marker gene, green fluorescent protein (wtGFP) in the binary construct pCASS, was used as a positive control.
166 AGL1 and AGL0 are hypervirulent, disarmed strains of Agrobacterium derived from EHA101, which carries the engin eered super virulent Ti-pla smid pTiBo542 (i.e., the super virulent plasmid was disarmed, enabling high tran sformation efficiency). It also contains a mutated recA gene, so that recombination events which might disrupt the plasmids are kept to a minimum. Since this strain of Agrobacterium carries carbenicillin resistance (CbR) for selection, but not spectinomycin resistance (SpR; Lazo et al., 1991), it was c hosen for use with the SmR/SpR -containing binary vectors pK7WG2 and pK7WG2D. After infiltration, plants in treatment groups incorporat ing the eGFP or wtGFP were observed daily for symptoms under both visi ble (daylight, fluorescent) and UV (366 nm, blacklight) light. Tissue from pE1-D2 inoculated N. sylvestris N. tabacum Xanthi [HCPro] and TSA was harvested 24, 72, and 120 hours post-inocul ation, triturated under Laemlli dissociation buffer, and frozen at -20 C in preparation fo r Western blotting. West ern blotting was carried out as described previously in this materials a nd methods section, except that the tissue extracts used were from the agro-infiltrated N. sylvestris N. tabacum To investigate sensitivity of TSA to agro-inoculation, A. tumefaciens strain ABI, carrying Ti plasmid pMP90RK, was cultured overnight in 5 mL of YEP broth, co ntaining 50 g per mL kanamycin sulfate (Virulent A. tumefaciens strain AT-3, grown in YE P broth without antibiotic, was used as a control.) Five hundred (500 L) of this culture was used to inoculate a 200 mL flask of YEP containing 10 mM MES (pH of MES adjusted to 5.85), and 20 M acetosyringone. This flask was incubated at 28C with agitation (200 rpm) overnight. The overnight cultures were centrifuged in a refrigerated Sorvall centr ifuge at 5,000 rpm for 15 minutes. Pellets were resuspended in 175 mL infiltration bu ffer (10 mM MES, (pH 5.85); 10 mM MgCl2), and recentrifuged at 6,000 rpm for 15 minutes. Supernat ant was discarded and the pellets were re-
167 suspended in 175 mL of the same buffer, with the addition of 150 M acetosyringone to induce vir-gene expression. Cultures thus prepared were allowed to inc ubate at RT for 3-5 hours prior to agro-infiltration. Plants of N. sylvestris N. tabacum cv. Samsun ( nn ), N. tabacum cv. Xanthi [HCPro], and TSA, 3-4 weeks old, were infiltrated with Agrobacterium suspension four times in three half-leaves per plant. Additionally, each plant was inoculated by inserting an Agrobacterium suspension-coated toothpick into the stem. In TSA, this was accomplished by piercing downward at an axial leaf node along the stem. Plants of N. benthamiana N. sylvestris and TSA were grown in Metro-mix potting mix, in 6.35 cm square plastic pots until the plants had at tained suitable size (~3 weeks) for inoculation. The plants were then used for biolistic infiltra tion. In subsequent experiments, plants were grown in expanded peat pelle ts, in a growth chamber (25C day, 20C night; 16 hour photoperiod), and subjected to a 24-hour period of darkness prior to bombardment to enhance sensitivity to the infiltrated material. The particle inflow gun (PIG) was similar to the design described by Gray et al., (1994). Essentially, it is a welded steel box with a thick Plexiglas door f itted with a thick gasket. The nozzle delivering the pressurized helium was located in the top and the vacuum outlet located in the side. The solenoid regulating the helium delivery was triggered by manually pressing an electronic switch. Five protocols for projectile-m ediated transient expression/transformation were examined, and a new protocol was adapted to the available facilities. Gold particles (1.5 3.0 m) were baked in an oven overnight and stored in a desiccat or before use. Ten (10) mg of gold particles prepared in this manner were suspended in NaCl/Tris (150 mM NaCl, 60 mM Tris) and
168 approximately 20 g of plasmid vector a dded. To this suspension, 100 L of 100 mM spermidine, 100 L of 100 mM PEG, and 100 L of 2.5 M CaCl2 were added, and the whole suspension was incubated for 10 minutes at room temperature. Particles were pulse-spun in a tabletop centrifuge and the supernatant discarde d. Particles were washed twice with 100% ethanol, and then resuspended in 20 0 L of absolute ethanol. Thr ee (3) L of DNA/gold particle suspension per shot were evaporated onto the filt er screen and used to bombard the plants. One to three (1-3) shots per plant were applied, under a vacuum of ~25 psi. Transient expression vectors were extracted us ing the Eppendorf miniprep kit from clones pE1-A2 (pKYWG2D + NZ TMGMV ORF3) an d pE1-D2 (p pKYWG2D + NZ ORF4) in Escherichia coli and clones pCASS [+2b] and pCASS [+wtGFP] in A. tumefaciens strain AGL0. Escherichia coli ., strain DH5 was cultured overnight at 37 C in lysogeny broth (LB) containing 50 g per mL kanamycin. Agrobacterium tumefaciens strain AGL0 was cultured 20 hours in YEP broth containing 100 g per mL ampicillin at 25C. Extracted plasmids were adjusted to an estima ted concentration (agarose gel estimation) of 0.1 g per L. Ten (10) L of this solution was added to 10 L of freshly-vortexed 50 mg 0.7m tungsten metal particles suspended in a steril e 50% solution of glycerol and distilled, DNase and RNase-free water. The projectile samples we re incubated at room temperature for 1 minute, before the plasmid DNA was precipitated onto th e particles by the addition of 10 L of 1.25 M CaNO3, pH 10.5. The suspension was then mixed by inversion, brief vortexing, and incubated at room temperature for 3 minutes. Samples were then centrifuged at 13,000 rpm in a tabletop centrifuge, and 2/3 (20 L) of the supernatant was removed from each sample and discarded. Samples (including a tungsten only negative control consisting of only tungsten prepared as above (glycerol, CaNO3) without plasmid DNA) were then immediately loaded into a Bim-
169 Laboratory inoculation device, (kindly provided by Dr. Avihai Ilan, Bio-Oz Technologies, Kibutz Yad-Mordechai, Israel) and injected (5 L per shot, pressure = 3 bar, distance 1 cm) into the leaves of N. benthamiana N. sylvestris and TSA (Table 9-2). The Bim-Lab is a device that utilizes compressed air to efficiently deliver small volumes of virus inoculum or DNA into intact plants or plant tissue (Anonymous, 2005). Following inoculation/infiltration, plants were returned to a cool greenhouse and observed visually, both under ambient lighting and a blacklight UV source. After 9 days, infiltrated tissue was collected and a portion extracted with LDS before freezing for future experimentation. Epidermal peels of construct and virus-infilt rated plants were mount ed in 10% glycerol, sealed with clear nail polish, and viewed unde r an epifluorescent microscope using a GFPoptimized filter set (kindly provided by Dr. Jeff Rollins, Associate Professor, Plant Pathology Department, University of Florida). Results Alignment of TMGMV PV-0113 ORF4 with the 30B CP nt sequence using Multalin (Corpet, 1988), revealed four nuc leotide variations (Figure 9-2), of which one was translationally important (Figure 9-3). TSA plants inoculated with LAU5, developed local lesions and either collapsed, developed leaf abscission, or, in a few instances, survived with mosaic symptoms. TSA plants inoculated with LAU2 either developed no symptoms, epin asty, local lesions, or mosaic, but did not succumb. TSA plants inoculated with PV-0113 consistently develope d local lesions, leaf abscission, and wilting, followed by complete defolia tion or collapse. TSA plants inoculated with 30B developed mosaic (Figure 9-4), but not local lesions, leaf abscission, or necrosis. TSA plants inoculated with TMV developed mosaic w ith a different appearance but not local lesions, leaf abscission, or necrosis. Nicotiana sylvestris plants inoculated with PV-0113 developed local
170 lesions (Figure 9-5), while N. sylvestris plants inoculated with TMV did not develop local lesions, but newly emergent leaves develope d mosaic and leaf di stortion (Figure 9-6). Western blotting (Figure 9-8) revealed tobamovirus CP accumulation in plants inoculated with LAU2, LAU5, PV-0113, TMV, and 30B. Good transient expression of wtGFP from agroinfiltration was accomplished with pCASS[+wtGFP] in AGL0 on N. benthamiana plants only. Fluorescence for wtGFP was seen both with and without the use of gene silencing suppressors, however gene silenc ing suppressors, particularly 2b, clearly enhanced expression (as measured by observed brightness) and persistence. Tropical soda apple plants inoculated with 30B GFPc3 infected tissue developed green fluorescent patches in the area of inoculation (Figure 9-9) that expanded, but usually not beyond the inoculated leaf. Eventually, 30B GFPc3 inoc ulated TSA plants developed systemic mosaic symptoms similar to those seen in 30B inoculat ed plants. Furthermore a chimeric tobamovirus provided by Dr. Shailaja Shivprasad iden tified as TTU2T-GFP, containing the 5 NTR, replicase, and movement genes of TMV, the CP subgenomic promoter and the CP of TMGMV-S, and the 3 NTR of TMV produced systemic symptoms only on inoculated TSA plants, but not HR/lethal systemic hypersensitive respons e (LSHR; data not shown). Nicotiana sylvestris plants inoculated with 30B developed local lesions with much the same appearance (Figure 9-7) as those induced by PV-0113. As with natural GFP (excita tion = 395-475 nm, emission = 510), GFPc3 can be visualized under a blacklight (Figure 9-10), while eGFP, th e red-shifted (488/509 nm) variant employed in the Gateway-compatible expression vectors, must be observed under the epifluorescent microscope with a 485/20 excitation filter and a 530/25 emission filter (Bio-Tek, 2004;
171 Olympus, 2004). Fluorescence was observed under th e microscope in both GFPc3-inoculated N. benthamiana (Figure 9-11) and TSA (Figure 9-12). TSA plants inoculated with the chimeric c onstruct 30B GFP expressed GFP, but did not develop delayed HR, and went on to develop a mosa ic after a period of ti me. Expression of GFP did not coincide with the mosaic, suggesting that the systemic virus consisted mostly of recombinants. No GFP-related fluorescence was observed on any TSA infiltrated with A. tumefaciens Fluorescence was observed on all (5/5) N. benthamiana and 1/3 N. sylvestris plants infiltrated with AGL0. This fluorescence was enhanced (muc h brighter) in the plan ts co-infiltrated 2b, where the period of expression was extended from 2 to 4 days post-inoculation (DPI) to nearly a month. Some N. sylvestris plants infiltrated with A. tumefaciens strain ABI carrying expression construct pE1-D2 developed necros is throughout the infiltrated leaf panels (Figure 9-13). No CP was detected in any of the Agrobacterium -infiltrated TSA or tobacco tissues by Western blotting (data not shown). Wild-type, pathogenic A. tumefaciens caused gall formation and other effects (i.e. adventitious root formation, pe rsistence of chlorophyll in se nescent leaves) on inoculated N. sylvestris and N. tabacum but not on TSA. On TSA, A. tumefaciens caused chlorosis, necrosis, or leaf abscission, but not gall formation. Bombardment with gold particle s caused considerable trauma to the leaves: at pressures and distances required to infiltrate TSA, N. benthamiana leaves were destroyed. Under the microscope, gold particles were observed embedded in sectioned host tissues (data not shown).
172 Tissue bombarded by uncoated control particles exhibited yellow-green fluorescence similar to that observed in treatment groups (data not shown). Discussion Based on the host response to inoculation with TMGMV PV-0113 and to 30B, it appears that TMGMV CP alone is not the elicitor of the HR-like response in the disease process observed in the TSA. Both PV-0113 and 30 B elicit local le sions typical of HR in the CP-sensitive host, N. sylvestris and the sequences of the CP of both PV0113 and 30B are nearly identical. Culture LAU5 also consistently produces necrotic sympto ms on TSA, but not letha lity. This culture was later demonstrated to contain STMV by serolo gical methods (data not shown). While PV-0113 produced local lesions in N. sylvestris and lethal response in TS A, TMV moves systemically through both hosts without causing necrosis or lethality. The chimera 30B, expressing TMGMV-L CP only, behaves similarly to TMV: moving systemically through TSA without causing necrosis. Serology (Western blotting) demonstrates a subs tantial accumulation of CP in 30B-infected TSA plants. Taken together, th ese data indicate that TMGMV CP is not responsible for the disease process produced by PV-0113 in TSA. Unlikely, but possible scenarios nullifying this hypothesis include: 1) so mething about 30B may cause a difference in timing of the infection cycle di srupting elicitor rec ognition (perhaps it is unavailable for recognition by host detection following aggregat ion); and 2) tobamovirus CP may form a complex with other TMGMV proteins in planta This protein complex may itself be what is recognized by the elicitor. The limited development of green fluorescen ce on TSA plants inoculated with 30B GFPc3-infected sap are consistent with the form ation of recombinants in tobamovirus vectors reported previously (Shivprasad et al., 1999; Dawson et al., 1989).
173 While expression of wtGFP was clearly enhan ced by the presence of the gene silencing suppressors, the system appears optimized to function only in N. benthamiana Only 1 out of 3 infiltrated N. sylvestris plants responded with fluorescence. Additionally, while both N. sylvestris and N. tabacum responded to inoculation with pa thogenic wild t ype strains of A. tumefaciens with galls and other symptoms, TSA was i mmune. Infiltration and injection of TSA with wild type A. tumefaciens cultures produced no gall form ation, or other effects (e.g., persistent green areas on sene scent leaves) seen after infilt ration, injection, and toothpickinoculations on tobacco species ( N. sylvestris and N. tabacum ). This suggests that the strains of A. tumefaciens used are not efficiently transforming TS A, and the procedure must be optimized until reliable transformation/transient expression in TSA with a strain of Agrobacterium can be achieved. It appears that different strains of A. tumefaciens work better with different hosts in binary vector expression systems (Dr. Greg Martin, Professor, Plant Pathology Department, Cornell University, NY). A strain of Agrobacterium pathogenic to TSA should be identified before future attempts at agroin filtration are attempted. There exis ts a report of hairy root culture of Solanum aculeatissimum, a species closely related to TSA (Jacob and Malpathak, 2004). Future work with Agrobacteria should follow the procedures outlined in this paper and the strain of A. rhizogenes used should be considered. Cotyledons and primary TSA leaves were easie st to infiltrate. Growth-chamber-grown plants tended to have fewer and more flexible prickles than greenhouse grown plants. Finally, greenhouse-grown TSA plants in filtrated with 10 mM MgSO4 buffer solution showed tissue collapse and bleaching in the infiltrated ar eas, necessitating a switch to 10 mM MgCl2 buffer. Transient expression of CP by pE1-D2 should be verified: The vector should be agroinfiltrated into an CP insensitiv e, yet permissive host such as Arabidopsis thaliana along with a
174 gene-silencing suppressor such as 2b, and po ssibly wtGFP to measure timing of expression, harvested, and tested using ELISA or Western bl otting. Western blotting appears useful to demonstrate transgenic CP expr ession, but perhaps ELISA would wo rk equally well as a protein detection method. For the other gene products, some other method of de monstrating transgene expression will be required, unless antiserum sp ecific to the TMGMV 12 6 and 183 kDa proteins and MP can be obtained. The visible synergy between gene silencing suppressor 2b and GFP suggest that revealing co-infiltration experiments may be possible (i.e., wtGFP and 2b might be infiltrated along with vectors carrying PV-0113 ORF3 and ORF4, compared to wtGFP and 2b along with the vector carrying an unrelated insert ). Transient expression of MP by pE1-A2 will be more difficult to establish without MP-sp ecific antiserum, but could perhaps be achieved using a dye of the right molecular weight, or by inoculating a small region of an infiltrated leaf panel on the host Eryngium planum with 30B GFPc3 and then looking for systemic movement of the virus. To summarize, when transient expression in an undomesticated host such as TSA is desired, expression of construct must first be demonstrated in a permissive host such as N. benthamiana then strains of A. tumefaciens must be screened against the (less-permissive) host plant to find a strain that will infect it. Using a gene gun had some drawbacks. Plants to be tested had to be produced in peat pots, to which not all species were adaptable. Pl ants in 2 pots literally had soil blown out of pots by the treatment, contaminati ng the entire field. Furthermor e, it was difficult to calibrate, there needs to be a variety of distances and/or gas pressures are required to achieve correct depth of tissue penetration. Particle bombardment n ecessarily causes trauma to the host tissue. Density of tissue varies from plant type to plant type. Nicotiana benthamiana tissue is so
175 succulent that it is actually destroyed at pres sure/vacuum settings required to embed particles in TSA. It would be best to use the cotyledons and heart-shaped primary leaves of TSA for any future experiments requiri ng biolistic infiltration. Evidence of successful particle delivery was observed under the microscope gold particles were seen embedded in hos t tissues (data not shown). It st ands to reason, therefore, that particles carrying DNA were getting into the cells. Extensive tissue trauma, however, may obscure the marker gene -tissue bombarded by unc oated control particles also exhibited yellowgreen fluorescence (data not s hown). Detection of eGFP, unlike GFPc3, requires special epifluorescent filter set. Perhaps the GUS-trans formed VIB vector (E1-gu s) could be used for calibration, as the blue pigment will be easier to visualize. Phenolic compounds released during wounding, particularly in TSA appear to fluoresce under the UV microscope, potentially obscuring eGFP. If projectile delivery is to be us ed in the future, it would be better to use galacuronidase as a marker, or perhaps a different color fluorescent protein such as blue or cyan. In conclusion, while all three methods (chime ric viruses, agroinfiltration, and particle bombardment) have the potential to express individua l PV-0113 proteins in planta they differ in their ease of application. The chimeric virus ap proach seems the easiest to apply, once the virus has been created. Agroinfiltration using binary vectors requires some optimization to the host, but is also facile. Particle bom bardment is a relative ly crude method that requires calibration and damages the plant tissue. It is perhaps better suited to transformations than transient expression.
176 Table 9-1. Tobacco mild green mosaic virus-containing tobamovirus chimeras. Identifier Component Viruses Author Toc-J ToMV/TMGMV-J Hori and Watanabe, 2003. Toc-J/GFP ToMV/TMGMV-J/GFP Hori and Watanabe, 2003. 30B TMV/TMGMV-L Shivprasad, et al. 1999. 30B-GFP(C3) TMV/TMGMV-L/GFP( C3) Shivprasad, et al. 1999. V23 TMV/TMGMV-S Nejidat, et al. 1990. Table 9-2. Bim-Lab-inoculated treatment plants. Plasmid construct (in ABI, AGL0) Infiltrated Host pCASS [+2b, +GFP] Nicotiana benthamiana (small) N. benthamiana (large) N. sylvestris pE1-A2 (PV-0113 ORF3) Nicotiana benthamiana N. sylvestris Solanum viarum (small TSA) S. viarum (large TSA) pE1-D2 (PV-0113 ORF4) Nicotiana benthamiana N. sylvestris Solanum viarum (small TSA) S. viarum (large TSA)
177 Figure 9-1. Map of TMV/TMGMV chimera, 30B (Shivprasad et al., 1999).
178 Figure 9-2. Nucleotide alignment between 30B CP domain and PV-0113 ORF4.
179 Figure 9-3. Amino acid alignment between 30B CP and PV-0113 CP.
180 Figure 9-4. Symptoms on TS A leaf following inoculation with the 30B construct.
181 Figure 9-5. Local lesions on N. sylvestris following inoculation with TMGMV PV-0113.
182 Figure 9-6. Symptoms (mosai c, leaf distortion) on Nicotiana sylvestris following inoculation with TMV.
183 Figure 9-7. Local lesions on Nicotiana sylvestris following inoculation w ith the 30B construct.
184 Figure 9-8. Western blot showing CP accumulation in TSA. 1) Empty lane, 2) 30B, 3) water blank, 4) LAU5, 5) LAU2, 6) TMGMV PV -0113, 7) healthy tissue, 8) TMV, 9) protein standard ladder, and 10) empty lane. A B Figure 9-9. Solanum viarum inoculated with 30B GFPc3. A) Uninoculated (healthy) TSA viewed under UV light (glowing objects in so il are perlite particles). B) TSA inoculated with 30B-GFPc3 s howing fluorescent areas. 1 2 3 4 5 6 7 8 9 25 kDa 20 kDa 10
185 A B Figure 9-10. Nicotiana benthamiana inoculated with sap containi ng infectious 30B GFPc3. A) Viewed under fluorescent light and B) under UV illumination.
186 A B C D Figure 9-11. Epifluorescent microsc opy of TMV/TMGMV chimera 30B-GFPc3 in Nicotiana benthamiana A) Healthy tissue viewed under visi ble light B) Healthy tissue viewed under UV light C) Inoculated tissue under vi sible light D) Inoculated tissue under UV light.
187 A. B C D Figure 9-12. Epifluorescent microscopy of TMV/TMGMV chimera 30B-GFPc3 in TSA. A) Healthy tissue viewed under visible light. B) Healthy tissue view ed under UV light. C) Inoculated tissue under visible light. D) Inoc ulated tissue under UV light.
188 A B C Figure 9-13. Necrotic areas on N. sylvestris in response to infiltrati on with pE1-D2. A) ABI [pE1-gus] ( -galacuronidase). B) ABI [pE1-A2] (PV-0113 MP), C) ABI [pE1-D2] (PV-0113 CP).
189 CHAPTER 10 GENERAL DISCUSSION Tropical soda apple (TSA) remains a major w eed problem in the Southeastern United States and in many countries around the world. Outs ide of Florida, TSA is expected to continue to spread into more temperate areas. The plan ts regenerate from the r oots after a hard frost (Mullahey et al., 1998), and, ind eed the range of TSA in South America includes areas that are not strictly tropical, but subt ropical. Recent studies have shown that the promising new herbicide, aminopyralid, (marketed as Milesto ne, and, in combination with 2-4 D, as Forefront), which is well-translocated in most weeds, is not well translocated within TSA and requires complete and thorough coverage at a rate 0.11% to be effective (Ferrell and Sellers, 2007). In contrast, inoculation of a single le af with a few micrograms of the bioherbicide Tobacco mild green mosaic virus (TMGMV) is us ually effective. The bioherbicide activity is then amplified until the host TSA is dead. As presented in this thesis, the Columnea isolate PV-0113 is the small-type TMGMV [TMGMV-S (based on the size of the 3 end of the genomic RNA)], and does not appear to contain Satellite tobacco mosaic virus (STMV) TMGMV PV-0113 appears to be a relatively stable quasi-species, without pat hogenic variants or c ontaminants. Open reading frames (ORFs) 3 (movement) and 4 (coat protein) of PV0113 are highly similar to the corresponding ORF sequences reported for other TMGMV strains. Ho wever, mutations resulting in a single amino acid substitution in toba movirus proteins have been demons trated to attenuate necrotic host response (Lewandowski and Dawson, 1993; Hamamo to et al., 1997; Hagiwara et al., 2002; Divki et al., 2004). Not all cultu res of TMGMV are lethal to TSA. TSA plants inoculated with (ATCC PV-586), which is reported to cont ain large-type TMGMV (TMGMV-L) and STMV, exhibited hypersensitive respons e (HR)-like symptoms but recove red with systemic symptoms.
190 Because of the unusual number of variables (for a tobamovirus) applicable to TMGMV, each culture used in experimentation mu st be carefully evaluated before use to determine whether it is large or small-type TMGMV, and whether or not STMV is present. Spurious and inconsistent results in experiments using cult ures obtained from the CREC in La ke Alfred, Florida are likely related. In the case of U2, contamination or replacement by another tobamovirus and, in the case of U5 to the presence of STMV (which ma y have been minor in the initial sample but accumulated to higher levels in inoculated plants) may have occurred. A simplified protocol for purification of TMGMV appears about as effective as the standard one. There is a re duction of pathogenicity upon lypohilization and storage. There is also a reduction of pathogenicity in unlyophilized virus stored over a long period of time. Virus held in storage (above 0C) retains enough poten cy to produce local lesions (LL) and kill TSA for at least several months. TSA is highly se nsitive to purified PV -0113 which displays a dilution curve typical for a tobamovirus. Cross-protection may prove an obstacle to de velopment of PV-0113 as a bioherbicide. While the mechanism for cross-protection is not well understood, it is possible that several tobamoviruses will cross-protect TSA against TMGM V. The possibility that STMV, as well as TMV both prevent lethality thr ough some cross-protection mechan ism is noteworthy. STMV CP is quite dissimilar from that of TMV and TM GMV, making simple competition between CPs unlikely. TMGMV and STMV share common elements in the 3 end, including multiple pseudoknots and two nearly identi cal 40-50 base domains (Dodds, 1998) It is possible that the mechanism of cross-protection depends on these elements. Although STMV does not appear to interfere with the TMGMV-elicit ed HR in the TSA, it is not known whether the subsequent recovery of the plants inoculated with PV586 is due to the presence of STMV or is a
191 characteristic of TMGMV-L. To determine th is, an STMV-free culture of TMGMV-L will be required. Another option would be to introdu ce STMV into a subcul ture of PV-0113 and inoculate TSA to observe any changes in disease progress. As an obligate parasite, it is not advantageous for TMGMV to kill its host. Unless related to some function enhancing transmission, (e.g. ence phalitis-related behavioral changes seen in rabies), death of the host reduces the reproductive fitness of the virus. As ho rrific as the effects of a mortal disease such as Ebola may appear to us, they must ulti mately be viewed as failure of the infectious agent to negotiate a more commensa l relationship with its host. As an example of such a failed host-pathogen in teraction, the TSA-TMGMV diseas e process may serve as a good model to investigate what is probably lethal systemic HR. Indeed, perhaps a better question might be: if a virus that cause s local lesions on a host goes on to produce systemic mosaic (e.g., TMGMV PV-586 on TSA), why does the entire plan t not become necrotic? or, put in another way, Why is the trailing HR form of lethal sy stemic HR not the norm? Either the HR in these cases must be initiated by some protein onl y transiently present du ring the virus life cycle (in this hypothesis, you would expect not to see it in R-genes responding to CP, which accumulates) OR successful viruses have some mechanism (HR-silencing?) by which they shut down the host defense mechanisms. Further study of the interaction between TSA and TMGMV PV-0113 might also reveal more about R-gene genetics and e volution. While it seems plausible, even probable that there is a gene-for-gene interaction occurring between TSA and TMGMV, it is also possible that multiple genes are involved, or multiple host fact ors. Replication, movement, host-range, and symptomatology are all under the cont rol of the virus genome. It ap pears likely that sequences in the tobamovirus genome must perform multiple func tions to co-ordinate such events (Dawson et
192 al., 1988). Recent phylogenetic work examining both nuclear and chloroplast DNA has grouped TSA ( Solanum viarum ) in a clade with S. aculeatissimum S. myriacanthum and S. incarceratum (Levin et al., 2005, 2006). A study of the reaction of these species to inoculation with PV-0113 might prove fruitful if a host showing no re sistance (systemic infection with no HR) or localization (LL only) can be hybrid ized with TSA and the reaction of the offspring analyzed. In addition, it appears that TSA has been su ccessfully hybridized with the eggplant, S. melongena (Daunay, 1999), members of which may respond to PV-0113 inoculation with either LL or symptomless systemic infection (unpublished data). Indeed, there is evidence to s uggest that heterozygosity of re sistance alleles contributes to LSHR, at least in some interactions. Holmes (1937), describes the response of Capsicum species to inoculation with a tobamovirus, (most likely TMV), non-resistan t (susceptible) hosts developed systemic symptoms, while resistant host s developed LL, usually confining the virus to the inoculated leaf. When hybridized, the F1 offspring showed a severe systemic necrosis resulting in the death of the plant. Holmes we nt on to demonstrate the resistance allele was inherited in a Mendellian fashion (Holmes, 1937). Similarly, the I gene of Phaseolus vulgaris may produce extreme resistance (no visible symptoms), hypersensitive resistance (LL), or systemic vascular necrosis in re sponse to Bean common mosaic virus (BCMV), depending upon ge ne dosage and environmental conditions. Plants heterozygous for I often develop a necrosis of the phloem when inoculated with BCMV and grown at a high temperat ure (Collmer et al., 2000). The effect of heterozygosity on disease phenot ype might be related to lower expression of the R-gene product. In the sink hypothesis translation of a protei n is balanced with catabolism. That is, the constant breakdown of proteins within the cell ma y require a certain equilibrium
193 between the R-gene product, the pathogen elic itor, and the complex between them at which programmed cell death (PCD) can o ccur in time to confine the pat hogen to the site of infection. The prospect of using chimeric tobamovirus ve ctors for elucidating the elicitor of HR in TSA remains attractive, especially following th e development of a full-length clone of PV-0113, by Dr. Ernest Hiebert (GenBank accession EF469769). In addition to determining the elicitor of HR, such chimeras might be useful, along with natural TMGMV, in examining the physiology of the response. Aside from various marker ge nes, virus progress and accumulation might be visualized by terminal deoxyribonucleotidyl tr ansferase mediated de oxyuridine triphosphate nick and labeling (TUNEL) assay or the possibly easier agarose electrophoresis of total DNA to look for laddering (Xu and Roossinck, 2000). Since binary vectors incorporat ing ORF3 (ex. pE1-A2) have al ready been created, it might be productive to re-evaluate these, especially if an agroinfection system efficient at transforming TSA could be established. While th ere are different ways to deliver genetic material to plants to get transient expression for analysis of the hostpathogen interaction, some approaches are better suited than others. Biolistic infiltration is messy and cumbersome. Agroinfiltration shows potential, but must be optimized to work with TSA. The chimeric virus approach, while requiring technical sophistication, is probably the most elegant. While the discovery of TMGMV PV-0113 mo rtality on TSA is a lucky find (Drs. Raghavan Charudattan and Ernest Hiebert, prof essors, Plant Pathology Department, University of Florida, personal communications) perhaps viru s-induced lethality is not so uncommon, just merely unnoticed. When inoculated mechanically, Nicotiana tabacum (tobacco) is known to support several species of Tobamovirus it is not normally associated with, so it may not be
194 unreasonable to expect an exotic tobamovirus to produce disease in, or even lethal systemic HR in a problematic solanaceous weed. Several relatives of TSA are listed as noxious weeds. In addition to TSA, two other Solanum species, wetland nightshade ( S. tampicense ) and turkey berry ( S. torvum ) are of concern in Florida (Cuda et al., 2002). To screen these and other weedy/invasive species of Solanum using tobamoviruses to see if any exhibit a similar lethal sy stemic HR might prove useful, especially in a situation in which much applie d science the same applied science knowledge accumulated in working with PV-0113 could be br ought to bear with a similar tobamovirus, preferably a tobamovirus of limited or no econom ic importance (e.g., Frangipani mosaic virus, FrMV). Candidate tobamoviruses are conti nually being discovered. A new tobamovirus, Tropical soda apple mosaic virus (TSAMV), wa s recently discovered associated with TSA in Florida (Adkins et al., 2007). After the large investment of time and rese arch into methods of isolating, screening, producing, processing, purifying, applying and evaluating the TMGMV tobamovirus, it seems reasonable to proceed to evaluation of other toba movirus species as biological controls against other solanaceous pests. The be st candidates for evaluation would probably be those of minor or no economic importance, or those for applicat ion in regions in which economically or environmentally important host species do not occur. Although reasons for the tendency for host susceptibility to plant viruses to run very roughly along family lines remain unknown, it is wo rth noting that new tobamoviruses such as TSAMV and Streptocarpus flower breaking virus (SFBV) (Heinze et al., 2006) continue to be discovered, possibly explaining bot h susceptibility in (e.g., the ge sneriads) and resistance in (e.g., Gomphrena ) species that would otherwise appear to be erratic hosts. It is tempting to
195 speculate that a tobamovirus ge ographically removed, or infecti ng a more distantly related host might be more likely to result in a maladaptive host response leading to mortality, based on that hypothesis. In the future, it may be possible, using site-d irected mutagenesis or other methods to create tobamoviruses tailor-made for use as a bioherbicid e against a particular host. Such use would, however, require a thorough unders tanding of the host and perhap s modifications (e.g., knockout of CP) that would prevent their escape into the environment. Based on the current state of knowledge, TMGMV seems to be an unusually complex species for a tobamovirus. With both large and sm all type variants being maintained in the wild, along with STMV (+) and STMV ( ) infections, the virus may have more evolutionary options than other tobamovirus species.
196 CHAPTER 11 SUMMARY AND CONCLUSIONS The objective of this research, to discover the way in which lethal plant-virus interactions occur in the tropical soda a pple (TSA)-Tobacco mild green mosaic virus (TMGMV) PV-0113 interaction, in the hope of harnessing this phenom enon to control other invasive plants with a species-specific biological c ontrol approach, remains unfulfi lled. Several avenues for investigation are now apparent however, including: physiol ogical studies using techniques designed to detect and locali ze programmed cell death (PCD). My work, however, has laid the foundation for the exploration of th e interaction between TSA and TMGMV, the accurate identification of the host and pathogen, is not as simple as it would seem. The identity of the pathogen PV0113, which was once classified as a strain of Tobacco mosaic virus (TMV) (Siegel and Wild man, 1954), is now known to be a small-type isolate of the species TMGMV (chapter 2). Sub-isolates of PV-0113 showed no variation in dis ease phenotype or in the progression of disease development in the hosts TSA and Nicotiana tabacum cv. Samsun ( nn ). Disease progress on TSA typically began around 5 days postinoculation with symptoms such as epinasty and local lesions on the inoculated leaves, followe d by wilting of the apical regions or collapse, and necrosis or abscission of both inoculated an d uninoculated leaves. Eventually, the stems of PV-0113 inoculated plants become necrotic and the entire plant dies (chapter 3). Reverse-transcriptase PCR has confirmed that the PV-0113 is a small-type isolate of TMGMV, and dsRNA extraction and immunodiffu sion using rabbit antiserum has confirmed that STMV is not present in this isolate (chapter 3). At least one culture of TMGMV, PV-586, a la rge type isolate containing STMV, produced symptoms on TSA that initially were sim ilar to those produced by TMGMV PV-0113, but TSA
197 plants inoculated with TMGMV PV-586 recovered with mosaic sy mptoms. No other species of tobamovirus besides TMGMV tested thus far has been found that causes death of the host, TSA (chapter 4). A strain of TMV discovered as a greenhouse contaminant, C-1, cross-protects TSA plants against TMGMV PV-0113 induced lethality. Th e non-tobamovirus, Cucumber mosaic virus (CMV), does not cross-protect TSA plants fr om TMGMV PV-0113, and plants die following the typical death phenotype (chapter 5). TMGMV PV-0113 could be conveniently purifi ed and the purified preparations had physical properties consistent with those of purified TMGMV (chapter 6). The dilution curve of TMGMV (flat at high concentration, steeply sloping at moderate concentration, and flat at low concentration) is consistent with the dilution curve of a tobamovirus. Disease in TSA is produced by TMGMV PV-0113 even at the dilution of 1:50,000 from 1 optical density at an ab sorbance of 260 nm (chapter 7). Coat protein (CP) and moveme nt protein (MP) genes cloned from the genome of TMGMV PV-0113 were sequenced and found to be highly similar to sequences of TMGMV on record with the National Center for Biotechnology Info rmation (NCBI). A nucleotide motif known to be common to tobamoviruses and characterist ic of the species TMGMV was found to be identical in isolate PV-0113 (chapter 8). TSA plants inoculated with the chimeric construct 30B, which expresses TMGMV coat protein, developed systemic mo saic symptoms, but not necros is or death, indicating that TMGMV CP alone is not an elicitor of hypers ensitive response (HR) in TSA (chapter 9).
198 LIST OF REFERENCES Abbink, T.E.M., Tjernberg, P.A., Bol, J.F., Li nthorst, H.J.M., 1998. Tobacco mosaic virus helicase domain induces necrosis in N gene -carrying tobacco in the absence of virus replication. Mol. Plant-Mi crobe Interact. 11, 1242-1246. Adkins, S., Kamenova, I., Rosskopf E.N., 2007. Id entification and characterization of a novel tobamovirus from Tropical soda appl e in Florida. Plant Dis. 91, 287-293. Agrios, G.N., 1997. Plant Pathology. 4th ed. New York: Academic Press. 635 p. Aguilar, I., Snchez, F., Ponz F., 2000. Di fferent forms of interference between two tobamoviruses. Plant Pathol. 49, 659-665. Akanda, R.U., Dowler, C.C., Mullahey, J.J., Shilling, D.G., 1997. Influence of postemergence herbicides on Tropical soda apple ( Solanum viarum ) and Bahiagrass ( Paspalum notatum ). Weed Technol. 11, 656-661. Albin, C.L., 1994. Non-indigenous pl ant species find a home in mined lands. In: Schmitz, D.C., Brown T.C. (Eds.) An assessment of invasive non-indigenous species in Floridas public lands. Department of Environmental Protec tion, Tallahassee, FL, US A. Technical Report TSS-94-100. p. 252-253. Alexandre, M.A.V., Soares, R.M., Rivas, E.B., Duarte, L.M.L., Chagas, C.M., Saunal, H., Van Regenmortel, M.H.V., Richtzehain L.J., 2000. Characterization of a strain of Tobacco mosaic virus from Petunia. J. Phytopathol. 148, 601-607. Alfieri, S.A., Jr., Langdon, K.R., Kimbr ough, J.W., El-Gholl, N.E., Wehlburg, C., 1994. Diseases and disorders of plants in Florid a. Florida Department of Agriculture and Consumer Services, Division of Plant Industry. Bulletin No. 14. PI93T-10. 1114 p. Allen, J.E. Kamenova, I. Adkins, S., Hanson S.F ., 2005. First report of Hibiscus latent Fort Pierce virus in New Mexico. Plant H ealth Progress. [cited 2007 July 1] http://www.plantmanagementnet work.org/pub/php/brief/2005/hlpv Anonymous, 2002. A global taxonomic resource for the Solanaceae Natural History Museum, London, UK. [cited 2002] http://internt.nhm.ac.uk/jdsml/botany/databases/solanum/ Anonymous, 2003. A virus worth sharing? In: Ha yes, L., (Ed.) Whats new in biological control? Landcare Research New Zealand Ltd. 24:6. Anonymous, 2004. Excitation and emission of green fluorescent protei ns: you gotta get your filter sets right. [cited 2004 Aug 4] http://www/biotek.de/products/ Anonymous, 2004. GRIN Taxonomy. USDA, ARS, National Genetic Resources Program. Germplasm Resources Information Network (GRIN) National Germplasm Resources Laboratory, Beltsville, Maryland. [cite d 2004 May 19] http://www.ars-grin.gov/
199 Anonymous, 2004. Applications: fluorescent prot eins and bioluminescent proteins. Olympus. [cited 2004 Aug 3] http://microscope.olympus.com/contentsDB/01world/01research/ Anonymous, 2005. Bim-Lab. Bio-Oz Biot echnologies Ltd. [cited 2007 Jul 4] http://www.biooz.co.il/products-bimfield.html Anonymous, 2007. Catalog. American Type Cultur e Collection (ATCC). Manassas, Virginia, USA. [cited 2007 Jun 23] http://www.atcc .org/common/catalog/wordSearch/results. Anonymous, 2007. Plant Virus Index. De utche Sammlung von Mikroorganismen und Zellkulturen GmbH (DSMZ). Brauns chweig, Germany. [cited 2007 Jun 23] http://www.dsmz.de/plant_viruses/ Anonymous, 2007. Sample labels for produc ts: Milestone and ForeFront R&P. DowAgroSciences. [cited 2007 Jun 16] http://www.dowagro.com/Label/index.htm Anonymous, 2007. Tobacco buyout information webs ite. College of Agriculture and Life Sciences, North Carolina State University. [cited 2007 Jun 19] http://www.cals.ncsu.edu/advancement/tobaccobuyout/ Aragao, F.J.L., Grossi de Sa, M.-F., Davey, M.R ., Brasileiro, A.C.M., Faria, J.C., Rech E.L., 1993. Factors influencing transient gene expression in bean ( Phaseolus vulgaris L.) using an electrical partic le acceleration device. Plant Cell Rep. 12, 483-490. Ashby, A.M., Watson, M.D., Shaw, C.H. 1987. A Ti-pla smid determined function is responsible for chemotaxis of Agrobacterium tumefaciens towards the plant wound product acetosyringone. FEMS Microbiol. Lett. 41, 189-92. Baker, C.A., Zettler F. W., 1988. Viruses inf ecting wild and cultivated species of the Commelinaceae. Plant Dis. 72, 513-518. Ball, E.M., 1964. Serology: Techniques used in pl ant virus research. In: Corbett, M.K., Sisler, H.D., editors. Plant Virology. University of Florida Press, Gainesville. p. 235-252. Bawden, F.C., 1956. Quantitative methods of assa ying for viruses. In: Plant Viruses and Plant Diseases, 3rd ed. Chronica Botanica Co., Waltham MA. p. 150-166. Bernard, P., Kzdy, K.E., Van Melderen, L., Stey aert, J., Wyns, L., Pato, M.L., Higgins, P.N., Couturier M., 1993. The F plasmid CcdB prot ein induces efficient ATP-dependent DNA cleavage by gyrase. J. Mol. Biol. 234, 534-541. Bodaghi, S., Yassi, M.N.A., Dodds, J.A., 2000. Hete rogeneity in the 3-terminal untranslated region of tobacco mild green mosaic tobamoviruses from Nicotiana glauca resulting in variants with three or six pseudoknots. J. Gen. Virol. 81, 577-586. Bodaghi, S., Mathews, D.M., Dodds, J.A., 2004. Natural incidence of mixed infections and experimental cross protection between two genot ypes of Tobacco mild green mosaic virus. Phytopathology. 94, 1337-1341.
200 Borovsky, D., Rabindran, S., Dawson, W.O., Po well, C.A., Iannotti, D.A., Morris, T.J., Shabanowitz, J., Hunt, D.F., DeBondt, H. L., DeLoof A., 2006. Expression of Aedes trypsin-modulating oostatic factor on the virion of TMV: a potential la rvicide. Proc. Natl. Acad. Sci. USA. 103, 18963-18968. Boubourakas, I. N., Hatziloukas, E., Antignus, Y., Katis, N. I., 2004. Etiology of leaf chlorosis and deterioration of the fru it interior of watermellon plants. J. Phytopathol. 152, 580. Bruce, W.B., Christensen, A.H., Klein, T., Fro mm, M., Quail P.H., 1989. Photoregulation of a phytochrome gene promoter from oat transfe rred into rice by particle bombardment. Proc. Natl. Acad. Sci. USA. 86, 9692-9696. Brunt, A.A., Crabtree, K., Dallwitz M.J., Gibbs, A.J., Watson, L. Zurcher, E.J., (1996 onwards). `Plant Viruses Online: Descriptions and Li sts from the VIDE Da tabase. Version: 20th August 1996' http://biology.anu.edu.au/Groups/MES/vide/. Burr, T.J., Otten, L., 1999. Crown gall of grap e: biology and disease management. Annu. Rev. Phytopathol. 37, 53-80. Canto, T., MacFarlane, S.A., Palukaitis P., 2004. OR F6 of Tobacco mosaic virus is a determinant of viral pathogenicity in Nicotiana benthamiana J. Gen. Virol. 85, 3123. Carrington, J.C., Kasschau, K.D., Mahajan, S. K., Schaad M.C., 1996. Cell-to-cell and longdistance transport of viruses in plants. The Plant Cell. 8, 1669-1681. Chagas, C.M., Grill, M., Noronha, A.B., Vin cente, M., 1978. Solanceas silvestres como hospedeiros experimentais de virus I Sola num ciliatum LAM., S. Palinacanthum Dun. E S. Viarum DUN. Arq. Inst Biol., Sao Paulo. 45, 177-182. Charudattan, R., Zettler, F.W., Cordo, H.A., Chri stie, R.G., 1976. Susceptibility of the Florida milkweed vine, Morrenia odorata to a potyvirus from Araujia angustifolia (Abstr. 319) Proc. Am. Phytopathol. Soc. 3, 272. Charudattan, R., Cordo, H.A., Silveira-Guido, A ., Zettler, F.W., 1978. Oblig ate pathogens of the milkweed vine, Morrenia odorata as biocontrol agents. In: Freeman, T.E. (Ed.) Proceedings of the 4th International Symposium of Bi olological Control of Weeds. University of Florida, Gainesville. p. 241. Charudattan, R., 1979. Charudattan, R., invent or and assignee. 1978 Feb 7. Composition and process for controlling milkweed vine. U.S. patent 4,162,912. Charudattan, R., DeValerio, J.T., 1996. Biol ogical control of tropical soda apple, Solanum viarum using plant pathogens. In: Mullahery, J.J. (Ed.) Proceedings of the Tropical Soda Apple Symposium, Jan. 9-10, 1996, Bartow, FL Institute of Food and Agricultural Sciences, University of Florida, Gainesville, FL, USA. p. 73-78.
201 Chenna, R., Sugawara, H., Koike, T., Lopez, R., Gibson, T.J., Higgins, D.G., Thompson J.D.. 2003. Multiple sequence alignment with the Clusta l series of programs. Nucleic Acids Res. 31, 3497-3500. Cohen, J., Rosner, A., Kagan, S., Lampel, M., Maslenin, L., Zeidan, M., Gera A., 2001. A new disease in Tabernaemontana associated with Tobacco mild green mosaic virus. Ann. Appl. Biol. 138, 153-159. Coile, Nancy C., 1993. Tropical Soda Apple, Solanum viarum Dunal: The Plant from Hell (SOLANACEAE). Fla. Dept. Agric. & Consum er Services, Division of Plant Industry. Botany Circular No. 27. Collmer, C.W., Marston, M.F., Taylor, J.C., Jahn M., 2000. The I gene of bean: a dosagedependent allele conferring extreme resistance, hypersensitive resi stance, or spreading vascular necrosis in response to the potyvi rus Bean common mosaic virus. Mol. PlantMicrobe Interact. 13, 1266-1270. Conti, M., Marte, M., 1983. Vi rus, virosi, e microplasmosi de l pepperone. Ital. Agric. 12c, 132. Corpet, F., 1988. Multiple sequence ali gnment with hierarchical clustering Nucleic Acids Res.16, 10881-10890. http://bioinfo.ge nopole-toulouse.prd.fr/multalin/ Cuda, J.P., Gandolfo, J.C., Medal, J.C., Cha udatan, R., Mullahey, J.J., 2002. Tropical Soda Apple, Wetland Nightshade, and Turkey Berry. In: Biological Control of Invasive Plants in the Eastern United States. USDA Forest Service publication FHTET-2002-04. Culver, J.N., Lindbeck, A.G.C., Dawson, W.O., 1991. 15 Virus-host interactions: induction of chlorotic and necrotic responses in plants by tobamoviruses. A nnu. Rev. Phytopathol. 29, 193-217. Culver J.N., Dawson W.O., 1991. Tobacco mosaic vi rus elicitor coat protein genes produce a hypersensitive phenotype in transgenic Nicotiana sylvestris plants. Mol. Plant-Microbe Interact. 4, 458-463. Dardick, C.D., Culver, J.N., 1997. Tobamovirus coat proteins: elicitors of the hypersensitive response in Solanum melongena (Eggplant). Mol. Plant-Microbe Interact. 10, 776-778. Daunay, M.-C. Dalmon, A. Lester, R.N. 1999. Management of Solanum species for eggplant ( Solanum melongena ) breeding purposes. In: M. Nee, D.E. Symon, R.N. Lester, J.J.P., editors. Solanaceae IV: Royal Bota nic Gardens, Kew. p. 369-383. Dawson, W.O., Lehto, K.M., 1990. Regulation of toba movirus gene expression. Adv. Virus Res. 38, 307-342. Dawson, W.O., Beck, D.L., Knorr, D.A., Granth am, G.L., 1986. cDNA cloning of the complete genome of tobacco mosaic virus and production of infectious transcript s. Proc. Natl. Acad. Sci. USA. 83, 1832-1836.
202 Dawson, W.O., Bubrick, P., Grantham, G.L., 1988. Modifications of the Tobacco mosaic virus coat protein gene affecti ng replication, movement, and symptomatology. Phytopathology. 78, 783-789. Dawson, W.O., Lewandowski, D.J., Hilf, M.E., B ubrick, P., Raffo, A.J., Shaw, J.J., Grantham, G.L., Desjardins, P.R., 1989. A tobacco mosaic virus-hybrid expresses and loses an added gene. Virology. 172, 285-292. Dawson, W.O., Tobamovirus-plant intera ctions: minireview. 1992. Virology. 186, 359-367. de la Cruz, A., Lpez, L.Tenllado, F., Daz-Ru z, J.R., Sanz, A.I., Vaquero, C., Serra, M.T., Garca-Luque, I., 1997. The coat protein is required for the elicitation of the Capsicum L2 gene-mediated resistance against the tobamovi ruses. Mol. Plant-Microbe Interact. 10, 107113. Delgado-Sanchez, S., Grogan, R. G., 1970. Potato virus Y. CMI/AAB Descriptions of Plant Viruses, Association of Applied Biologists: Institute of Horticultural Research, UK. Bulletin No. 37. Dinesh-Kumar, S.P., Tham, W.H., Baker, B.J., 2000. Structure-function analysis of the tobacco mosaic virus resistance gene N Proc. Natl. Acad. Sci. USA. 97, 14789-14794. Divki, Z., Salnki, K., Balzs, E., 2004. The necr otic pathotype of the Cucumber mosaic virus (CMV) Ns strain is solely determined by amino acid 461 of the 1a protein. Mol. PlantMicrobe Interact. 17, 837-845. Dodds, J.A., 1998. Satellite tobacco mosaic virus. Annu. Rev. Phytopathol. 36, 295-10. El-Gholl, N. E., Schubert, T.S., Coile, N.C., 1997. Diseases and Disorders of Plants in Florida. Bulletin No. 14, Supplement No. 1. 138 p. Epstein, A.H., Hill, J.H., Nutter, F.W., Jr., 1997. Augmentation of rose rosette disease for biocontrol of multiflora rose ( Rosa multiflora ). Weed Sci. 45, 172-178. Ferrell, J.A., Sellers B.A., 2007. Developing spot -spray reccomendations for TSA control with aminopyralid. 30th Annual Florida Weed Science So ciety Meeting, 2007 in Maitland, FL. Ferrell, J.A., Mullahey, J.J., Langeland, K.A., Klin e, W.N., 2006. Control of tropical soda apple ( Solanum viarum ) with aminopyralid. Weed Technol. 20, 453-457. Fillhart, R.C., Bachand, G.D., Castello, J.D., 1998. Detection of infectious tobamoviruses in forest soils. Appl. Environ. Microbiol. 64, 1430-1435. Finer, J.J., Vain, P., Jones, M.W., McMullen, M.D., 1992. Plant Cell Reports. 11, 323-328. [FLEPPC] Florida Exotic Pest Pl ant Council. 2005. List of Florid a's Invasive Species. [cited 2007 Jun 25] http://www.fleppc.org/05list.htm
203 Fraile, A., Malpica, J.M., Aranda, M.A., Rodr guez-Cerezo, E., Garca-Arenal F. 1996. Genetic diversity in Tobacco mild green mosaic tobamovirus infecting the wild plant Nicotiana glauca Virology. 223, 148-155. Fraile, A., Escriu, F., Aranda, M.A., Malpica, J.M., Gibbs, A.J., Garca-Arenal, F. 1997. A century of Tobamovirus evolution in an Australian population of Nicotiana glauca J. Virol. 71, 8316-8320. Francki, R.I.B., Mossop, D.W., Hatta, T., 1979. Cucu mber mosaic virus. CMI/AAB Descriptions of Plant Viruses, Association of Applied Biol ogists: Institute of Ho rticultural Research, UK. Bulletin No. 213. Fuerstenau, S.D., Benner, W.H., Thomas, J. J., Brugidou, C., Bothner, B., Siuzdak, G., 2001. Mass spectrometry of an intact vi rus. Angew. Chem. Int. Ed. 40, 541-544. Fujiyama, K., Saejung, W., Yanagihara, I., Naka do, J., Misaki, R., Honda, T., Watanabe, Y., Seki, T., 2006. In planta production of imm unogenic poliovirus peptide using Tobacco mosaic virus-based vector sy stem. J. Biosci. Bioeng. 101, 398-402. Fukuda, M., Meshi, T., Okada, Y., Otsuki, Y., Takabe, I., 1981. Correla tion between particle multiplicity and location on virion RNA of the assembly initiation site for viruses of the tobacco mosaic virus group. Proc. Natl. Acad. Sci. USA. 78, 4231-4235. Gallie, D.R., Sleat, D.E., Watts, J.W., Turner, P.C., Wilson, T.M.A., 1987. Nucleic Acids Res. 15, 3257-3273. Gasteiger E., Gattiker A., Hoogland C., Ivanyi I., Appel R.D., Bairoch, A., 2003. ExPASy: the proteomics server for in-depth protein know ledge and analysis. Nucleic Acids Res. 31, 3784-3788. Available from: http://ca.expasy.org/ Gibbs, A.J., Armstrong, J.S., Gibbs, M.J., 2004. A type of nucleotide motif that distinguishes tobamovirus species more efficiently than nucleotide signatures. Arch. Virol. 149, 19411954. Gibbs, A., 1999. Evolution and origins of tobamoviru ses. Philos. Trans. R. Soc. Lond, Ser B:. 354, 593-602. Gibbs, A.J., 1977. Tobamovirus group. CMI/AAB Descri ptions of Plant Viruses, Association of Applied Biologists: Institute of Hort icultural Research, UK. Bulletin No. 184. Gibbs, C., 1986. Tobamovirus classi fication. In: Van Regenmortel, M.H.V., Fraenkel-Conrat H. editors. The Plant Viruses. Vol. 2. The rod-shap ed plant viruses. Plenum Press, New York. p. 167-180. Gilardi, P., Garcia-Luque, I., Serra, M.T., 1998. Pe pper mild mottle virus coat protein alone can elicit the Capsicum spp. L3 gene-mediated resistance. Mol. Plant-Microbe Interact. 11, 1253-1257.
204 Gilardi, P., Garcia-Luque, I., Serr a, M.T., 2004. The coat protein of tobamovirus acts as elicitor of both L2 and L4 gene-mediated resistance in Capsicum J. Gen. Virol. 85, 2077-2085. Giritch, A., Marillonnet, S., E ngler, C., van Eldik, G., Botterm an, J., Kilmyuk, V., Gleba, Y., 2006. Rapid high-yield expression of full-size Ig G antibodies in plants coinfected with noncompeting viral vectors. Proc. Natl. Acad. Sci. USA. 103, 14701-14706. Goelet, P., Lomonossoff, G.P., Butler, J.G., Ak am, M.E., Gait, M.J., Karn J., 1982. Nucleotide sequence of Tobacco mosaic virus RNA. Proc. Natl. Acad. Sci. USA. 79, 5818-5822. Gray, D.J. Hiebert, E., Lin, C.M., Compton, M. E., McColley, D.W., Harrison, R.J., Gaba, V.P., 1994. Simplified contstruction and performan ce of a device for particle bombardment. Plant Cell, Tissue Organ Cult. 37, 179-184. Hagiwara, K., Ichiki, T. U.,Ogawa, Y., Om ura, T., Tsuda, S., 2002. A single amino acid substitution in 126-kDa protein of Pepper m ild mottle virus associates with symptom attenuation in pepper; the co mplete nucleotide sequence of an attenuated strain, C-1421*. Arch. Virol. 147, 833-840. Hajdukiewicz, P., Svab, Z., Maliga, P ., 1994. The small, versatile family of Agrobacterium binary vectors for plant transfor mation. Plant Mol. Biol. 25, 989-994. Hajimorad, M.R., Eggenberger, A.L., Hill, J.H ., 2005. Loss and gain of elicitor function of Soybean Mosaic Virus G7 provoking Rsv1 -mediated lethal hypersensitive response maps to P3. J. Virol. 79, 1215-1222. Hamamoto, H., Watanabe, Y., Kamada, H., Okad a, Y., 1997. A single amino acid substitution in the virus-encoded replicase of Tomato mosaic tobamovirus alters host specificity. Mol. Plant-Microbe Interact. 10, 1015-1018. Heinlein, M., 2002. The spread of Tobacco mosaic virus infection: insights into the cellular mechanism of RNA transport. Ce ll. Mol. Life Sci. 59, 58-82. Heinze, C., Lesemann, D.-E., Ilmberger, N., Willingmann, P., Adam, G., 2006. The phylogenetic structure of the cluster of tobamovirus speci es serologically related to Ribgrass mosaic virus (RMV) and the sequence of Streptocarpus flower break virus (SFBV). Arch. Virol. 151, 763-774. Hellens, R., Mullineaux, P., Klee, H., 2000. Technical Focus: A guide to Agrobacterium binary Ti vectors. Trends Plant Sci. 5, 446-451. Hollings, M., Huttinga, H., 1976. Tomato mosaic virus. CMI/AAB Descriptions of Plant Viruses, Association of Applied Biologists: Institute of Horticultural Research, UK. Bulletin No. 156. Holmes, F.O., 1937. Inheritance of resistance to tobacco-mosaic disease in the pepper. Phytopathology. 27, 637-642.
205 Holmes, F.O., 1938. Inheritance of resistance to tobacco-mosaic disease in tobacco. Phytopathology. 28, 553-561. Hori, K., Watanabe, Y., 2003. Construction of a toba movirus vector that can systemically spread and express foreign gene products in Solanaceous plants. Plant Biotechnol. 20, 129-136. Hull, R., 2002. Matthews Plant Virol ogy. New York: Academic Press. 1001 p. Izhevsky, S.S., 1979. The applicatio n of pathogenic microorganisms for control of weeds in the U.S.S.R. Proceedings of the Joint American-Soviet Conference on Use of Beneficial Organisms in the Control of Crop Pests. Entomological Society of America. p.35 Jacob, A., Malpathak, N., 2004. Green hairy root cultures of Solanum khasianum Clarke a new route to in vitro solasodine production. Curr. Sci. 87, 1442-1447. Johnson, J., 1947. Virus attenuation and the se paration of strains by specific hosts. Phytopathology. 37, 822-837. Kado, C.I., Heskett, M.G., 1970. Selective me dia for isolation of Agrobacterium, Corynebacterium, Erwinia, Pseudomon as, and Xanthomonas. Phytopathology. 60, 969975. Kamenova, I., Rosskopf, E.N., Lewandowski, D. J., Adkins, S., 2004. Characterization of a tobamovirus from tropical soda apple. Phytopathology. 94, S48. Karimi, M., Inze, D., Depicker, A., 2002. Gateway vectors for Agrobacterium -mediated transformation. 2002. Trends Plant Sci. 7, 193-195. Kearny, C.M., Thomson, M.J., Roland, K.E., 1999. Ge nome evolution of t obacco mosaic virus populations during long-term pa ssaging in a diverse range of hosts. Arch. Virol. 144, 15131526. Knight, C.A., Silva, D.M., Dahl,D., Tsugita, A., 1962. Two distinctive strains of Tobacco mosaic virus. Virology. 16, 236. Koncz, C., Schell, J., 1986. The promoter of TL-DNA gene controls the tissue-specific expression of chimaeric genes carried by a novel type of Agrobacterium binary vector. Mol. Gen. Genet. 204, 383-396. Kreiser, B.R., Bryson, C.T., Usnick, S.J., 2004. Genetic variation in native and introduced populations of tropi cal soda apple ( Solanum viarum ). Weed Technol. 18, 1120-1124. Kucharek, T., Percifull D., Hiebert, E., 2003. Viruse s that have occurred naturally in agronomic and vegetable crops in florida. University of Florida EDIS. http://edis.ifas.ufl.edu/BODY_PG101
206 Kumagai, M.H., Turpen, T.H., Weinzettl, Della -Cioppa, G., Turpen, A.M., Donson, J., Hilf, M.E., Grantham, G.L., Dawson, W.O., Chow, T. P., Piatak, M. Jr., Grill, L. K., 1993. Rapid, high-level expression of biologically active -trichosanthin in transfected plants by an RNA viral vector. Proc. Na tl. Acad. Sci. USA. 90, 427-430. Laemmli, U. K., 1970. Cleavage of structural pr oteins during the assembly of the head of bacteriophage T4. Nature 227, 680-685. Lamb, E.M., Adkins, S., Shuler, K.D., Roberts, P.D., 2001. Pepper mild mottl e virus. IFAS Plant Pathology Fact Sheet PP-55. Lanfermeijer, F.C., Dijkhuis, J., Sturre, M. J.G., de Haan, P., Hille, J., 2003. Cloning and characterization of the durable toma to mosaic virus resistance gene Tm-22 from Lycopersicon esculentum Plant Mol. Biol. 52, 1037-1049. Lapido, J.L. Koenig, R., & Lesemann, D.E., 200 3. Nigerian tobacco latent virus: a new tobamovirus from tobacco in Nigeria. Eur. J. Plant Pathol. 109, 373-379. Lazo, G.R., Stein, P.A. Ludwig, R.A., 1991. A DNA transformation-competent Arabidopsis genomic library in Agrobacterium. Biotechnology. 9, 963-967. Leathers, V., Tanguay, R., Kobayashi, M., Gal lie, D.R., 1993. A phyllogenetically conserved sequence within viral 3 untranslated pseudoknots regulates translation. Mol. Cell. Biol. 13, 5331-5347. Letschert, B., Adam, G., Lesemann, D., Willin gmann, P., Heinze, C., 2002. Detection and differentiation of serologically cross-reacting tobamoviruse s of economical importance by RT-PCR and RT-PCR-RFLP. J. Virol. Methods 106, 1-10. Levin, R.A., Watson, K., Bohs, L., 2005. A four-gen e study of evolutionary relationships in Solanum section Acanthophora Am. J. Bot. 92, 602-612. Levin, R. A., Myers, N. R., Bohs, L., 2006. Phylogenetic relationships among the spiny solanums ( Solanum subgenus Leptostemonum Solanaceae). Am. J. Bot. 93,157-169. Lewandowski, D.J., Dawson, W.O., 1993. A single amino acid change in Tobacco mosaic replicase prevents symptom production. Mo l. Plant-Microbe Interact. 6, 157-160. Lewandowski, D.J., Dawson, W.O., 1999. Tobamoviruses. Academic Press. Li, C.Y., Chang Y.C., 2005. Fi rst report of Tobacco mild green mosaic virus on Capsicum annuum in Taiwan. BSPP. p.258. Lorenzi, H. 2000. Plantas daninhas do Brasil: terrestres, aquticas, parasitas, e txicas. 3rd Ed. Nova Odessa, SP: Instituto Plantaru m de Estudos da Flora LTDA. p.568. Malpica, J.M., Fraile, A., Moreno, I., Obies, C.I ., Drake, J.W., Garca-Arenal, F., 2002. The rate and character of spontaneous mutation in an RNA virus. Genetics. 162, 1505-1511.
207 Mathews, D.M., Dodds J.A., 1998. Naturally occu ring variants of tobacco mosaic virus. Phytopathology. 88 (6), 514-519. Mayers, C.N., Palukaitis, P., Carr, J.P., 2000. Sub cellular distribution anal ysis of the cucumber mosaic virus 2b protein. J. Gen. Virol. 81, 219-226. McGovern, R.J., Polston, J.E., Mullahey J.J., 1994. Solanum viarum : weed reservoir of plant viruses in Florida. Int. J. Pest Manage. 40 (3), 270-273. McKinney, H.H., 1929. Mosaic diseases in the Canary Islands, West Africa, and Gibraltar. J. Agric. Res. 39 (8), 557-578. McKinney, H.H., 1952. Two strains of tobacco mosaic virus one of which is seed-borne in an etch-immune pungent peppe r. Plant Dis. 36, 184. Medal, J.C., Gandolfo, D., Cuda, J.P., 2004. Biology of Gratiana boliviana the First Biocontrol Agent Released to Control Tropical Soda A pple in the USA. EDIS [Internet]. 2004. [cited 2007 Jul 3]; ENY 826. Available from: http://edis.ifas.ufl.edu/IN487 Meshi, T., Motoyoshi, F., Atsuko, A., Watana be, Y., Takamatsu, N., Okada, Y., 1988. Two concomitant base substitutions in the putativ e replicase genes of tobacco mosaic virus confer the ability to overcome the effects of a tomato resistance gene, Tm-1. EMBO Journal. 7 (6), 1575-1581. Metzer, M.L., Caskey, T.C., 2007 Polymerase chain reaction (PCR). En cyclopedia of Life Sciences, Nature Publications. [cited 2007 Jul 4] http://www.els.com Min, B.E., Chung, B.N., Kim, M.J., Ha, J.H., Lee, B.Y., Ryu, K.H., 2006. Cactus mild mottle virus is a new cactus-infecting toba movirus. Arch. Virol. 151, 13-21. Mislevy, P., Mullahey, J.J., Colvin, D.L., 1996. Ma nagement practices for tropical soda apple control: Update. In: Mullahey, J. J., (Ed.) Proceedings of the Tropical Soda Apple Symposium, Jan 9-10, 1996. Bartow, FL. Instit ute of Food and Agricultural Sciences, University of Florida, Gainesville, FL, USA. p. 61-67. Mislevy, P., J.J. Mullahey, Martin F.G., 1997. Tropi cal soda apple (Solanum viarum) control as influenced by clipping frequency and herbicide rate. Weed Science Society of America Abstracts. 37, 30. Morikawa, H., Iida, A., Yamada Y., 1989. Transien t expression of foreign genes in plant cells and tissues obtained by a simple biolistic device (particle gun ). Appl. Microbiol. Biotechnol. 31, 320-322. Morishima, N., Ido, T., Hamada, H., Yoshimoto, E., Mizumoto, H., Takeuchi, S., Kiba, A., Hikichi, Y., Okuno T., 2003. Infectious in vi tro transcripts from a cDNA clone of Tobacco mild green mosaic tobamovirus and its biologi cal activity in host and nonhost plants and in their protoplasts. J. Gen. Plant Pathol. 69, 335-338.
208 Morozov, S.Y., Denisenko, O.N., Zelenina, D.A., Fedorkin, O.N., Solovyev, A.G., Maiss, E., Casper, R., Atabekov, J.G., 1993. A novel open r eading frame in tobacco mosaic virus genome coding for a putative small, positively charged protein. Biochimie. 75, 659-665. Morris, T.J., Dodds, J.A., 1979. Isolation and an alysis of double-stra nded RNA from virusinfected plant and fungal tissue. Phytopathology. 69, 854-858. Muller, H.J., 1932. Some genetic asp ects of sex. The Am. Nat. 66, 118-138. Mueller, K., 2004. The Lycopersicon peruvianum complex: those species which are not easily crossed with Lycopersicon esculentum [cited 2004 Oct 7] http://www.kcinter.net/~tomato/Tomato/pcomplex.html Mullahey J.J., Shilling, D.G., Mislevy, P., Akanda R.A., 1998. Invasion of Tropical Soda Apple ( Solanum viarum ) into the U.S.: Lessons Learned. Weed Technol. 12, 733-736. Mullahey J.J., Colvin, D.L., 2002. Tropical Soda Apple: A New Noxious Weed in Florida. University of Florida IFAS. h ttp://edis.ifas.ufl.edu/BODY_UW097 Nagai, Y., Choi, Y-M., Tochihara H., 1987. TMVU, a new strain of Tobacco mosaic virus isolated from sweet pepper. Ann. Phytopathol. Soc. Jpn. 53, 540-543. Nejidat, A., Cellier, F., Holt, C.A., Gafny, R., Eggenberger, A. L., Beachy, R.N., 1991. Transfer of the movement protein gene between tw o tobamoviruses: infl uence on local lesion development. Virology. 180, 318-326. Olesen, K., 2003. pDRAW32. AcaClone Software. [cited 2007 May 23] http://www.acaclone.com Okuno, T., Hamada, H., Takeuchi, S., Morishima, N., Yoshimoto, E., Hikichi, Y., 2002. Tobacco mild green mosaic virus complete genome strain: Japanese. Direct submission to GenBank. Accession number AB078435. [cited 2007 May 21] http://www.ncbi.nlm.nih.gov Ozeki, J., Takahashi, S., Komatsu, K., Kagiwa da, S., Yamashita, K., Mori, T., Hirata, H., Yamaji, Y., Ugaki, M., Namba, S., 2006. A si ngle amino acid in the RNA-dependent RNA polymerase of Plantago asiatica mosaic virus contributes to systemic necrosis. Arch. Virol. 151, 2067-2075. Padgett, H.S., Beachy, R.N., 1993. Analysis of a t obacco mosaic strain capable of overcoming N gene-mediated reistance. Plant Cell. 5, 577-586. Paliwal, Y.C., Nariani, T.K., 1965. Propert ies of the inhibitors of sunnhemp ( Crotalaria juncea L.) mosaic virus in certain plant extracts. Acta Virologica 9:455-458. Palukaitis, P., Zaitlin, M., 1986. Tobacco mosaic virus: infectivity and replication. In: Van Regenmortel M.H.V., Fraenkel-Conrat H., ed itors. The Plant Viruses. Vol. 2. The rodshaped plant viruses. Plenum Press, New York. p. 105-126.
209 Parrella, G., Verdin, E., Gognalons, P., Marchou x, G., 2006. Detection and characterization of Tobacco mild green mosaic virus (TMGMV) la rge type isolate from trailing petunia in France. Communications in Agricultural a nd Applied Biological Sciences. 71(3b), 12371244. Pelham, J., 1966. Resistance in tomato to to bacco mosaic virus. Euphytica. 15, 258-267. Pettersen, M. S., Charudattan, R., Hiebert, E., Zettler, F.W., Elliott, M.S., 2000. Tobacco mild green mosaic tobamovirus strain U2 causes a lethal hypersensitive response in Solanum viarum Dunal (tropical soda apple). Weed Sc ience Society of America Abstracts. 40, 84. Pettersen, M.S., Charudattan, R., Hiebert, E., Ze ttler, F.W., 2001. Tobacco mild green mosaic virus (TMGMV) induces a lethal hypersensi tive response in tropical soda apple ( Solanum viarum Dunal). Phytopathology 91(Suppl.):S71(Abstr.). Pfitzner, U.M., Pfitzner, A.J.P., 1992. Expression of a viral avirulence gene in transgenic plants is sufficient to induce the hypersensitive defe nse reaction. Mol. Plan t-Microbe Interact. 5, 318-321. Pickersgill, B. 1988. The genus Capsicum: a mu ltidisciplinary approach to the taxonomy of cultivated and wild plants Biol. Zentralbl. 107, 381-389. Porter, M.B., MacKay, R.J., Uhl, E., Platt, S.R., de Lahunta, A., 2003. Neurologic disease putatively associated with ingestion of Solanum viarum in goats. J. Am. Vet. Med. Assoc. 223, 501-504. Purcifull, D.E., Batchelor D.L., 1977. Immunodiffus ion tests with sodium dodecyl sulfate (SDS)treated plant viruses and plant viral inclusions. Fla. Agric. Exp. Stn. Technical Bulletin 788. 39 p. Purcifull, D.E., Hiebert, E., 1982. Tobacco etch vi rus. CMI/AAB Descriptions of Plant Viruses, Association of Applied Biologists: Institute of Horticultural Research, UK. Bulletin No.258. Randles, J.W., 1986. Susceptibility of Echium plantagineum L. to tobacco mosaic, alfalfa mosaic, tobacco ringspot, and tobacco necrosis viruses. Australas. Plant Pathol. 15,74-77. Rast, A. TH. B., 1979. Infection of tomato seed by different strains of tobacco mosaic virus with particular reference to the symptomless mu tant MII-16. Neth. J. Plant Pathol. 85, 223-233. Ream, W., 1989. Agrobacterium tumefaciens and interkingdom genetic exchange. Annu Rev. Phytopathol. 27, 583-618. Rhie, M.J., Min, B.E., Hong, J.S., Song, Y.S ., Ryu, K.H., 2007. Complete genome sequence supports Bell pepper mottle virus as a species of the genus Tobamovirus. Arch. Virol. 152, 1401-1407.
210 Roberts, D.A., 1964. Local-lesion assa y of plant viruses. In: Corb ett,M.K., Sisler, H.D., editors. Plant Virology. University of Florid a Press, Gainesville, FL. p. 194-210. Robinson, D. 2004. A quarter century of progr ess in plant virology. Oxford Press. Rohozinski, J., Epstein, A.H., Hill, J.H., 2001. Pr obable mechanical transmission of a virus-like agent from rose rosette disease-infected multiflora rose to Nicotiana species. Ann. Appl. Biol. 138, 181-186. Romano, A., Raemakers, K., Visser, R., M ooibroek, H., 2001. Transformation of potato ( Solanum tuberosum ) using particle bombardmen t. Plant Cell Rep. 20, 198-204. Ruether, J., Sauer, B., 1989. A screen for SVgpt in E. coli DH5 lac and DH5 : small colony phenotype. Nucleic Acids Res. 17, 10499. Ryu, K.H., Min, B.E., Choi, G.S., Choi, S.H., Kwon, S.B., Noh, G.M., Yoon, J.Y., Choi, Y.M., Jang, S.H., Lee, G.P., Cho K.H., Park, W.M., 2000. Zucchini green mottle mosaic virus is a new tobamovirus; comparison of its coat pr otein gene with that of Kyuri green mottle mosaic virus. Arch. Virol. 145, 2325-2333. Saito, T., Meshi, T., Takamatsu, N., Okada, Y ., 1987. Coat protein gene sequence of tobacco mosaic virus encodes a host response determ inant. Proc. Natl. Acad. Sci. USA. 84, 60746077. Sambrook, J., Russell, D.W., 2001. Molecu lar Cloning: A laboratory manual. 3rd ed. Cold Spring Harbor Laboratory Press. Cold Spring Harbor, NY. Sawant, S.V., Singh, P.K., Tuli, R., 2000. Pretreat ment of microprojectiles to improve the delivery of DNA in plant transformation. BioTechniques 29, 246-248. Schneider, W.L., Roosinck, M.J., 2000. Evoluti onarily related Sindbis-like plant viruses maintain different levels of population di versity in a common host. J. Virol. 74, 31303134. Shimamoto, I., Sonoda, S., Vazquez, P., Minaka N., Nishiguchi, M., 1998. Nucleotide sequence analysis of the 3 terminal region of a wasabi strain of crucifer tobamovirus genomic RNA: subgrouping of crucifer tobamovi ruses. Arch. Virol. 143, 1801-1813. Shivprasad, S., Pogue, G.P., Lewandowski, D.J ., Hidalgo, J., Donson, J., Grill, L.K., Dawson, W.O., 1999. Heterologous sequences greatly a ffect foreign gene expression in Tobacco mosaic virus-based ve ctors. Virology 255, 312-323. Siegel, A., Wildman, S.G., 1954. So me natural relationships among strains of Tobacco mosaic virus. Phytopathology 44, 277-282. Silvestro S.R., Chapman, G.B., 2004. A tran smission electron micros cope study of New Dawn climber rose ( Rosa wichuraiana x safrano ) exhibiting rose rosette disease. Plant Cell Rep. 23, 345-351.
211 Smith, P.G., Heiser, C.B. Jr., 1951. Taxonomic and genetic studies on the cultivated peppers, Capsicum annuum L. and C. frutescens L. Am. J. Bot. 38, 362-368. Solis, I., Garca-Arenal, F. 1990. The complete nucleotide sequence of the genomic RNA of the tobamovirus, Tobacco mild green mosaic virus. Virology 177, 553-558. Song, Y.S., Min, B.E., Hong, J.S., Rhie, M.J., Ki m, M.J., Ryu, K.H., 2006. Molecular evidence supporting the confirmation of Maracuja mo saic virus as a species of the genus Tobamovirus and production of an infecti ous cDNA transcript. Arch. Virol. 151, 23372348. Srinivasan, K.G., Narendrakumar, R., Wong, S. M.. 2004. Hibiscus virus S is a new subgroup II tobamovirus: evidence from its unique coat protein and movement protein sequences. Arch. Virol. 147, 1585-1598. Stanley, W. M., 1935. Isolation of a crystalline protein possessing the properties of tobacco mosaic virus. Science 81, 644-645. Sturgis, A.K., Colvin D.L.. 1996. Co ntrolling tropical soda apple in pastures. In: Mullahey, J.J., (Ed.) Proceedings of the Tropical Soda Apple Symposium. Jan 9-10, 1996. Bartow, FL. Institute of Food and Agricultural Sciences, Univ ersity of Florida, Gainesville, FL, USA. p. 79. Sun, C-B., Kong, Q-L., Xu, W.-S., 2002. Efficient transformation of Penicillium chrysogenum mediated by Agrobacterium tumefaciens LBA4404 for cloning of Vitreoscilla hemoglobin gene. EJB Electronic Journal of Bi otechnology 5, 21-28. [cited 2007 Jul 3] http://www.ejb.org/content/vol5/issue1/full/5 Szab, B., Hori, K., Nakajima, A., Sasagawa, N ., Watanabe, Y., Ishiura S., 2004. Expression of amyloid1-40 and 1-42 peptides in Capsicum annuum var. angulosum for oral immunization. ASSAY and Drug De velopment technologies. 2, 383-388. Takamatsu, N., Watanabe, Y., Yanagi, H., Meshi, T., Shibal, T., Okada, Y., 1990. Production of enkephalin in tobacco protoplasts using tob acco mosaic virus RNA vector. FEBS Lett. 269, 73-76. Taliansky, M.E., Ryabov, E.V., Robinson, D.J., Pal ukaitis, P., 1998. Tomato cell death mediated by complimentary plant viral satellite RNA se quences. Mol. Plant-Microbe Interact. 11, 1214-1222. Thompson, J.D., Gibson, T.J., Plewniak, F ., Jeanmougin, F., Higgins, D.J., 1997. The CLUSTAL_X windows interface: flexible strate gies for multiple sequence alignment aided by quality analysis tools. Nucleic Acids Res. 25, 4876-4882. Thompson, J.D., Higgins, D.G., Gibson, T.J., 1994. CLUSTAL W: improving the sensitivity of progressive multiple sequence alignment th rough sequence weighting, position-specific gap penalties and weight matrix choice. Nucleic Acids Res. 22, 4673-4680.
212 Tsuda, S., Kirita, M., Watanabe, Y., 1998. Character ization of a Pepper mild mottle tobamovirus strain capable of overcoming the L3 gene-mediated resistance, distinct from the resistancebreaking Italian isolate. Mol. Plant-Microbe Interact. 11, 327-331. [USDA-NASS] USDA-National Ag ricultural Statistic s Service. 2007. QuickStats: US & All States Data Tobacco (All Classes). [cited 2007 Jun 20] http://www.nass.usda.gov/ Valverde, R.A., Heick, J.A., Dodds, J.A., 1991. Inte ractions between satellite tobacco mosaic virus, helper tobamoviruses, a nd their hosts. Phytopathology 81, 99-104. Valverde, R.A., 1990. Analysis of Double-Stranded RNA for Plant Virus Diagnosis. Plant Dis. 74, 255-258. Varma, A., 1986. Sunn-hemp mosaic virus. In: Van Regenmortel M.H.V., Fraenkel-Conrat H., editors. The Plant Viruses. Vol. 2. The rod-shap ed plant viruses. Plenum Press, New York. p. 249-266. Voinnet, O., Rivas, S., Mestre, P., Baulcomb e, D., 2003. An enhanced transient expression system in plants based on suppression of ge ne silencing by the p19 protein of tomato bushy stunt virus. Plant J. 33, 949-956. Weber, H., Haeckel, P., Pfitzner, J.P., 1992. A cDNA clone of Tomato mosaic virus is infectious in plants. J. Virol. 66, 3909-3912. Wetter, C., Conti, M., Altschuh, D., Ta billion, R., vanRegenmortel, M.H.V., 1984. Phytopathology. 74, 405. Wetter, C., Conti, M., 1988. Pepper mild mottle vi rus. CMI/AAB Descriptions of Plant Viruses, Association of Applied Biologists: Institute of Horticultural Research, UK. Bulletin No. 330. Wetter, C., 1986. Tobacco mild green mosaic vi rus. In: Van Regenmortel M.H.V., FraenkelConrat H., editors. The Plant Viruses. Vol. 2. The rod-shaped plant viruses. Plenum Press, New York. p. 205-219. Wetter, C., 1989. Tobacco Mild Green Mosaic Viru s. CMI/AAB Descriptions of Plant Viruses, Association of Applied Biologists: Institute of Horticultural Research, UK. Bulletin No. 351. Wetter, C., 1984. Antigenic relationships between is olates of mild dark-g reen tobacco mosaic virus, and the problem of host-i nduced mutation. Phytopathology 74, 1308-1312. Win, J., Kamoun, S., 2004. pCB301 -p19: A binary plasmid vector to enhance transient expression of transgenes by agroinfiltrati on. Kamoun Lab, Department of Plant Pathology, Ohio State University. [cited Ju l 4 2007] http://www.KamounLab.net Xu, P., Roossinck, M.J., 2000. Cucumber mosaic virus D satellite RNA-induced programmed cell death in tomat o. Plant Cell 12, 1079-1092.
213 Yamanaka, T., Komatani, H., Meshi, T., Na ito, S., Ishikawa, M., Ohno, T., 1998. Complete nucleotide sequence of the genomic RNA of To bacco mosaic virus strain Cg. Virus Genes 16, 173-176. Yoon, J.Y., Min, B.E., Choi, S.H., Ryu, K.H., 2001. Completion of nucleotide sequence and generation of highly infectious transcripts to cucurbits fr om full-length cDNA clone of Kyuri green mottle mosaic virus. Arch. Virol. 146, 2085-2096. Yoon J.Y., Min, B.E., Choi, J.K., Ryu, K.H ., 2002. Genome struct ure and production of biologically active in vitro transcripts of cucurbit-infecting Zucchini green mottle mosaic virus. Phytopathology. 92, 156-163. Yoon, J.Y., Ahn, H.I., Kim, M., Tsuda, S ., Ryu, K.H., 2006. Pepper mild mottle virus pathogenicity determinants and cross protecti on effect of attenuated mutants in pepper. Virus Res. 118, 23-30. Yordanova, A., Donev, T., Stoimenova, E., 2002. A model for longevity of freeze-dried tobamoviruses. Biotechnol. Lett. 24, 1505-1508. Yu, H.H., Wong S.M., 1998. A DNA clone encoding the full-length infectious genome of odontoglossum ringspot tobamovirus and mutagenesis of its coat protein gene. Arch. Virol. 143, 163-171. Yusibov, V., Hooper, D.C., Spitsin, S.V., Fleys h, N., Kean, R.B., Mikheeva, T., Deka, D., Karasev, A., Cox, S., Randall, J., Koprowski, H., 2002. Expression in plants and immunogenicity of plant virus-based expe rimental rabies vaccine. Vaccine 20, 3155. Zaitlin, M., Israel H.W., 1975. Tobacco Mosaic Virus (Type Strain.) Department of Plant Pathology. Cornell University. Ithaca, NY. CM I/AAB Descriptions of Plant Viruses, Association of Applied Biologists: Institute of Horticultural Research, UK. Bulletin No. 151. Zettler, F.W., Nagel, J., 1983. In fection of cultivated gesneria ds by two strains of tobacco mosaic virus. Plant Dis. 67, 1123-1125. Ziemienowicz, A., Rlich, D.G., Lanka, E., Hohn, B., Rossi, L. 1999. Import of DNA into mammalian nuclei by proteins originating from a plant pathogenic bacterium. Proc. Natl. Acad. Sci. USA. 96, 3729-3733.
214 BIOGRAPHICAL SKETCH Jonathan Robert Horrell was born to Robert and Cynthia Ho rrell in Fort Lauderdale, Florida. He lived and attended school in Lake Worth, Florida, until 1983. From 1983 until 1989 he lived various places in southwestern New Yo rk State, thereafter returning to Lake Worth where he attended high school at Lake Worth Community High School. Upon graduating high school, Jonathan travel ed the country alone, stopping to work occasionally as a handyman or farm laborer. Returning to Lake Worth, he then worked in advertising before deciding to pursue a higher education at the University of Florida. Working his way through college, Jonathan earned a Bachelor of Science in Plant Pathology in 2002. At the encouragement of his a dvisor, Dr. Francis William Zettler, Jonathan decided to pursue further education in gradua te school, under the mentorship of Dr. Raghavan Charudattan, who kindly provided him with an assistantship. Applying himself to many projects, his primary research goal has been the investigation of the host-parasite interaction between tropical soda apple and Tobacco mild green mosaic virus, and the identification of the predicted ge ne-for-gene interaction that takes place.