NOTE TO USERS
This reproduction is the best copy available.
UMI*
SUB-CELLULAR LOCALIZATION OF THE GRAPEVINE
RUPESTRIS STEMPITTING-ASSOCIATED VIRUS REPLICASE
A Thesis
Presented to
The Faculty of Graduate Studies
of
The University of Guelph
By
SEAN PROSSER
In partial fulfillment of requirements
for the degree of
Master of Science
November, 2009
©Sean Prosser, 2009 Library and Archives Bibliotheque et 1*1 Canada Archives Canada Published Heritage Direction du Branch Patrimoine de I'edition
395 Wellington Street 395, rue Wellington Ottawa ON K1A 0N4 OttawaONK1A0N4 Canada Canada
Vour Tile Votre reference ISBN: 978-0-494-58413-2 Our file Notre reference ISBN: 978-0-494-58413-2
NOTICE: AVIS:
The author has granted a non L'auteur a accorde une licence non exclusive exclusive license allowing Library and permettant a la Bibliotheque et Archives Archives Canada to reproduce, Canada de reproduire, publier, archiver, publish, archive, preserve, conserve, sauvegarder, conserver, transmettre au public communicate to the public by par telecommunication ou par I'lnternet, preter, telecommunication or on the Internet, distribuer et vendre des theses partout dans le loan, distribute and sell theses monde, a des fins commerciales ou autres, sur worldwide, for commercial or non support microforme, papier, electronique et/ou commercial purposes, in microform, autres formats. paper, electronic and/or any other formats.
The author retains copyright L'auteur conserve la propriete du droit d'auteur ownership and moral rights in this et des droits moraux qui protege cette these. Ni thesis. Neither the thesis nor la these ni des extraits substantiels de celle-ci substantial extracts from it may be ne doivent etre imprimes ou autrement printed or otherwise reproduced reproduits sans son autorisation. without the author's permission.
In compliance with the Canadian Conformement a la loi canadienne sur la Privacy Act some supporting forms protection de la vie privee, quelques may have been removed from this formulaires secondaires ont ete enleves de thesis. cette these.
While these forms may be included Bien que ces formulaires aient inclus dans in the document page count, their la pagination, il n'y aura aucun contenu removal does not represent any loss manquant. of content from the thesis.
1+1 Canada ABSTRACT
SUB-CELLULAR LOCALIZATION OF THE GRAPEVINE RUPESTRIS STEMPITTING-ASSOCIATED VIRUS REPLICASE
Sean William John Prosser Advisor: University of Guelph, 2009 Dr. Baozhong Meng
Co-Advisor: Dr. Peter Krell
The replicase polyprotein of Grapevine rupestris stem pitting-associated virus
(GRSPaV) was expressed in Nicotiana tabacum cv. Bright Yellow-2 cells and found to form punctate structures in the cytoplasm, which were hypothesized to represent viral replication vesicles. A region within the methyltransferase (MTR) domain was found to be responsible for the formation of punctate structures that aligned along actin microfilaments, but did not associate with the ER, Golgi stacks, or peroxisomes. An arginine residue (R173) and a tryptophan residue (W184) within the MTR domain were tested as putative critical residues required for membrane association based on the formation of the observed puncta; however, replacement of these residues had no effect on punctal formation. The RNA-dependent-RNA-polymerase domain of the replicase was found to form aggregate structures in the cytoplasm when expressed alone, but co- localized with the puncta formed by the MTR during co-expression experiments, suggesting a possible interaction between these two proteins. ACKNOWLEDGMENTS
I would like to thank my advisor, Dr. Baozhong Meng, for giving me the opportunity to work in his lab, as well as for his advice and help throughout the course of my degree. I would like to thank my advisory committee, Dr. Peter Krell, who was also my co-advisor, and Dr. Paul Goodwin. I would also like to thank Dr. Jay Subramarian for providing Vitis rupestris suspension cells and callus tissue. Finally I would like to thank the current and former members of the Meng, Krell, and Mullen labs, family, and friends for their advice and support.
i TABLE OF CONTENTS
TABLE OF CONTENTS.. ii LIST OF TABLES v LIST OF FIGURES vi LIST OF ABBREVIATIONS viii LIST OF VIRUS ABBREVIATIONS ix CHAPTER I: INTRODUCTION 1 1.1 Viticulture 1 1.2 Rugose Wood Complex 2 1.3 Rupestris Stem Pitting 5 1.4 Flexiviridae 6 1.4.1 Genera and Taxonomy 6 1.4.2 Physical Characteristics 6 1.4.3 Host Range 7
1.4.4 Symptoms of Infection ; 7 1.4.5 Transmission 10 1.4.6 Genome Structure 11 1.5 Positive Sense RNA Virus Replication 16 1.5.1 Positive RNA Virus Superfamilies 16 1.5.2 Replicase Expression and Processing 17 1.5.3 Host Proteins , 21 1.5.4 Site of Replication 24 1.5.5 Membrane Association 26 1.5.6 Replication of Viral RNA 30 1.6 Grapevine Rupestris Stem Pitting-associated Virus 38 1.6.1 Physical Characteristics 38 1.6.2 Molecular Characteristics 38 1.6.3 ORF1: The viral replicase 41 1.6.4 ORF2, ORF3, and ORF4: The viral movement proteins 42 1.6.5 ORF5: The viral capsid protein 43
n 1.6.6 Symptoms and Transmission.. 44 1.6.7 Population Structure , 45 1.7. Hypotheses 46 CHAPTER 2: MATERIALS AND METHODS 48 2.1 Polymerase Chain Reaction (PCR) 48 2.2 Restriction Digestions 48 2.3 PCR Purification and Gel Extraction of DNA 49 2.4 De-phosphorylation, Ligation, and Transformation of DNA 49 2.5 Generation of Chemically Competent E. coli 51 2.6 Generation and Transformation of Electrocompetent.fi'. coli 51 2.7 Tobacco Bright Yellow (BY-2) Tissue Culture 52 2.8 Isolation of Tobacco BY-2 Protoplasts 53 2.9 Transfection of Tobacco BY-2 Protoplasts 54 2.10 Biolistic Bombardment and Immunostaining of Tobacco BY-2 Cells 55 2.10.1 Biolistic Bombardment , 55 2.10.2 Fixation and Immunostaining 58 2. 11 Epifluorescent Microscopy 59 2.11.1 Preparation of BY-2 Cells for Microscopic Examination 59 2.11.2 Preparation of BY-2 Protoplasts for Microscopic Examination 59 2.11.3 Epifluorescent Microscopy 60 2.12 Construction of Plant Expression Constructs 61 2.12.1 Full-length Replicase (pRSP-REP:GFP) 61 2.12.2 Truncation Constructs (pMTRASD:GFP, pMTR:GFP, pMTR:mRFP, pSD:GFP, pPOL:GFP) 66 2.12.3 Construction of pMTR:GFP (R173A), pMTR:GFP (W184A), and pMTR:GFP(R173A,W184A) 72 2.12.4 Construction of pER-mRFP 73 CHAPTER 3: RESULTS 75 3.1 The GRSPaV replicase forms punctate structures in the cytoplasm of tobacco BY-2 cells and protoplasts 75 3.2 A sub-domain within the MTR domain is important for the formation of the viral puncta 81
iii 3.3 The GRSPaV replicase polyprotein residues R173 and W184 are not critical for the formation of the viral puncta 90 3.4 The viral puncta appear to localize to actin microfilaments and adjacent to the ER lumen 91 3.5 The RdRp domain of the GRSPaV replicase forms aggregate structures in the cytoplasm when expressed alone 103 CHAPTER 4: DISCUSSION AND CONCLUSIONS 106 4.1 The GRSPaV replicase forms punctate structures when expressed in tobacco BY-2 cells and protoplasts 106 4.2 Amino acid residues 132-207 of the MTR domain of the GRSPaV replicase are responsible for the formation of viral puncta 107 4.3 The GRSPaV replicase polyprotein residues R173 and W184 are not critical for the formation of viral puncta 109 4.4 The viral puncta localize adjacent to the ER lumen and appear to align along actin microfilaments 113 4.5 The GRSPaV RdRp domain forms aggregate structures in tobacco BY-2 protoplasts when expressed alone and may interact with the MTR domain 115 4.6 Conclusions : 117 REFERENCES 123 APPENDIX I: INSTABILITY OF pREP:GFP IN E. coli DH5a 131 APPENDIX II: GENERATION OF GRAPEVINE PROTOPLASTS 141 APPENDIX III: IMMUNOSTAINING OF TOBACCO BY-2 PROTOPLASTS 145
IV LIST OF TABLES
Table Page
Table 1: Primers used in this study. 62
v LIST OF FIGURES Figure Page
Figure 1: Symptoms of the four diseases comprising the Rugose Wood Complex. 3
Figure 2: Electron micrograph images showing virions of six representative members of the family Flexiviridae. 8
Figure 3: Genome structure of five representative members of the family
Flexiviridae. 12
Figure 4: Replicase structures of the four positive RNA virus superfamilies. 18
Figure 5: Three dimensional structure of the Semliki Forest virus monotopic membrane association signal. 28 Figure 6: Replication of a positive RNA viral genome utilizing sub-genomic RNA production. 31
Figure 7: Physical and molecular structure of Grapevine rupestris stem
pitting-associated virus. 39
Figure 8: Vector map of pRTL2. 67
Figure 9: Full-length and truncated replicase constructs used in this study. 76
Figure 10: Epifluorescent images of tobacco BY-2 cells and protoplasts expressing the full-length GRSPaV replicase. 78 Figure 11: Predicted amphipathic regions of the GRSPaV methyltransferase domain. 82
Figure 12: Nucleotide alignment of the GRSPaV methyltransferase domain with the Semliki Forest virus nsPl replication protein and its relative position to a predicted amphipathic region within the "sub-domain". 85
Figure 13: Epifluorescent images of tobacco BY-2 protoplasts expressing
regions of the methyltransferase domain. 88
Figure 14: Mutant truncation constructs used in this study. 92
Figure 15: Epifluorescent images of tobacco BY-2 protoplasts expressing mutant truncation constructs. 94
vi Figure 16: Epifluorescent images of tobacco BY-2 protoplasts co-expressing the methyltransferase domain with an ER marker protein. 97
Figure 17: Epifluorescent images of tobacco BY-2 protoplasts co-expressing the methyltransferase domain with Golgi and peroxisome marker proteins. 99
Figure 18: Epifluorescent images of tobacco BY-2 protoplasts co-expressing the methyltransferase domain with an actin marker protein. 101
Figure 19: Epifluorescent images of tobacco BY-2 protoplasts expressing the GRSPaV RNA-dependent-RNA-polymerase domain alone or with the GRSPaV methyltransferase domain. 104
Figure 20: Strategy used to sequentially clone the GRSPaV replicase into pRTL3 and position of a putative prokaryotic cryptic promoter. 133
Figure 21: Schematic diagram showing the insertion of the E. coli IS1 transposon into the GRSPaV genome during cloning procedures using E. coli DH5ct as a host strain. 137
vii LIST OF ABBREVIATIONS
BY-2 Bright Yellow-2
CB Corky bark
CP Capsid protein/coat protein
ER Endoplasmic reticulum
HA Hemagglutinin
HEL Helicase
HVR Highly variable region
HPT Hours post transfection
KSG Kober stem grooving
LNSG LN-33 stem grooving
MES 2[N-morpholino] ethanesulfonic acid mRFP Monomeric red fluorescent protein
MTR Methyltransferase
O-Pro Ovarian tumour protease
P-Pro Papain-like cysteine protease
RdRp RNA-dependent-RNA-polymerase
RSP Rupestris stem pitting
RWC Rugose wood complex
SDM Site directed mutagenesis sgRNA Sub-genomic RNA
TGB Triple gene block
UTR Un-translated region
Vlll LIST OF VIRUS ABBREVIATIONS
AMV Alfalfa mosaic virus
BIScV Blueberry scorch virus
BMV Brome mosaic virus
BYV Beet yellows virus
CaMV Cauliflower mosaic virus
CMV Cucumber mosaic virus
CNV Cucumber necrosis virus
GRSPaV Grapevine rupestris stem pitting-associated virus
GVA Grapevine virus A
GVB Grapevine virus B
PVX Potato virus X
PVY Potato virus Y
SFV Semliki Forest virus
TBSV Tomato bushy stunt virus
TCV Turnip crinkle virus
TEV Tobacco etch virus
TMV Tobacco mosaic virus
TYMV Turnip yellow mosaic virus
IX CHAPTER 1: INTRODUCTION
1.1 Viticulture
Grapes are one of the largest and oldest commercially grown fruit crops in the world, globally covering approximately eight million hectares worth billions of dollars annually (Mullins et al, 1992). The common grape (Vitis vinifera) has been cultivated for over 7000 years, dating back to the area currently known as Iran. Throughout its history, the grape has had a significant impact on human civilizations, economics, politics, trade, and religion (This et al, 2006). The cultivation of grapes (viticulture) spread with the
Roman Empire across Europe and Asia, and eventually spread to North America with the
New World colonists. While North America has several native species of grapes, Vitis vinifera was unable to grow in the non-native soils due to a root damaging insect called grape phylloxera. Subsequent transport of phylloxera infested Vitis vinifera stocks back to Europe spread the invasive insect to the well established European Vitis crops, resulting in almost complete collapse of the grape and wine industries (This et al, 2006).
To overcome this devastating infestation, European grapevines were grafted to wild
North American rootstocks which were resistant to phylloxera (Granett et al, 2001).
Starting with this desperate tactic, the now common practice of grafting was born.
While grafting saved European viticulture from collapse, it came with negative consequences which, unknown at the time, we are only now beginning to understand. Of all the woody fruit crops of commercial and economic significance, grapes are among the most susceptible to viruses, hosting up to fifty five distinct viruses from twenty genera and seven families (Meng and Gonsalves, 2008). It is likely that the process of grafting, which is now known to transmit many grapevine viruses, led to the unusually large 1 dissemination of viruses among the world's grape stocks. Indeed, Vitis cultivars from which rootstocks have been derived tend to harbour a more uniform population of viruses compared to scion cultivars (vines grafted onto the rootstock), which usually contain multiple species of viruses (Meng et al, 2006). Regardless of how grapevines came to be susceptible to so many plant viruses, the end result is that global viticulture requires consistent monitoring and the constant employment of methods to protect the important crops from numerous viral diseases.
1.2 Rugose Wood Complex
Many viruses that infect woody plants elicit xylem modification in infected hosts, which can manifest in the forms of shallow strips of pits or grooves on the xylem beneath the outer bark. The Rugose Wood Complex (RWC) describes a set of graft-transmissible, xylem-modifying viral diseases of grapevines (Osman and Rowhani, 2008). RWC is an economically important and widespread phenomenon causing severe reduction in the growth and yield of grapevines (Meng and Gonsalves, 2008; Osman and Rowhani, 2008).
RWC is comprised of four distinct diseases: Rupestris stem pitting (RSP), Kober stem grooving (KSG), LN stem grooving (LNSG), and corky bark (CB) (Meng and Gonsalves,
2008). RSP, KSG, and CB are associated with, and thought to be caused by Grapevine rupestris stem-pitting-associated virus (GRSPaV), Grapevine virus A (GVA), and
Grapevine virus B (GVB) respectively (Figure 1) (Boscia et al, 1993; Chevalier et al,
1995; Meng et al, 1998; Zhang et al, 1998; Osman and Rowhani, 2008). GVA and GVB are phloem restricted and all three viruses show a seasonal fluctuation in titre (positively correlated with the growing season) in infected grapevines (Osman and Rowhani, 2008).
2 Fig. 1. Symptoms of the four diseases that comprise the Rugose wood complex. LN-
Stem Grooving (LNSG), Kober Stem Grooving (KSG), Rupestris Stem Pitting (RSP), and Corky Bark (CB). Viruses suspected of being the causal agent of each disease are identified below each image (adapted from Martelli et al, 2007). LNSG KSG RSP
BBS »••»•„• •'
< * Hfflffi t 1 v™ . ..I , ! i « ''• H * v«V m •'. ! •'",' H . t' ! fil!' ' i .'i'. HE- ;:• > t.. • MI; " i. 1 ')'
•! :i 1 f - • ••':i- ':r:i HI '. *<' f • . ,i • B' - • 11, HI Hr \» BE • \ ' i He '"' ' **•" 1 > . ! HI '• •-' '1 Unknown Grapevine Grapevine Grapevine causative virus A rupestris virus B agent stem pitting- associated virus
4 In the past, RWC diseases were diagnosed by grafting the suspected vine onto known uninfected rootstocks, waiting several years, and then visually inspecting the xylem of the rootstock for symptoms of the suspected disease (Martelli, 1993). This method is time consuming, expensive, and requires special resources and skills. The current method of choice for RWC diagnosis is a reverse transcription polymerase chain reaction (RT-PCR) based assay which specifically detects the putative causal virus in the suspected plants.
The RT-PCR method is significantly faster, cheaper, easier, and more reliable and is now a common diagnostic tool (Osman and Rowhani, 2008).
1.3 Rupestris Stem Pitting
RSP was discovered in the 1970's and is the most prevalent disease of the RWC
(Martelli, 1993). RSP derives its name from its characteristic symptoms, which are small pits found in strips under the outer bark of the indicator Vitis rupestris "St. George" that has been graft inoculated (Figure 1). The graft transmissible nature of RSP led scientists to assume the causal agent was a virus, a theory which was supported when, in 1998, two independent groups isolated double-stranded RNA (dsRNA) from RSP infected plants
(Meng, et al, 1998; Zhang, et al, 1998). The dsRNA was found to be the replicative intermediate form of a single-stranded RNA (ssRNA) virus, named GRSPaV. Since then,
GRSPaV has been consistently associated with RSP, however it has also been found in grapevines showing little to no symptoms of RSP. Whether GRSPaV is solely responsible for RSP is unknown, as the complex nature of plant viruses and the high number of viruses that infect grapevines results in many variables and hence makes
Koch's postulates very difficult to fulfill. However, the fact that GRSPaV is associated
5 with RSP suggests GRSPaV must play an important, if not critical, role in the formation of this disease.
1.4 Flexiviridae
1.4.1 Genera and Taxonomy
GRSPaV is a member of the plant virus family Flexiviridae. Flexiviridae contains eight genera (Allexivirus, Capillovirus, Carlavirus, Foveavirus, Mandarivirus,
Potexvirus, Trichovirus, Citrivirus and Vitivirus) and some unassigned viruses (Adams et al, 2004; Martelli et al, 2007; Carstens, 2009). Members of Flexiviridae are structurally most related to the families Potyviridae and Closterviridae; however molecular analysis has revealed that members of Flexiviridae most likely evolved from the family
Tymoviridae (Martelli et al, 2007). Indeed, the replicase proteins (see section 1.4.6) of potexviruses, mandariviruses, and allexiviruses are more closely related to Tymoviridae than to other genera of the Flexiviridae.
1.4.2 Physical Characteristics
All members of Flexiviridae are characterized by flexuous filamentous particles of 12-13 nm in diameter and approximately 470-1000 nm in length. The virions have been suggested to possess viral protein "tails" similar to members of Potyviridae and
Closteroviridae (Torrance et al, 2006; Martelli et al, 2007). The virion "tails" of potyviruses are formed by the interaction of capsid proteins, viral proteases, and the potyviral 5' VPg cap, and it has been suggested the tails are involved in intracellular
6 movement (Torrance et al, 2006). The potyviral tails are also thought to be involved in
cell-to-cell and systemic movement, as well as transmission between plants via aphid
vectors. It has also been suggested that the potyvirus "tails" interact with host factors to
promote the translation of viral RNA (Torrance et al, 2006). The degree of plasticity of
the virions varies among genera with potexviruses and carlaviruses being the least
flexible (Figure 2).
1.4.3 Host Range
Members of Flexiviridae infect many herbaceous and woody dicotyledonous
plants, and to a lesser extent, monocotyledonous plants. Potexviruses, carlaviruses, and
allexiviruses tend to infect herbaceous plants while foveaviruses, capilloviruses,
vitiviruses, mandariviruses, and trichoviruses tend to infect woody hosts (Martelli et al,
2007). Allexivirus infections are limited to Allum species (onion family), while trichovirus and vitivirus infections are mostly limited to stone fruit trees and Vitis (grape)
species respectively. Collectively, members of Flexiviridae infect many economically and commercially important crops.
1.4.4 Symptoms of Infection
Depending on the particular virus and host species, Flexiviridae infections can result in mild to no symptoms, parenchymatous tissue modification such as leaf rolling or wrinkling, or xylem modification such as the RWC. Carlavirus infections are often symptomless but can produce mild mosaic patterns in some instances (Bristow et al,
2000; Martelli et al, 2007). Likewise, allexivirus infections are usually asymptomatic in
7 . ' Fig. 2. Electron micrograph images showing virions of six representative members of the family Flexiviridae. Potexviruses and carlaviruses can be seen as the most rigid virions. Bar = 50 nm. Adapted from Martelli et al. (2007). 51iH
Mandarivirus
Carlavirus
Foveavirus
» .;*•* -'4'. ~-.^:^ '? K* .^ ^^^^^-^p#f| Vitivirus Potexvirus 9 onions. Mandarivirus infections result in chlorotic ring-spots on leaf tissue, leading to the rapid decline of the infected host. The symptoms of a trichovirus infection can range from moderately stunted growth of the affected tree to deformation and/or necrosis of the fruit (Martelli et al, 2007). Potexvirus infections are by far the most pathogenic and detrimental to the host. Potato virus X (PVX), a potexvirus, is among the top ten most biologically, economically, and commercially damaging plant viruses worldwide. Additionally, the virulence of PVX is known to be increased in the presence of mixed infections. Mixture of Potato virus Y (PVY), a member of the family Potyviridae, with PVX was found to increase the pathogenicity of PVX in potato (Martelli et al, 2007). As mentioned previously, foveavirus, capillovirus, and vitivirus infections can lead to xylem modification resulting in the gradual decline of the affected plant; however, many foveavirus infections are asymptomatic (Martelli et al, 2007; Meng and Gonsalves, 2008). Some foveavirus infections are symptomless and non-pathogenic in one species but are highly detrimental in another species. For example, the foveavirus Apricot latent virus is asymptomatic in apricots but highly pathogenic to peach trees (Gentit et al, 2001). Xylem modifications from certain foveaviruses and capilloviruses are extreme enough to cause graft incompatibility in apple and pear cultivars respectively (Martelli et al, 2007). 1.4.5 Transmission In general, members of Flexiviridae that replicate primarily in parenchymatous tissue (potexviruses, carlaviruses, allexiviruses, capilloviruses, foveaviruses, mandariviruses and trichoviruses) are more easily transmitted than members that do not 10 (vitiviruses). Vitiviruses are phloem restricted in their natural hosts and therefore rely on graft transmission for dissemination, as do foveaviruses though they are not phloem restricted (Martelli et al, 2007; Meng and Gonsalves, 2008). Capilloviruses and mandariviruses also rely on graft transmission. Allexiviruses and trichoviruses are transmitted by mites in an unknown manner, while carlaviruses are transmitted by aphids. Potexviruses are very stable, replicate to a high titre, and are highly infectious, rendering them capable of transmission simply by mechanical contact between plants. Most members of Flexiviridae are not spread in seeds, or if they are, then the spread is negligible compared to the above mentioned methods. 1.4.6 Genome Structure All members of Flexiviridae possess positive sense, ssRNA genomes which are monopartite and contain a poly(A) tail (Figure 3). The Flexiviridae genome can contain five or more open reading frames (ORFs) required for replication, movement, encapsidation, and virion assembly. As with all (+)RNA viruses, members of Flexiviridae must express a replicase protein directly from the viral genome upon infection (David et al, 1992; van der Heijden and Bol 2002; Dolja et al, 2006; Martelli et al, 2007). At a minimum, the replicase protein must contain those domains indicative of the alphavirus-like superfamily (see section 1.5.1): a methyltransferase (MTR), helicase (HEL), and RNA-dependent-RNA-polymerase (RdRp) (van der Heijden and Bol, 2002). The MTR domain is thought to be involved in the addition of a 7-methylguanoside cap to the viral RNA, and is required by the virus because the host enzyme and cellular RNA capping occur in the nucleus while viral replication takes place in the cytoplasm 11 Fig. 3. Genome structure of five representative members of the family Flexiviridae. Genome structure of Grapevine rupestris stem pitting-associated virus (GRSPaV) compared to other members of Flexiviridae: Potato virus M (PVM, Carlavirus), Potato virus X (PVX, Potexvirus), Indian citrus ringspot virus (ICRSV, Mandarivirus), and Shallot virus X(ShVX, Allexivirus). Open reading frame (ORF) 1 shows various domains associated with alphavirus-like replicases: Methyltransferase (MTR), cysteine-like protease (PRO), helicase (HEL), and RNA-dependent-RNA-polymerase (RDRP). Triple gene block (TGB) and capsid (CP) genes are also shown. Sixth (shaded red) ORF of PVM, ICRSV, and ShVX represents a gene for an unknown RNA-binding protein. Kb, kilobases. Adapted from Meng and Gonsalves, (2007). 12 MTR PRO HEL RDRP TGEpl C P GRSPaV PVM U)B PVX (A)n ICRSV ShVX 9kb 13 (Rozanov et al, 1992; Nagy and Pogany, 2008). The HEL domain is thought to unwind the double-stranded RNA (dsRNA) intermediate and to remove ssRNA hairpin structures during viral replication (Gorbalenya et al, 1988; van der Heijden and Bol, 2002; Nagy and Pogany, 2008). The MTR and HEL domains are conserved across alphavirus-like replicases (Koonin and Dolja, 1993). The RdRp domain is highly conserved across all (+)RNA virus families and is involved in polymerizing both positive and negative sense RNA during replication (Koonin, 1991; van der Heijden and Bol, 2002). In addition to the MTR, HEL, and RdRp domains, carlaviruses, capilloviruses, foveaviruses and trichoviruses also contain a papain-like cysteine protease (P-Pro) for proteolytic processing of the polyprotein (see section 1.5.2). While less common in plant virus replicases, all alphavirus-like animal virus replicases contain a P-Pro domain, and it has been speculated that the P-Pro domain was once in all members of Flexiviridae but lost over evolutionary time so that some now lack the domain (Koonin and Dolja, 1993; Martelli et al, 2007). In addition to the P-Pro domain, most carlaviruses and foveaviruses contain a second protease domain, the ovarian tumour-like protease (O-Pro). The lack of the O-Pro domain in most members of Flexiviridae suggests that O-Pro was likely acquired by a virus from its host and horizontally transferred to other carlaviruses and foveaviruses via co-infection and recombination (Martelli et al, 2007). An AlkB-like domain has also recently been discovered within the replicase of viruses from each genus of Flexiviridae except Capillovirus. The precise function(s) of the AlkB-like domain is currently unknown, however it has been shown to repair both ssRNA and dsRNA methylation damage, similar to cellular AlkB, and is thus theorized to be involved in counter- 14 measures against host defence (Aas et al, 2003; van den Born et al, 2008). Keeping with this theory is the fact that the AlkB-like domain seems to be present mostly in viruses which infect woody hosts. Alkylating agents are often incorporated into the plant phloem and surrounding cells, where woody plant-infecting viruses must often survive for long periods of time prior to transmission (van den Born et al, 2008). The resulting "confinement" of the virus within a single plant would entail more opportunity for methylation damage and thus having a suitable mechanism of repair would be a great advantage to these types of viruses. Indeed, the AlkB-like domain is rarely found in potexviruses, carlaviruses, and allexiviruses, which infect herbaceous hosts. Similar to the O-Pro domain, the AlkB-like domain was likely acquired by a single virus from its host and horizontally transferred to other co-infecting viruses. Unlike the replicase, which is always the 5' most proximal gene, the 3' proximal genes are quite diverse among members of Flexiviridae. All members contain one or more genes encoding movement protein(s), as well as a gene encoding a capsid protein. The movement protein(s) can be "p30-like" or arranged in a "Triple Gene Block" (TGB). Within Flexiviridae, the p30-like movement protein is found only in capilloviruses, trichoviruses, and vitiviruses. The p30-like movement protein is translated to a single protein from a single ORF while the TGB proteins, as the name suggests, comprise three different proteins translated from three different ORFs. While both types of movement proteins serve to translocate the virion or viral RNA between adjacent cells, they seemed to have evolved separately (Martelli et al, 2007). The TGB proteins are more conserved among members of Flexiviridae than the p30-like movement proteins. The CP proteins form the viral coat, encapsidating the viral RNA. More recently, 15 the CP has also been implicated in viral replication and movement (Martelli et al, 2007; Verchot-Lubicz et al, 2007). Like the movement proteins, CP of viruses within Flexiviridae is divided into two separate phylogenetic clusters (Martelli et al, 2007). Since the two types of CP cluster with the p30-like and TGB movement proteins, it is thought that the CP likely co-evolved with the movement proteins. All members of Flexiviridae contain the above mentioned genes, however some also contain one or more additional ORFs putatively encoding an RNA-binding protein and/or an unknown protein (Figure 3). RNA binding proteins, whose function is unknown, have been found in some carlaviruses, allexiviruses, vitiviruses, and mandariviruses. Additional ORFs with completely unknown functions have been reported in some allexiviruses and vitiviruses. The limited occurrence of these extra ORFs in Flexiviridae members suggests they were likely assimilated from a host and horizontally transferred between viruses. l.S Positive Sense RNA Virus Replication 1.5.1 Positive RNA Virus Superfamilies Plant, animal, invertebrate, and fungal (+)RNA viruses differ greatly in morphology, host range, pathology, genome structure, and life cycle. However, they do share a similar mode of genome replication, exemplified by a universally conserved RdRp that is always expressed directly from the viral genome upon infection (David et al, 1992; van der Heijden and Bol, 2002; Dolja et al, 2006; Martelli et al, 2007). While the RdRp is always required for (+)RNA virus replication, domains like the MTR, HEL, P-Pro, and O-Pro are not. Consequently, all (+)RNA viruses have been demarcated into 16 four superfamilies based on the absence/presence and type of domains present in their replicase protein(s). The specific composition of the replicase. determines whether a specific virus belongs to the picornavirus-like, flavivirus-like, carmovirus-like, or alphavirus-like superfamily (Figure 4). Alphavirus-like and flavivirus-like replicases contain MTR, HEL, and RdRp domains, though they differ in type. Picornavirus-like replicases contain only HEL and RdRp domains, while carmovirus-like replicases contain only an RdRp domain (van der Heijden and Bol, 2002). GRSPaV belongs to the alphavirus-like superfamily (Meng and Gonsalves, 2007), and so the following section on (+)RNA virus replication will concentrate on alphavirus-like replication. 1.5.2 Replicase Expression and Processing Regardless of the specific strategy of replicase activation and function, the role of all (+)RNA virus replicase protein(s) is to specifically associate with the viral RNA, localize to a specific site of the host cell, mediate the assembly of the replication complex, and finally replicate the positive-sense, negative-sense, and sub-genomic viral RNA (Nagy and Pogany, 2008). To accomplish these tasks, alphavirus-like viruses utilize an MTR, HEL, RdRp, and in some cases additional domains (e.g. proteases). The required domains are expressed, upon entry into the cell, directly from the viral genomic (+)RNA. In members of the family Bromoviridae, the host translation initiation factors eIF4E and eIF4G interact with the 3' un-translated region (UTR) of the viral (+)RNA, likely to circularize the RNA for replicase translation. The capsid protein of Alfalfa mosaic virus (AMV, family Bromoviridae), but not Brome mosaic virus (BMV, Tamily Bromoviridae), has been shown to associate with the 3' UTR, suggesting a possible role 17 Fig. 4. Replicase structures of the four positive RNA virus superfamilies. All positive-strand RNA viruses are separated into the alphavirus-like, picornavirus-like, flavivirus-like, and carmovirus-like superfamilies based on the viral replicase composition. MTR, methytransferases; HEL, helicases; PRO, protease; POL, RNA- dependent-RNA-polymerases; VPg, viral protein (genome-linked). 18 Alphavirus-like fe*^^*^^^^^^^^ Picornavirus-like Flavivirus-like i^-at-swA*. -. listfJ* "tHu-wil Carmovirus-like 19 in translation in some viruses but not others (Sztuba-Solinska and Bujarski, 2008). The specific manner of expression varies among alphavirus-like viruses. For example, BMV encodes two replicase proteins on separate RNA molecules, RNA1 and RNA2. RNA1 produces a protein containing the MTR and HEL domains, while RNA2 produces a protein containing the RdRp domain (den Boon et al, 2001; Nagy and Pogany, 2008; Sztuba-Solinska and Bujarski, 2008). Tobacco mosaic virus (TMV, unassigned to a family) encodes a single replicase protein containing the MTR and HEL domains on a single RNA molecule. Suppression of a stop codon is used to produce a larger fusion protein containing the MTR, HEL, and RdRp domain (Buck, 1999; Nagy and Pogany, 2008). Turnip yellow mosaic virus (TYMV, family Tymoviridae) and Blueberry scorch virus (BIScV, family Flexiviridae) encode a single polyprotein containing the MTR, HEL, and RdRp domains from a single RNA molecule. Post-translational auto- proteolytic processing via one or more protease domains in the viral replicase results in the separation of the RdRp domain from the MTR and HEL domains (David et al, 1992; Lawrence et al, 1995; Prod'homme et al, 2001; van der Heijden and Bol, 2002). Some viruses combine multiple methods to produce an active replicase. For example, Beet yellows virus (BYV, family Closteroviridae) utilizes a.+1 ribosomal frameshift, rather than suppression of a stop codon, to produce the fused RdRp domain, which is then auto- catalytically cleaved off via a viral protease domain (van der Heijden and Bol, 2002). Thus, a common feature of alphavirus-like replicases, regardless of the taxonomic family, seems to be that the active replicase must contain an RdRp domain on a separate protein, whether the RdRp domain be separated at the RNA level (e.g. BMV), the translational level (e.g. TMV), the post-translational level (e.g. TYMV and BIScV), or a 20 combination of multiple levels (e.g. BYV). Indeed, inhibition of RdRp separation via deletion of the protease domain of BYV resulted in a reduction of viral replication to 0.1% of wildtype levels (Peng and Dolja, 2000). The active replicase, formed from the interaction of the MTR-HEL and RdRP-containing proteins, has been reported to contain an MTR-HEL:RdRp molecular ratio ranging from 25:1 to 2:1 ( Watanabe et al, 1999; van der Heijden and Bol, 2002; Nagy and Pogany, 2008). The actual interacting domains were found to be the HEL and RdRp domains in most alphavirus-like replicases examined to date, however the TYMV replicase was found to re-assemble via an interaction between the RdRp and protease domains (Jakubiec et al, 2004; Jonczyk et al, 2007). 1.5.3 Host Proteins While the active form of the viral replicase requires specific expression strategies and/or proteolytic processing (see section 1.5.2), the active replication complex requires, in addition to the active viral replicase, one or more host factors (van Kammen, 1985; David et al, 1992; van der Heijden and Bol, 2002). Originally, cellular plant RdRp was thought to be involved, but this theory has since been discounted. Cellular RdRp is soluble and non-specific regarding the RNA to be polymerized, while viral RdRp is now known to be membrane bound (see section 1.5.4), has a high degree of specificity to viral RNA, and will polymerize RNA of much longer lengths than cellular RdRp (David et al, 1992). While possibly sharing a common ancestor, viral RdRp, not cellular plant RdRp, has been tailored to meet the high demands required by the virus during replication. The complete constituents of any alphavirus-like replication complex has yet to be 21 determined, however components of the replication complex have been identified as conserved translation initiation factors and chaperone proteins (Restrepo-Hartwig and Ahlquist, 1996; Van Der Heijden et al, 2001; Wang et al, 2009). The proposed function of host factors within a viral replication complex range from the stimulation of the viral RdRp, enhancement of replication, activation of viral reverse transcriptase (in the case of hepadnaviruses), formation of viral inclusion bodies, to the assembly of virions (Wang et al, 2009). Currently, the most extensively studied host factors known to be involved in the viral replication complex are from bromoviruses, tobamoviruses, potexviruses, and tombusviruses. About ten host factors have been implicated as being part of the BMV replication complex. Dedlp interacts with the 5' UTR of RNA2 to down-regulate viral RNA replication (Nagy and Pogany, 2008). Another host protein involved in the BMV replication complex, p41, is an eIF-3 type translation initiation factor and likely plays a role in BMV RNA replication (Nagy and Pogany, 2008). The TMV replication complex contains two host factors: GCD10, an eIF-3 type initiation factor, and eEFIA, an elongation factor, both putatively involved in viral RNA synthesis (Nagy and Pogany, 2008). TOM1 and TOM3, from Arabidopsis thaliana, and a TOM1 homologue from Nicotiana benthamiana are critical for TMV replication (Yamanaka et al, 2000; Yamanaka et al, 2002; Chen et al, 2007). In Arabidopsis, TOM1 and TOM3 have been shown to interact with the HEL domain of the replicase and anchor it to the endoplasmic reticulum (ER) for replication (see section 1.5.4) (Yamanaka et al, 2000; Yamanaka et al, 2002). A heat shock protein 70 (HSP70) homologue was also found to be required for TMV replication in Nicotiana benthamiana (Wang et al, 2009). 22 Host factors involved in the replication complex have been extensively studied in tombusviruses. Experiments in Saccharomyces cerevisiae have identified seven host factors required for Cucumber necrosis virus (CNV, family Tombusviridae) and Tomato bushy stunt virus (TBSV, family Tombusviridae) replication: Ssal/2p (HSP70 homologues), Tdh2/3p (glyceraldehyde 3-phosphate dehydrogenase, GAPDH), Pdclp (pyruvate decarboxylase), Pexl9p (involved in peroxisome biogenesis), and an unknown 35 kDa protein (Serva and Nagy, 2006; Pathak et al, 2008). Ssal/2p, the HSP70 homologue, is also critical for TBSV and Turnip crinkle virus (TCV, family: Tombusviridae) replication in Nicotiana benthamiana (Wang et al, 2009). In HSP70 deletion mutant yeast strains, the TBSV replicase remained cytosolic, failing to associate with the peroxisomal membrane (see section 1.5.4). HSP70 has therefore been hypothesized to promote the association of the tombusvirus replication complex with peroxisome membranes. Sepcifically, HSP70 has been suggested to shield a hydrophobic region of the tombusvirus replicase, preventing aggregation and allowing access by another host factor, Pexl9p, which is thought to be responsible for the actual peroxisome membrane association (Pathak et al, 2008; Wang et al, 2009). However, HSP70 has been shown to remain associated with the tombusvirus replication complex even after membrane association, suggesting it may perform additional functions during replication. While much of the replication cycle of potexviruses is unknown, it has been demonstrated that a host protein binds to the 3' UTR of PVX, suggesting a possible regulatory role in (+)RNA synthesis (Verchot-Lubicz et al, 2007). It has been suggested that a host protein may also be involved in localizing the PVX replication complex to the site of replication, however the specific host protein is currently unknown and the site of 23 replication has only recently been identified (Bamunusinghe et al, 2009). 1.5.4 Site of Replication An active alphavirus-like replication complex is comprised of a processed viral replicase and supplementary viral and host proteins. However, this active replication complex cannot replicate viral RNA when simply diffused throughout the cytoplasm. A host membrane is the last component required for the replication complex to fulfill its function (David et al, 1992; Jonczyk et al, 2007). Indeed, every (+)RNA virus, whether of plant, animal, fungus, or invertebrate, has been found to require a host membrane for replication (den Boon et al, 2001). The particular membrane seems to be dependent upon the particular virus. For example, the majority of (+)RNA viruses associate with the ER membrane, including many picornaviruses (e.g. Poliovirus), flaviviruses (e.g. West Nile virus), bromoviruses (e.g. BMV), tobamoviruses (e.g. TMV), potexviruses (e.g. PVX) and potyviruses (e.g. PVY) (Chen and Ahlquist, 2000; Wei and Wang, 2008; Bamunusinghe et al, 2009). However, viruses within a given family are not restricted to a particular membrane. The bromoviruses AMV and Cucumber mosaic virus (CMV) replicate on the tonoplast (Sztuba-Solinska and Bujarski, 2008). TYMV and Tobacco etch virus (TEV, family Potyviridae) replicate on the outer membrane of chloroplasts and the nuclear envelope respectively (Buck, 1996; Prod'homme et al, 2001). Interestingly, TEV replication was inhibited when the COPI and COPII secretory systems were abolished in dominant-negative mutant plants (Wei and Wang, 2008). The COPI and COPII secretory systems are responsible for Golgi to ER retrograde transport and ER to Golgi transport respectively, suggesting TEV replication also requires the early secretory 24 pathway. For many viruses, notably members of Flexiviridae such as foveaviruses, the intracellular site of replication is still unknown. Research into the site of tombusvirus replication has been extensive and has revealed membrane associations and rearrangements much more complex than initially thought. TBSV replicates in vesicles derived from peroxisomes via an inward invagination of the peroxisomal membrane (McCartney et al, 2005). At the same time, the number of peroxisomes per cell decreases as individual peroxisomes aggregate. The aggregates will then fuse, forming large "peroxisomal multivesicular bodies" in which TBSV replication takes place. Despite the fact that no other organelles were altered by TBSV infection, TBSV could still replicate in peroxisome deficient yeast cells (McCartney et al, 2005; Jonczyk et al, 2007). It was later discovered that TBSV can switch to utilizing ER membranes in the absence of peroxisomes. This phenomenon suggests that the particular membrane used for replication by a (+)RNA virus may only be preferential, not fixed and unalterable. It is still unknown why (+)RNA viruses must replicate in association with a host membrane. However, a generally accepted theory is that viral replication in host derived vesicles serves as a method to increase the local concentration of the viral replication complex, viral RNA, and auxiliary proteins in an effort to increase replication efficiency. It has also been suggested that the vesicles protect the virus from host defence responses such as RNA interference (RNAi), proteases, ribonucleases, and other antiviral responses (dos Reis Figueira et al, 2002; Molnar et al, 2005; Jonczyk et al, 2007). A third suggested function of viral replication vesicles is based on the fact that the viral (+)RNA molecule is a common substrate for both the translation of the replicase and as a template 25 for (-)RNA synthesis. While some animal (+)RNA viruses will inhibit host and eventually viral translation to allow for a switch to viral RNA replication, this temporal separation is rarely seen in plant viruses (Bushell and Sarnow, 2002). The physical separation of viral replication from viral protein translation would circumvent the potential problem of competition for the viral (+)RNA template by the replication complex and host ribosomes, as the latter would be excluded from the replication vesicles (Buck, 1996; Prod'homme et al, 2001). 1.5.5 Membrane Association There are two main strategies used by (+)RNA viruses to associate with a host membrane. One strategy is to possess a membrane association sequence within the viral replicase, while the other is to utilize a host factor that performs the actual membrane association. Potyviruses are a good example of viruses which mediate their own membrane association. The potyvirus 6 kDa replicase protein contains a typical transmembrane domain which associates the replication complex with the target membrane (Jonczyk et al, 2007). However, many (+)RNA viruses utilize host proteins to associate with their target membranes. For example, the replicase of TMV has no membrane spanning domains, but was found to interact with the Arabidopsis proteins TOM1 and TOM3 (see section 1.5.3), which contain transmembrane domains and anchor the replication complex to the ER membrane (Yamanaka et al, 2000; dos Reis Figueira et al, 2002; Yamanaka et al, 2002). The tombusvirus replication complex contains the host proteins HSP70 and Pexl9p, which are thought to promote the integration of the replication complex into peroxisome membranes (see section 1.5.4) (Pathak et al, 2008; 26 Wang et al, 2009). Semliki Forest virus (SFV, genus Alphavirus, family Togaviridae) is the model alphavirus for membrane association via a monotopic signal (Spuul et al, 2007). The nsPl replicase protein of SFV contains the MTR and HEL domains, and is responsible for the association of the viral replication complex with the endosomal membrane. The nsP 1 protein does not contain any transmembrane domains and does not interact with any known host factors that could account for membrane associations. Nuclear magnetic resonance, fluorescent quenching experiments, and mutational assays revealed that a small region within nsPl formed an amphipathic a-helix which associated with the upper leaflet of a lipid bilayer in a monotopic fashion (Figure 5A) (Lampio et al, 2000; Salonen et al, 2005; Spuul et al, 2007). The amphipathic a-helix was found to form in such a way that the cytoplasmic surface contained polar residues while the opposite surface contained non-polar hydrophobic residues (Figure 5B). The hydrophobic surface was found to sink under the upper leaflet of the bilayer with a critical tryptophan residue penetrating approximately 10 carbon atoms deep along the acyl chains of the phospholipids (Spuul et al, 2007). The hydrophilic surface, which is predominantly positively charged, is thought to interact with the phospholipid heads, specifically through a critical arginine residue. This monotopic interaction alone was found to be sufficient for membrane associations, however palmitoylation of the amphipathic a-helix, rendering it hydrophobic, enhances the membrane association and subsequent pathogenicity of SFV (Laakkonen et al, 1996; Ahola et al, 2000). The plant virus BMV also contains an amphipathic monotopic signal within the replicase protein la, which is required for the association with, and the rearrangement of, the ER during BMV infection 27 Fig. 5. Three dimensional structure of the Semliki Forest virus monotopic membrane association signal. Mechanism of membrane association by an amphipathic monotopic signal (A) and the 3-D structure of the Semliki Forest vims nsPl monotopic signal showing the amphipathic a-helix and important residues (circled) (B). The R253 residue is critical and responsible for enhancing membrane association by interacting with phosphoplipid head groups. The W259 residue is critical and responsible for anchoring the a-helix to the membrane by interacting with the lipid bilayer core. Cytoplasmic surface (above dotted line) of the amphipathic a-helix consists of polar (hydrophilic) residues while opposite surface (below dotted line) consists of non-polar (hydrophobic) residues. Adapted from Spuul et al, (2007). 28 (A) R253 V249 '-f" N-term C -term "V "*" ,L256 %W''sl)$Sh*. (B) R253J Hydrophilic ¥249^) XL^6 Hydrophobic 29 (Liu etal, 2009). Though SFV has been extensively studied with regards to . its method of membrane association, it is still unknown how plant viruses that lack transmembrane domains and do not utilize a membrane associating host proteins are able to interact with their target membrane. Other plant viruses have predicted monotopic membrane association signals within their replicase proteins; however these have yet to be demonstrated as functional (den Boon et al, 2001; dos Reis Figueira et al, 2002). In the past few years, amphipafhic monotopic signals are increasingly being predicted in, and used to explain, many instances where the current method of membrane attachment is unknown. 1.5.6 Replication of Viral RNA Despite the differences separating (+)RNA viruses into four superfamilies (see section 1.5.1), the underlying mechanism of RNA replication is essentially the same for all (+)RNA viruses. All (+)RNA viruses must first synthesize a (-)RNA strand to be used as a template for (+)RNA, and in many cases, sub-genomic (sg) RNA production (Figure 6). Positive-sense RNA, (-)RNA, and sgRNA molecules are all synthesized by the viral replication complex in an asymmetrical manner, with approximately 100 (+)RNA strands being produced from a single (-)RNA template (Nagy and Pogany, 2008; Sztuba-Solinska and Bujarski, 2008). Positive-sense RNA synthesis is thought to be favoured because the (+)RNA molecule can be used for further translation of more replicase, more (-)RNA production, or encapsidation to produce progeny virions. Since the molecular mechanics 30 Fig. 6. Replication of a positive RNA viral genome utilizing sub-genomic RNA production. Step 1: Positive RNA strand [(+)RNA, green] is used as a template,for negative RNA [(-)RNA, red] synthesis. Negative RNA synthesis is initiated at the (-) RNA promoter (yellow) located at the 3' terminus of the (+)RNA strand. Step 2: Negative RNA strand is used as a template for further (+)RNA synthesis. Positive RNA synthesis is initiated at the (+)RNA promoter (blue) located at the 3' terminus of the (-) RNA strand. Step 3: Negative RNA strand is used as a template also for sub-genomic RNA (sgRNA) synthesis. Sub-genomic RNA synthesis is initiated from sub-genomic promoters (brown) located in the interior of the viral genome. Step 4: Newly synthesized (+)RNA is used as a template for further (-)RNA synthesis and the replication cycle repeats. 31 5' 3' (+) RNA 5' (-) RNA ®l (+) RNA sgRNA 32 governing the synthesis of (+)RNA, (-)RNA, and sgRNA's are similar but not identical, they will be discussed separately. 1.5.6.1 Negative Sense Viral RNA Synthesis All (+)RNA viruses must produce a (-)RNA strand as the first step in the genome replication cycle. Transcription of the (-)RNA strand is controlled by a viral promoter that is specifically recognized by the viral replication complex. The (-)RNA promoter is located within the 3' UTR of the (+)RNA strand and has a complex secondary structure (Nagy and Pogany, 2008). Studies on potexviruses have revealed the presence of three critical stem loop structures within the 3' UTR, termed SL1, SL2, and SL3 (Verchot- Lubicz et al, 2007). SL1 is not required for (-)RNA synthesis, while SL2, SL3, and the viral poly(A) tail are essential. Two additional elements located upstream of the poly(A) tail, termed "near upstream elements 1 and 2" (NUE1 and NUE2) were found to negatively impact (-)RNA synthesis upon their removal, however not to the extent observed with the removal of SL2 or SL3 (Verchot-Lubicz et al, 2007). Experiments using members of Bromoviridae have shown that the (-)RNA promoters form tRNA-like structures which require low concentrations of the viral capsid protein to initiate (-)RNA synthesis (Sztuba-Solinska and Bujarski, 2008). It was discovered, however, that high concentrations of CP actually inhibited (-)RNA synthesis, which was later explained as a method of (-)RNA regulation (see section 1.5.6.3). Currently, a lack of knowledge and inconsistent results even among members of the same virus family preclude the construction of a universal model of (-)RNA transcription, if such a model indeed exists. However, most alphavirus-like viruses likely require the coordinated interactions of the 33 viral replicase, host factors, and viral ancillary proteins at the viral promoter, which is itself highly unique to allow for stringent selection by the replication complex. 1.5.6.2 Positive Sense and Sub-genomic Viral RNA Synthesis In contrast to the (-)RNA promoter, the (+)RNA viral promoter, located in the 5' UTR, is relatively simple and less structured (Nagy and Pogany, 2008; Sztuba-Solinska and Bujarski, 2008). No viral (+)RNA promoters have been discovered that are similar to their (-)RNA counterparts in either sequence or structure, demonstrating that the viral replication complex can recognize multiple promoters in a very specific manner. Although less complex than (-)RNA promoters, work with potexviruses have revealed that (+)RNA promoters also contain stem loop structures, although only two, SL1 and SL2 (Verchot-Lubicz et al, 2007). Both SL1 and SL2 are essential for viral replication, with SL1 being multifunctional and able to bind host factors. Since SL1 is required for (+)RNA synthesis, it has been suggested that host transcription factors likely bind to this element during replication. The 3' UTR is also required for (+)RNA synthesis, suggesting a possible interaction between the 5' and 3' UTR's during replication. Sub-genomic RNAs are produced by most alphavirus-like viruses to efficiently translate downstream ORFs. Viruses with monopartite genomes, such as TMV, TBSV, and PVX, usually produce two or more sgRNAs to express movement and capsid proteins (dos Reis Figueira et al, 2002; Verchot-Lubicz et al, 2007; Nagy and Pogany, 2008; Sztuba-Solinska and Bujarski, 2008). Viruses with multipartite genomes, such as BMV and AMV, usually produce one sgRNA to express the capsid protein (Sztuba- Solinska and Bujarski, 2008). sgRNAs are transcribed using 30-60 nt sub-genomic 34 promoters, which are generally located on the (-)RNA strand immediately upstream of the gene to be transcribed. Specifically, sub-genomic promoters can be located partly within the upstream ORF and/or in an intergenic region. The replication complex recognizes stem loops and surrounding sequence within sub-genomic promoters and transcribes sgRNAs via internal initiation on the (-)RNA strand (Sztuba-Solinska and Bujarski, 2008). The resulting sgRNAs are then used solely as translational templates to express a single protein, although some members of Flexiviridae produce bicistronic sgRNA molecules (Martelli et al, 2007; Sztuba-Solinska and Bujarski, 2008). Various components of the viral replication complex have been found to interact with sub-genomic promoters. The p33 protein of TBSV, 126K protein of TMV, and la protein of BMV have all been shown to bind sub-genomic promoters in vitro (Goregaoker and Culver, 2003; Pogany et al, 2005; Nagy and Pogany, 2008). In addition to the viral replication complex, the viral RNA itself has been found to interact with sub- genomic promoters in a cw-acting manner. The 5' UTR and 3'UTR of PVX, required for sgRNA production, were both found to interact with the sub-genomic promoters in what is thought to be direct base-pairing (Verchot-Lubicz et al, 2007). 1.5.6.3 Regulation of Replication Upon infection, the viral (+)RNA is first used by ribosomes to translate the replicase protein(s). It is then used by the replication complex to produce (-)RNA, which is in turn used by the replication complex to produce (+)RNA and sgRNA. These RNAs are then used by ribosomes to produce more viral proteins, and finally packaged into progeny virions (Shin et al, 2009). Not surprisingly, (+)RNA virus replication is finely 35 regulated. While the specificity of replication is governed primarily by the unique viral promoters, the required switch between replication and translation is controlled by many factors including but not limited to the physical separation of the replication complex and ribosomes (see section 1.5.5), viral RNA, viral proteins, host factors, and enzymatic modifications (Jakubiec and Jupin, 2007; Nagy and Pogany, 2008; Sztuba-Solinska and Bujarski, 2008). Among plant viruses, regulation has been studied most extensively in bromoviruses. The BMV la protein, which contains the MTR and HEL domains of the replicase, down-regulates its own production by shuttling the (+)RNA away from the ribosomes to the ER for replication (Sztuba-Solinska and Bujarski, 2008). During BMV replication, the 5' UTR, sub-genomic promoter, and CP protein have all been implicated in the switch from (+)RNA to sgRNA production, as well as the control of strand synthesis asymmetry. Strand asymmetry in AMV is controlled in part by CP binding of the tRNA-like (-)RNA promoter in the 3' UTR. CP binding causes a conformational change in the tRNA-like structure, resulting in the RdRp region of the replicase no longer being able to recognize the (-)RNA promoter. Thus, at high concentrations, the CP protein acts to repress (-)RNA synthesis, effectively shutting down sgRNA and viral protein production. Indeed, the CP was found to be required for (+)RNA release from the replication complex into the cytoplasm for encapsidation (Sztuba-Solinska and Bujarski, 2008). Phosphorylation of the replication complex has also been shown to substantially affect viral replication. Many viruses phosphorylate the RdRp domain to downregulate or completely shut off viral genome replication (Jakubiec and Jupin, 2007). The animal virus, Dengue virus 2 (family Flaviviridae), produces multiple forms of the RdRp protein 36 (NS5), differing only among their degrees of phosphorylation (Jakubiec and Jupin, 2007). The hyperphosphorylated form localizes to the cell nucleus, while the hypophosphorylated form localizes to the cytoplasm. The HEL domain in the replicase protein NS3 would interact with only cytosolic NS5, suggesting that phosphorylation of the RdRp is used to inhibit/disassemble the replicase. Similar results were reported for the plant virus CMV. Studies on TYMV have also shown phosphorylation of the RdRp to be responsible for the inhibition and disassembly of the replicase late in the infection cycle. However, rather than inhibiting interactions with other replicase components, phosphorylation of the TYMV RdRp protein (66K) activates a degradation signal (Jakubiec et al, 2006). When bound to the HEL domain, the 66K protein cannot be phosphorylated, suggesting another mechanism must first separate the replicase components before the 66K protein can be phosphorylated and degraded. When TYMV was used to infect phosphorylation deficient yeast mutants, in which the replicase should theoretically be ultra-stable, replication was inhibited. Critical alterations to strand asymmetry were discovered, reinforcing the notion that phosphorylation is used to fine- tune the mechanics of viral replication. Phosphorylation is also used to control RNA binding. The CNV p33 RNA- binding protein was shown to be phosphorylated in the replication complex. Phosphorylation of p33 in vitro, as well as in yeast and plant cells, showed a reduced capacity to bind RNA, likely due to the neutralizing of a positively charged RNA binding pocket (Stork et al, 2005). Phosphorylation was therefore speculated to be involved in shutting off RNA replication late in the infection cycle. Indeed, phosphorylation of p33 37 of CNV and nsP3 of SFV was found to be required for wildtype levels of replication and pathogenicity (Shapka et al, 2005). These results are similar to those observed for . TYMV, and given the differences among these plant and animal viruses, the notion that phosphorylation of the replicase components may be a common feature among all (+)RNA viruses is not unfounded. 1.6 Grapevine Rupestris Stem Pitting-associated Virus 1.6.1 Physical Characteristics GRSPaV is a member of the family Flexiviridae and the genus Foveavirus, "fovea" being Latin for "pit", a symptom characteristic of foveavirus infections. Other members of Foveavirus include the species Apple stem pitting virus, Apricot latent virus, Peach sooty ringspot virus, and putatively African oil palm ringspot virus, Asian prunus virus-1, Asian prunus virus -2, and Asian prunus virus -3 (Martelli and Jelkmann, 1998; Adams et al, 2004; Martelli et al, 2007). GRSPaV is a flexuous filamentous particle displaying helical symmetry and forming a particle of approximately 723 nm long by 13 nm in diameter (Figure 7A) (Petrovic et al, 2003). GRSPaV ranks among the members of Flexiviridae with the largest genomes, a position which is echoed by the virus' large replicase polyprotein (see section 1.6.3). 1.6.2 Molecular Characteristics Like all members of Flexiviridae, GRSPaV has a monopartite positive sense ssRNA genome of 8725 nt containing a putative 5' 7-methylguanosine cap and a 3' poly- 38 Fig. 7. Physical and molecular structure of Grapevine rupestris stem pitting- associated virus. Electron micrograph image showing GRSPaV virion (A) and schematic diagram of GRSPaV genome (B). GRSPaV genome diagram shows the 5' and 3' untranslated regions (UTR), five replicase domains, the triple-gene block (TGB) genes, and the capsid (CP) gene. Numbers are the amino acid (aa) boundaries of replicase regions. MTR, methyltransferase; HVR, highly variable region; O-Pro, ovarian tumour protease; P-Pro, papain-like cysteine protease; HEL, helicase; POL, RNA-dependent- RNA-polymerase. Electron micrograph adapted from Petrovic et al, (2003). Bar = 100 nm. 39 tt H 3 CO IT) Q. m O 1 -1 1 • ^* •" • * • • r^ • * o*H • • • • • oI_ : Q. - • 1 •• 0. <* a • or-o» « o • •• o • •• • • Q1. • oo • • • o CQ • £ o r*IT)- a cc a > X 45 1 g i- S m r-\ Q: \- CO > • =3 TOCD in 40 adenylate tail (Meng et al, 1998; Zhang et al, 1998). GRSPaV has five ORFs arranged in tandem, encoding the viral replicase polyprotein (see section 1.6.3), three movement proteins (TGBpl, TGBp2, and TGBp3, see section 1.6.4), and the coat protein (see section 1.6.5). The ORFs are flanked by a 5' UTR and 3' UTR of 60 nt and 176 nt respectively (Meng et al, 1998; Zhang et al, 1998; Meng et al, 2005). The 5' UTR is highly conserved among all GRSPaV variants (see section 1.6.7) sequenced to date, supporting the belief that it is involved in viral replication (Meng et al, 2005). In contrast, the 3' UTR is more variable among sequence variants, suggesting a degree of flexibility in the importance of functionality. 1.6.3 ORF1: The viral replicase ORF1 immediately follows the 5' UTR and spans nucleotides 61-6546 (Meng et al, 1998; Zhang et al, 1998). Putatively encoded by this ORF is a polypeptide of approximately 244 kDa (2161 aa) containing three signature and three additional motifs of an alphavirus-like replicase: a methyltransferase (MTR) domain, an AlkB-like domain, an ovarian tumour-like protease (O-Pro), a papain-like cysteine protease (P-Pro), a helicase (HEL) domain, and an RdRp (POL) domain (Figure 7B) (Gorbalenya et al, 1988; Koonin, 1991; Rozanov et al, 1992; Meng et al, 1998; Zhang et al, 1998; Martelli et al, 2007). The MTR domain, located at aa 1-372, is a type 1 methyltransferase domain likely involved in the addition of a 5' cap to the viral (+)RNA and sgRNA molecules (Meng et al, 1998; Zhang et al, 1998; van derHeijden and Bol, 2002). Following the MTR domain is a highly variable region (HVR, aa 451-750) of unknown significance (Meng et al, 41 2005). Directly downstream of the HVR is a putative AlkB-like domain (aa 780-874), possibly involved in RNA methylation damage repair (see section 1.4.6). The O-Pro and P-Pro domains reside downstream of the AlkB-like domain at aa 1074-1175 and aa 1184- 1265 respectively. One or both of these domains are expected to auto-catalytically cleave the polyprotein somewhere between the HEL and POL domains to activate the replicase (see section 1.5.2), however this has yet to be demonstrated. Indeed, cleavage by the O- Pro domain has not been observed in any plant virus to date despite its prevalence in carlavirus and foveavirus genomes (Martelli et al, 2007). The HEL domain (aa 1346- 1624) likely encodes a type 1 RNA helicase involved in removing RNA secondary structure and separating the dsRNA intermediate molecules during replication (Gorbalenya et al, 1988; Meng et al, 1998; Zhang et al, 1998). The POL domain (aa 1870-2155) is located at the C-terminus of the replicase polyprotein and is putatively a type III RNA-dependent-RNA-polymerase and is thought to be responsible for the replication of (+)RNA, (-)RNA, and sgRNA in a viral specific manner (Koonin, 1991; Meng et al, 1998; Zhang et al, 1998). 1.6.4 ORF2, ORF3, and ORF4: The viral movement proteins ORF2, ORF3, and ORF4, collectively known as the "triple gene block" (TGB), encode putative viral movement proteins responsible for cell-to-cell and systemic spread of the virus throughout the infected plant (Meng et al, 1998; Zhang et al, 1998). The product of ORF2 (nt 6577-7242) is TGBpl, a 24.4 kDa protein containing signature motifs associated with NTPase, helicase, and RNA-binding activity. TGBpl is therefore thought to be the primary movement protein (Meng and Gonsalves, 2008). The product 42 of ORF3 (nt 7244-7597) is TGBp2, a 12.8 kDa ancillary protein which was found to interact with the ER when expressed alone (Rebelo et al, 2008). Indeed, TGBp2 contains two transmembrane domains likely for this purpose (Meng et al, 1998; Zhang et al, 1998). ORF4 (nt 7519-7761) overlaps ORF3 and is putatively translated via leaky scanning of ORF3, albeit at a much lower level than ORF3. The resulting product, TGBp3, is an 8.4 kDa protein also thought to assist TGBpl in viral translocation. TGBp3 contains a single transmembrane domain and was also found to localize to the ER when expressed alone (Meng et al, 1998; Zhang et al, 1998; Rebelo et al, 2008). While it is unknown how all three TGB proteins work in concert to translocate viral RNA or virions cell-to-cell, it has been proposed that TGBpl interacts with the viral cargo and associates with TGBp2-derived vesicles, which are then transported to the cell periphery by TGBp3 and exported by TGBpl (Rebelo et al, 2008). The interaction of the GRSPaV TGBpl with either TGBp2 or TGBp3 has yet to be demonstrated. 1.6.5 ORF5: The viral capsid protein The fifth and last ORF (nt 7770-8549) encodes the viral capsid protein, indicated by the presence of a conserved (+)RNA plant viral coat protein motif (Meng et al, 1998; Zhang et al, 1998; Meng and Gonsalves, 2008). Additionally, antibodies have been raised against the recombinant GRSPaV capsid protein which are capable of specifically decorating the GRSPaV virion, confirming its identity and function (Petrovic et al, 2003). Possible additional functions of the GRSPaV CP have not yet been elucidated. 43 1.6.6 Symptoms and Transmission GRSPaV is the putative causal agent of RSP, and is therefore associated with the formation of strips of small pits on the woody tissue directly under the bark of graft- inoculated V. Rupestris "St. George" (Martelli, 1993; Meng and Gonsalves, 2008). Whether GRSPaV is the sole agent responsible for RSP is currently unknown. Since grapevines can be infected with up to 55 viruses from seven different families, and many viruses act in concert to produce disease symptoms, it is very likely that GRSPaV is required for but not solely responsible for the symptoms seen in RSP infected vines. Indeed, GVA has been proposed to work in concert with GRSPaV to produce RSP symptoms (Petrovic et al, 2003). GRSPaV is thought to be an ancient virus that has co-evolved with grapevines for thousands of years. The viral replicase is quite complex, having at some point over evolutionary time acquired an Alk-B-like domain and an O-Pro domain, both of which are absent in most plant viruses (Martelli et al, 2007; Rebelo et al, 2008). The Alk-B-like domain is similar to that found in all higher eukaryotes, including Vitis species, suggesting GRSPaV may have acquired this domain due to prolonged interactions with its host (Martelli et al, 2007; Rebelo et al, 2008). Perhaps more importantly, GRSPaV is not very pathogenic, producing little to no symptoms in some instances and never killing the infected host (Meng and Gonsalves, 2008). Such behaviour is indicative of a long co- evolutionary history between the virus and its host. While the natural mode of transmission of GRSPaV is unknown, grafting has been shown to effectively transmit the virus (Martelli et al, 2007; Meng and Gonsalves, 2008; Osman and Rowhani, 2008). Graft transmission is common among members of 44 Flexiviridae that infect woody plants, and is exemplified by the fact that GRSPaV is found in many different grape cultivars following the widespread use of grafting in the viticulture industry (see section 1.1) (Meng and Gonsalves, 2008). GRSPaV displays a seasonal titre fluctuation in infected grapevines, attaining the highest titre in the summer months when the grapevine is actively growing (Martelli et al, 2007; Meng and Gonsalves, 2008; Osman and Rowhani, 2008). During this time, the virus has been detected in the leaves and phloem, and is likely also present in pollen spores. However, transmission via pollen is not common with plant viruses, so it is not likely that this route represents the main mode of transmission for GRSPaV. Until an insect or other type of vector is discovered, how GRSPaV is spread in the wild will remain a mystery. 1.6.7 Population Structure As more GRSPaV isolates were cloned and sequenced, two main aspects of the virus' population structure were revealed: 1. GRSPaV exists not as a single strain, but as a population of sequence variants; 2. certain sequence variants are found in certain Vitis species but not others (Meng et al, 1999; Meng et al, 2005; Meng et al, 2006; Lima et al, 2006). Based on nucleotide and amino acid sequence alignments, GRSPaV can be divided into four sequence variant families: GRSPaV-1, GRSPaV-SG, GRSPaV-VS, and GRSPaV-BS. Interestingly, these four sequence variants seem to be associated with specific Vitis species. For example, GRSPaV-1, GRSPaV-SG, and GRSPaV-VS are associated with V. riparia, V. rupestris, and V. sylvestris respectively (Meng and Gonsalves, 2008). GRSPaV-BS is not associated specifically with any one Vitis species. Many grapevine varieties are propagated via grafting, and so the plant itself is 45 made up of a rootstock from one species and a scion (the grape producing vine above the graft junction) from another species. Studies examining the population structure of GRSPaV in rootstocks versus scions revealed that while rootstock vines were infected by only one sequence variant, scion vines were more often than not infected with a mixed population of GRSPaV sequence variants (Meng et al, 2006; Meng and Gonsalves, 2008). The most likely explanation for these findings is that the mixed populations are a result of decades of scion grafting, trading, and further grafting onto different rootstocks. All Vitis species may have originally contained a single sequence variant of GRSPaV, but over time, as scions were grafted onto different rootstocks, the individual GRSPaV sequence variants from each rootstock and scion were mixed, eventually leading to a large random mixture of GRSPaV sequence variants in modern day scion varieties (Meng and Gonsalves, 2008). This theory is supported by the findings that wild North American grapevines tend to be infected with only one GRSPaV sequence variant per species (Meng and Gonsalves, 2008). However, it is not unlikely that further analysis and sequencing of GRSPaV isolates may reveal additional sequence variant families and different population structures within wild and cultivated Vitis species. 1.7. Hypotheses Based on results reported for other similar plant viruses and the above information on positive RNA virus replication in general, it was hypothesized that upon transient expression of the GRSPaV replicase polyprotein in Nicotiana tabacum cv. Bright Yellow (BY)-2 cells, the replicase would undergo auto-proteolytic processing and localize to viral replication complexes (likely on the ER) via a sub-domain (aa 132-207) located in 46 the methyltransferase domain. Furthermore, the sub-domain was expected to be able to localize independently of the rest of the replicase due to the presence of a monotopic membrane-assoication signal. An arginine residue (R173) and a tryptophan residue (W184) were thought to be critical for the proper functioning of the monotopic signal, and thus hypothesized to be required for localization. 47 CHAPTER 2: MATERIALS AND METHODS 2.1 Polymerase Chain Reaction (PCR) Unless otherwise stated, all primers were ordered from Sigma-Aldrich (Oakville, Ontario, Canada) and hydrated to a stock concentration of 100 uM. Standard PCR reactions used to amplify DNA sequences for cloning contained the following: 50 ng of template DNA, IX KOD Hot Start reaction buffer (Novagen, La Jolla, California, USA), 0.2 mM of each dNTP, 0.2 uM of each primer, 1 unit of KOD Hot Start DNA Polymerase (Novagen), and sterile dH20 in a final volume of 50 uL. Thermocycling was performed in a Peltier Thermal Cycler PTC-200 (MJ Research, Waltham, Massachusetts, USA) using the following parameters: 95°C for 5 minutes, 25 cycles of {95°C for 30 seconds, 55°C for 30 seconds, 72°C for 30 seconds}, and finally 72°C for 10 minutes. Successful amplification was verified by mixing 5 uL of the PCR product with 6X DNA loading buffer (0.25% bromophenol blue, 30% glycerol) and electrophoresing on a 1% (w/v) agarose gel (Bioshop, Burlington, Ontario, Canada) containing 1 ug/uL ethidium bromide. Electrophoresis was performed in 0.5X electrophoresis buffer (45 mM Tris- HC1, 89 mM boric acid, 1 mM EDTA) at 85 V for 60 minutes followed by visualization with a BioRad (Mississauga, Ontario, Canada) GelDoc. 2.2 Restriction Digestions Standard restriction digestions were performed by mixing 1 [ig of DNA with the appropriate IX digestion buffer (Promega, Madison Wisconsin, USA), 10-12 units of the appropriate restriction enzyme (Promega), and sterile dH20 in a total volume of 50 uL. If Invitrogen (Burlington, Ontario, Canada) restriction enzymes were used, 100 ug/mL 48 BSA was also added to the reaction. The reaction was then incubated at the permissive temperature for the enzyme in use, usually 37°C, for 2-4 hours and 5 uL of the reaction was checked on an agarose gel to determine if digestion was complete. Once complete, the reactions were stopped via heat inactivation at 90°C for 10 minutes. For restriction digestions used to analyze potentially recombinant plasmids, the above reaction was performed in a total volume of 20 uL, and incubation times were reduced to 1 hour. 2.3 PCR Purification and Gel Extraction of DNA All PCR purification and gel extractions were performed using a Wizard PCR Cleanup/Gel Extraction Kit (Promega) following the manufacturer's instructions. Purified or gel extracted DNA was assayed for quantity and quality via spectrophometry on a NanoDrop ND-1000 (Fisher Scientific, Ottawa, Ontairo, Canada). Five microlitres of DNA was also examined by agarose gel electrophoresis along with an O'GeneRuler 100 bp or 1 kb DNA ladder (Fermentas, Burlington, Ontario, Canada) for visual estimation of quality and quantity. 2.4 De-phosphorylation, Ligation, and Transformation of DNA To prevent re-ligation of vectors digested with only one restriction enzyme, completed digestion reactions were mixed with 3 units of Antarctic phosphatase (New England Biolabs (NEB), Pickering, Ontario, Canada), IX phosphatase buffer, and sterile dFhO to bring the total reaction volume to 60 uL. The phosphatase reaction was allowed to incubate at 37aC for 15 minutes, and then stopped via incubation at 70°C for 10 minutes. Five microlitres was then examined on an agarose gel to estimate quality and 49 quantity. Various ligation "insert to vector" ratios were performed depending on the particular construct being cloned. Typical ratios used were 1:1 and 3:1, with a 0:1 ratio always being included as a control to assess to degree of incomplete digestion and/or re- ligation of the vector. Three hundred nanograms of vector DNA was used and the appropriate weight of "insert" DNA was calculated accordingly and added to the reaction. Ligation reactions also contained 1 unit of T4 DNA ligase (Promega), IX final ligation buffer, and sterile dF^O in a total volume of 10 uL. Ligation reactions were incubated at 4°C for 18 hours. After 18 hours at 4°C, 2 uL of each ligation reaction was used to transform 50 uL of chemically competent E. coli DH5a or JM109. The cells were mixed with the ligation reaction, incubated on ice for 20 minutes, heat shocked in a 42°C water bath for 45 seconds, and incubated on ice for a further 2 minutes. Eight hundred microlitres of Luria-Bertani (LB, 1% tryptone (Fisher Scientific), 0.5% yeast extract (Fisher Scientific), 1% NaCl, pH 7.0) was then added to the cells, which were allowed to recover at 37°C for 45 minutes prior to plating on LB agar plates containing 100 ug/mL ampicillin (Sigma- Aldrich). The plates were incubated at 37°C for 16 hours and the resulting colonies were used to inoculate 5 mL of LB containing 100 [ig/mL ampicillin, which was then cultured at 37°C with shaking (250 RPM) for 12-18 hours. To make glycerol stocks, 500 uL of the 5 mL overnight culture was removed, mixed with 500 uL of 80% sterile glycerol, flash frozen in liquid nitrogen, and stored at -80°C. The rest of the 5 mL overnight cultures were then used for both colony PCR screens and recombinant plasmid recovery via miniprep. 50 2.5 Generation of Chemically Competent E. coli To prepare chemically competent E. coli DH5a, a single colony was used to inoculate 50 mL of LB containing 10 raM MgS04 in a 125 mL flask. This culture was grown at 37°C, 250 RPM, for 16 hours, after which 1 mL was sub-cultured into 100 mL LB containing 10 tnM MgS04 in a 250 mL flask. The 250 mL culture was incubated at 37°C, 250 RPM, until the OD600 was between 0.5 and 0.6 (~3 hours). The OD60o was measured by removing a 1 mL aliquot and measuring cell density on a SmartSpec Plus (BioRad) spectrophotometer. The culture was then added to two 50 mL conical tubes that were pre-chilled on ice. All materials and reagents are kept on ice and under sterile conditions from this point on. The cells were centrifuged at 2000 X g for 10 minutes at 4°C and re-suspended via pipetting, in 17 mL of filter sterilized RF1 solution (100 mM RbCl, 50 mM MnCl2-4H20, 30 mM KOAc, 10 mM CaCl2-2H20, 15% glycerol, pH 5.8). After a 1 hour incubation on ice, the cells were centrifuged at 1000 X g for 15 minutes at 4°C and re-suspended in 4 mL of filter sterilized RF2 solution (10 mM MOPS (Sigma- Aldrich), 10 mM RbCl, 75 mM CaCl2-2H20, 15% glycerol, pH 6.8). The cells were incubated on ice for 15 minutes and then aliquoted into pre-chilled 1.5 mL microfuge tubes, snap frozen in liquid nitrogen, and stored at -80°C. 2.6 Generation and Transformation of Electrocompetent E. coli E. coli JM109 and SURE2 (Stratagene, La Jolla, California, USA) cells were used to house the full-length replicase construct (pRSP-REP:GFP, see section 2.12.1) as DH5a proved unstable for this purpose (see Appendix I). To make the cells competent for transformation, a single colony was used to inoculate 50 mL of LB in a 125 mL flask, and 51 cultured at 37°C for 16 hours with shaking (250 RPM). One millilitre of this culture was then used to inoculate 500 mL of LB in a 2 L flask, which was then cultured at 37°C with shaking (250 RPM). When the OD600 was between 0.4 and 0.6, the culture was chilled on ice for 10 minutes. The culture was then centrifuged in an Avanti J-E floor centrifuge (Beckman Coulter, Mississauga, Ontario, Canada) (JS-5.3 rotor) at 4°C, 4000 x g, for 10 minutes, and the pellet was re-suspended in 100 mL cold, sterile dLbO. The cells were centrifuged again as above and washed twice more with cold, sterile dH^O. A final wash with 20 mL cold, sterile 10% glycerol was performed before the pellet was re-suspended in 1 mL of cold, sterile 10% glycerol and aliquoted (in 40 uL) into pre-chilled 1.5 mL microfuge tubes. The aliquoted cells were flash frozen using liquid nitrogen and stored at -80°C. For transformation, an aliquot of cells (40 uL) was thawed on ice and mixed with 10-200 ng of DNA. After incubation on ice for 1 minute, the cell/DNA mixture was transferred to a pre-chilled sterile 1 mm gap electroporation cuvette (BioRad) and electroporated in a Gene Pulser Xcell electroporation unit (BioRad) using the following parameters: 1800 V, 25 uF, 200 Q, 1 pulse. The cells were then returned to ice for one minute, after which 800 uL of LB was added to the cuvette and the cells were allowed to recover at 37°C for 1 hour with shaking (250 RPM). Dilutions of the cells were then plated on LB agar containing the appropriate antibiotic and incubated at 37°C for 16 hours. 2.7 Tobacco Bright Yellow (BY-2) Tissue Culture Nicotiana tabacum cv Bright Yellow (BY-2) cell cultures were maintained in the 52 following manner: BY-2 culture media [4.3g/L Murashige Skoog (MS) Salts (Sigma- Aldrich), 87.6 mM sucrose (Fisher Scientific), 0.56 mM myo-inositol (Sigma-Aldrich), 1.87 mM KH2P04, 200 ng/L 2,4-dichlorophenoxyacetic acid (2,4-D), 2.96 uM thiamine- HC1, pH 5.0] was prepared by aliquoting 50 mL into 125 mL tissue culture flasks (Corning Inc., Corning, New York, USA) and autoclaving at 120°C for 20 minutes. Sterile aliquots were stored at 4°C in the dark for up to one month prior to use. BY-2 suspension cell cultures were maintained by sub-culturing 2-4 mL into fresh sterile media every 7 days. Sub-culturing was performed in a class II type A2 Nuaire bio-hood using sterile 5 mL pipettes. The volume of cells used to sub-culture was dependent on the growth rate of the cells each week, and was thus determined independently each week. A packed-cell-volume of 10 mL resulting from centrifugation of 50 mL of BY-2 cell culture on the forth day was considered optimal growth. 2.8 Isolation of Tobacco BY-2 Protoplasts The following protocol was originally based on that of Zhong, et al. (2005). Twenty-five millilitres of four day old BY-2 cells were collected by centrifugation in a 50 mLconical tube at 150 x g for 6 minutes at room temperature. All centrifugations were performed in a swing bucket rotor (A-4-44, Eppendorf, Mississauga, Ontario, Canada) in an Eppendorf 581 OR bench top centrifuge. The supernatant was removed with a 25 mL pipette to avoid disturbing the cell pellet, which should be about 5 mL if the cell growth was optimal. Meanwhile, 20 mL of digestion buffer [0.45 M D-Mannitol (Sigma- Aldrich), 3.6 mM (2[N-morpholino] ethanesulfonic acid) (MES) (Sigma-Aldrich), 1.5% Onozuka RS Cellulase (Yakult Pharmaceutical, Tokyo, Japan), 0.2% Macerase 53 (Calbiochem, EMD Biosciences, La Jolla, CA,), pH 5.5] was prepared in a 125 mL flask by swirling on a shaker at 25°C, 200 RPM for 10-20 minutes until the enzymes were completely dissolved. The digestion buffer was then gently poured onto the cell pellet and re-suspended via gently rocking by hand. All subsequent re-suspensions were performed in this manner. The 50 mL conical tube was sealed with parafilm and rocked on a fixed speed Gyromini nutating mixer (Labnet International Inc., Edison, New Jersey, USA) at 37°C in the dark for 2 hours. After 2 hours, the state of digestion was checked under a microscope. All microscope slides were prepared using 1 mL wide-mouth pipette tips to avoid shearing of the protoplasts. Digestion was considered complete when 90%- 100% of the BY-2 cells were protoplasts. If digestion was incomplete, the cells were incubated as above for an addition 20 minutes and checked again until digestion was complete. Upon complete digestion, the protoplasts were collected by centrifugation at 150 x g for 8 minutes at room temperature. The supernatant was decanted with a 25 mL pipette and the protoplasts were washed with 20 mL of wash buffer [0.45 M D-Mannitol (Sigma-Aldrich), 3.6 mM MES (Sigma-Aldrich), pH 5.5] by re-suspending the pellet via gentle rocking. The protoplasts were centrifuged as above and the supernatant removed with a 25 mL pipette. The pellet was washed/centrifuged a second time as above, then gently re-suspended in 5 mL of electroporation buffer [0.45 M D-Mannitol (Sigma- Aldrich), 3.6 mM MES (Sigma-Aldrich), 0.051 mM CaCl2, pH 5.5]. The protoplasts were then incubated on ice for 1-2 hours. 2.9 Transfection of Tobacco BY-2 Protoplasts Transfection of protoplasts was accomplished via electroporation. For each 54 sample, 1 mL of protoplasts was added to a 4 mm gap electroporation cuvette that has been pre-chilled on ice. Thirty micrograms of plasmid DNA was added to the protoplasts and mixed via gentle inversion of the capped cuvette. The plates of the cuvette were dried with a Kim Wipe prior to electroporation and the protoplast-DNA mixture was further mixed. Electroporation was then performed using the following parameters: 1 pulse, 180 V, 150 |iF, 100 Q. Immediately after electroporation, the cuvette was placed back on ice for one minute. The protoplasts were mixed via inversion and poured into a 1.5 mL microfuge tube. Five hundred microlitres of electroporation buffer was added to the cuvette to collect residual protoplasts and poured into the same 1.5 mL microfuge tube as the original sample. The protoplasts were centrifuged at room temperature at 50 X g for 10 minutes in a swing bucket rotor by inserting the 1.5 mL microfuge tubes into empty 15 mL conical tubes that fit the rotor. The supernatant was removed using a 1 mL pipette, and the pellet re-suspended in 750 (J.L of protoplast culture media [0.45 M D- Mannitol (Sigma-Aldrich), 4.3g/L MS salts (Sigma-Aldrich), 87.6 mM sucrose (Fisher Scientific), 0.073 mM KH2P04, 0.56 mM myo-inositol (Sigma-Aldrich), pH 5.5]. The protoplasts were then poured into 30 mm tissue culture plates (Sarstedt, Montreal, Quebec, Canada) that had been layered with protoplast culture media containing 1.5% agar. The plates were then sealed with parafilm and incubated at 25°C in the dark for the appropriate time post transfection. 2.10 Biolistic Bombardment and Immunostaining of Tobacco BY-2 Cells 2.10.1 Biolistic Bombardment Fluorescent immunostaining (see section 2.10.2) required prior transfection of 55 tobacco BY-2 cells via biolistic bombardment. Spermidine (Sigma-Aldrich) was diluted in sterile water to a final concentration of 0.1 M and stored in 500 uL aliquots at -20°C. Prior to use, the required amount of spermidine was thawed in a 42°C water bath. Stocks of 2.5 M CaCb were prepared by diluting anhydrous CaCb in sterile water and storing as 500 uL aliquots at -20°C. 10% Triton X-100 stocks were made by diluting Triton X-100 (Sigma-Aldrich) in IX PBS and storing at room temperature. Tungsten M17 microcarriers (BioRad) were prepared by placing 100 mg in a glass test tube and baking at 180°C for 12-18 hours. The particles were then transferred to a 1.5 mL microfuge tube and vortexed vigorously with 1 mL of 100% isopropanol. The microcarriers were then deagglomerated in a sonicating water bath for at least 40 seconds and then vortexed for 3 minutes. Following incubation at room temperature for 15 minutes, the microcarriers were centrifuged at 16,000 X g for 15 seconds. The supernatant was removed using a pulled Pasteur pipette and the pellet was re-suspended in 1 mL of sterile water via vortexing for 1 minute. The microcarriers were then incubated at room temperature for 1 minute and centrifuged and washed as above twice more for a total of three washes. The pellet was finally re-suspended in 1 mL of sterile 50% glycerol via vortexing for one minute. The microcarriers were stored as 50 uL aliquots in 0.5 mL microfuge tubes at - 20°C. For biolistic bombardment, 10 ug of DNA was mixed with an aliquot (50 uL) of microcarriers by vortexing for 30 seconds. Fifty microlitres of CaCb was then added to the microcarrier/DNA mixture and vortexed for 30 seconds. Twenty microlitres of spermidine was added to the mixture and vortexed for 3 minutes then allowed to settle at room temperature for 1 minute. The microcarriers were pelleted via centrifugation at 56 16,000 X g for 15 seconds, and the supernatant was removed using a pulled Pasteur pipette. One hundred and fifty microlitres of 100% isopropanol was then mixed with the microcarriers by vortexing for 1 -3 minutes, followed by incubation at room temperature for 1 minute. The, microcarriers were centrifuged as above, and the pellet was re- suspended in 48 uL of 100% isopropanol by vortexing for 1 minute. The prepared microcarriers were stored on ice until use. To prepare the tobacco BY-2 cells for biolistic bombardment, 50 mL of four day old culture was transferred to a 50 mL conical tube and centrifuged at room temperature for 5 minutes at 200 X g. The supernatant was poured off and the cells were re- suspended in an equal volume (to the packed cell volume) of BY-2 transformation buffer [4.3g/L MS Salts (Sigma-Aldrich), 87.6 mM sucrose (Fisher Scientific), 0.56 mM myo inositol. (Sigma-Aldrich), 0.935 uM KH2P04, 2.96 uM thiamine-HCl, 4 M sorbitol (Sigma-Aldrich), 4 M D-Mannitol (Sigma-Aldrich)]. The medium for bombardment was prepared by lining a Petri dish lid with three pieces of sterile 9 cm (diameter) filter paper (Fisher Scientific). The filter paper stack was wet with 4 mL of BY-2 transformation buffer and flattened to remove any air bubbles. Five millilitres of the prepared BY-2 cells was added to the filter paper and allowed to incubate in the dark for 45 minutes. Biolistic bombardment was then performed using a PDS1000 He Biolistic Particle Delivery system (BioRad) following the manufacturer's instructions. Briefly, 20 uL of the microcarrier/DNA mixture was added to the centre of a sterile macrocarrier disk (BioRad) and allowed to dry. Duplicate samples were prepared and bombarded into each cell sample, turning the dish 180 degrees between each shot. Following bombardment, the dish was covered, wrapped in parafilm, and incubated in the dark at 25°C for the 57 desired length of time. 2.10.2 Fixation and Immunostaining Fresh fixative solution (4% formaldehyde in 0.5X BY-2 transformation buffer) was prepared prior to fixation (10 mL/sample). The BY-2 cells were scraped off the filter paper using a bent scoopula and transferred to a 15 mL conical tube containing 10 mL of fixative. The cells were mixed at room temperature on a nutating mixer for 45 minutes. To collect the cells, the sample was spun in an IEC clinical swing bucket centrifuge on setting 4 for 45 seconds. The cells were then re-suspended in 10 mL of IX PBS and centrifuged as above. The cells were re-suspended again in IX PBS, rocked for 1.5 minutes, and centrifuged as above. The pellet was washed twice more for a total of 4 washes. For the antibodies to penetrate into the cells, the cell walls had to first be removed and the cell membrane permeabilized. To remove the cell wall, the cell pellet was re-suspended in 10 mL of 0.1 mg/mL pectinase Y-23 (Sigma-Aldrich, dissolved in IX PBS) and incubated on a nutating mixer at room temperature for 2 hours. The cells were then washed with IX PBS as above for a total of 3 washes and re-suspended at the end in 5 mL of IX PBS. To permeabilize the cell membrane, 1 mL of cell was transferred to a 1.5 mL microfuge tube and mixed with 33 uL 10% triton X-100. The cells were incubated on a nutating mixer at room temperature for 15 minutes. The cells were washed 3 times as above by re-suspending the pellet in 1 mL of IX PBS. After the final centrifugation, the cells were re-suspended in 1 mL of anti-RdRp rabbit primary antibody (diluted in IX PBS by a factor of 1:1000) and incubated on a nutating mixer at room temperature for 1 hour. The anti-RdRp antibody was 58 commercially produced by GenScript Corp by first synthesizing the polypeptide 'GKGLDELNEWVRTRG', which covers aa 1927-1941 of the GRSPaV replicase polyprotein. The synthetic peptide was then injected into rabbits and the antiserum was collected and stored at -20°C. The anti-RdRp serum was used directly as mentioned above by diluting in IX PBS. The cells were washed as above in IX PBS for a total of 3 washes. The cell pellet was then re-suspended in 1 mL of Alexafluor 488 goat anti-rabbit IgG secondary antibody (Invitrogen) diluted 1:1000 in IX PBS. The cells were incubated on a nutating mixer at room temperature in the dark for 1 hour, then washed 3 times as above and re-suspended in 1 mL of IX PBS. 2.11 Epifluorescent Microscopy 2.11.1 Preparation of BY-2 Cells for Microscopic Examination Biolistically bombarded and immunostained BY-2 cells were prepared for fluorescent microscopy by placing a drop (-150 uL) on a glass microscope slide (Fisher Scientific). A glass coverslip was lowered over the drop, followed by 2-3 Kim Wipe papers. The cells were then compacted by applying intense pressure (50-100 kg of weight) to the coverslip and Kim Wipe papers by leaning on a flat object such as a heating block insert. The Kim Wipes were carefully removed and the coverslip was sealed with clear nail polish. Once sealed, the slide was ready for viewing. 2.11.2 Preparation of BY-2 Protoplasts for Microscopic Examination Protoplasts were prepared for viewing by placing a single drop (-50 uL) on a 59 glass microscope slide. A glass coverslip was then gently lowered onto the cells using forceps. Excess liquid was gently blotted from the edges of the coverslip using a Kim Wipe paper. Coverslips were not sealed as the process of sealing was found to be too rough and often resulted in protoplast shearing. Therefore, microscopy was completed with haste as the protoplasts would survive for only 30-50 minutes before succumbing to dehydration. 2.11.3 Epifluorescent Microscopy Transfected protoplasts were viewed on a Leica DM4500 B epifluorescent microscope (Leica Microsystems, Wetzlar, Germany). For enhanced green fluorescent protein (GFP)-tagged constructs, the GFP fluorophorewas excited with the Leica laser module which accompanies the above microscope. The resulting emission was filtered though a GFP filter cube (Leica cat. #11513899). For monomeric red fluorescent protein (mRFP)- and dsRed-tagged constructs, the red fluorescent fluorophore emission was filtered though a RFP filter cube (Leica cat. # 11513894). To identify transfected fluorescent protoplasts, the sample was first visually scanned under 100X total magnification. However, once a transfected cell was located, brightfield and fluorescent images were captured using a 100X oil immersion lens for a total magnification of 1000X. Images were processed using the Leica Application Suite software (v. 2.3.3) and arranged using Microsoft PowerPoint 2007 (v. 12.0.6504.5000). 60 2.12 Construction of Plant Expression Constructs 2.12.1 Full-length Replicase (pRSP-REP:GFP) pRSP-REP:GFP was constructed by modifying pRSP28A (Meng et al, 2009) and fusing the GFP gene to the 3' terminus of ORF1. Since pRSP28A already contained the replicase ORF driven by the Cauliflower mosaic virus (CaMV) 35S promoter, cloning of ORF1 into another plant expression vector was not necessary. Site-directed mutagenesis (SMD) was used to replace the stop codon of ORF1 with a BamRl restriction site. The SDM reaction contained 50 ng of pRSP28A, IX KOD Hot Start reaction buffer, 0.35 mM of each dNTP, 0.2 uM of each of the primers REPATAG(Bam)-F and REPATAG(Bam)-R (see Table 1), 1 unit of KOD Hot Start DNA Polymerase, and sterile dH20 in a final volume of 50 uL. Thermocycling consisted of 95°C for 3 minutes, then 18 cycles of {95°C for 30 seconds, 50°C for 30 seconds, 70°C for 6 minutes}. Successful amplification was verified by electrophoresing 5 uL of the SDM reaction on a 1% agarose gel, and 10 units of Dpnl (NEB), which digests only methylated and hemi- methylated DNA, was added directly to the remainder of the SDM reaction to destroy the parental pRSP28A template. The Dpnl reaction was allowed to incubate at 37°C for 1 hour. Following Dpnl digestion, 5 uL of the SDM reaction was used to transform 50 uL of chemically competent E. coli JM109 cells. Plasmids from the resulting clones were extracted and screened for the presence of the BamRl restriction site via a BamRl restriction analysis. Plasmid DNA (1.5 ug) testing positive for the BamRl restriction site was then digested with BamHl and subject to de-phosphorylation. To concentrate the plasmid DNA following de-phosphorylation, 0.1 volumes of sodium acetate (3 M, pH 5.2) and 61 Table 1. Primers used in this study. The name, nucleotide position in the GRSPaV genome (Genbank accession number AF026278) if applicable, sequence (5'->3'), and purpose of each primer are shown. Underlined sequence indicates restriction enzyme recognition site. Italicized sequence indicates epitope tag sequence. NA, not applicable: primers anneal to non-GRSPaV sequence. 62 a _o •c +•< c Primer O Sequence Purpose "E LU a. < 13 cn U- 03 GCCTATTTGAAGCGATTGATGGATCCGCCTTAAG 6521-6570 Replace stop codon of ORF1 with BamH\ site rATTTGGCAT ATGCCAATTACTTAAGGCGGATCCATCAATCGCT REPATAG(Bam)-R 6570-6521 Replace stop codon of ORF1 with SomHI site FCAAATAGGC ? 2. <3 Ll_ < < 1- 13 2 U- < "5. >- -C -a < U- 13 Q. \n +-* CO X 4-» v E 3 c O i/i \— u o o E AAAAGGATCCGTGAGCAAGGGCGAGGAG rz o c co c CD t TTTTGGATCCTYATGCATAATCTGGAACGTCATAC <3 ID E U <0 (3 r- < u U U <3 U < 1- 15 r- U < <_) 13 13 GFPAATG(Bam)-R < Amplify GFP with no stop codon, 3' HA tag, and BamH\ site < O 13 < < < U < 1- 13 s 1 < s I I TdllA s < 13 13 u u r- <_> I- u 1- u < 13 u 61-81 13 U Amplify nt 62-455 or 62-683 of ORF1 with 5' myc tag i § <* MTR3 Amplify nt 62-455 of ORF1 with A/col site 1 U 13 u < h- 13 U (3 < < < u 1- <3 U 13 13 u < (3 683-663 < < Amplify nt 62-683 of ORF1 with Ncol site GTGACAAGTCGTCATGTTAGTTCTAGACACAGTC Prepare nt 62-455 for C-terminal GFP fusion via insertion of SDMMT1 433-450 AACGCTAACAT Xba\ site ATGTAAGCGTTGACTGTGTCTAGAACTAACATCA Prepare nt 62-455 for C-terminal GFP fusion via insertion of SDMMT2 450-433 CGACTTGTCAC Xba\ site 63 CTCTTGGCAACTGCAGTAATCTCTAGACACTCATC Prepare nt 62-683 for C-terminal GFP fusion via insertion of E 13 u <3 (3 < SDMMT3 661-681 < Xba\ site CTGTGCCAAAGATGAGTGTCTAGAGATTACTGCA Prepare nt 62-683 for C-terminal GFP fusion via insertion of <3 t= 13 u U SDMMT4 681-661 < < < (3 Xba\ site < JGFPXbaF AAAATCTAGAGTGAGCAAGGGCGAGGA Amplify GFP with no start codon and Xba\ site < GFPXbaR rmTCTAGATTACTTGTACAGCTCGTCC Amplify GFP with Xba\ site .c <. U- 0 ^- lO E a u o 1 AAAACCATGGTTAGTAGGTATGGGTCTGAG Amplify nt 449-683 of ORF1 with Wcol site t ctgfpSTOPSpeR GGAaAGTTTACTTGTACAGCTCGTCCA Amplify GFP with 3' Spe\ site I 6282-6303 AAAAACCATGGGTTGTTTATrTAAGGAAGGG Amplify nt 6282-6548 of ORF1 with A/col site m 0 lO LD fN U3 < (3 < u u u < 1- < < t- U <3 u u < < < 13 |POLR2 i Amplify nt 6282-6548 of ORF1 with Xba\ site ID in (N m UD in GGCCTATTTGAAGCCATTGATTCTAGAGTCCGCA Prepare nt 6282-6548 for C-terminal GFP fusion via deletion SDMREP1 AAAATCACC of stop codon GGTGATTTTTGCGGACTCTAGAATCAATCGCTTC Prepare nt 6282-6548 for C-terminal GFP fusion via deletion SDMREP2 5544-6523 AAATAGGCC of stop codon RFPEcoF z < AAAAGAATTCGCCTCCTCCGAGGACGTCATCAAG Amplify mRFP with no start codon and EcoR\ site 1 1 1 1 1GAGCTCTTAAAGCTCATCATGGGCGCCGGTG RFPSacR Amplify mRFP with Socl site GAGTGGCG ER-For z < AAAAGAATTCTAAGAGGAGTCCACCATGG Amplify ER signal sequence with EcoRI site ER-Rev i 1 1 1 1GAATTCGGCCGAGGATAATGATAGG Amplify ER signal sequence with EcoRI site LCACG 1G 1GA1 1 1CCACAGG 1GCCGCGACCC1 1 1 1 REPMutR173A-F 553-595 R173A mutation of ORF1 fCTGCATG CATGCAGAAAAAGGTTCGCGGCACCTGTGGAAAT REPMutR173A-R 595-553 R173A mutation of ORF1 CACACGTGG CTGCATGATGAGATTCACTACGCGTCAATTAGTGA REPMutW184A-F 589-630 W184A mutation of ORF1 FCTGATC GATCAGATCACTAATTGACGCGTAGTGAATCTCAT REPMutW184A-R 630-589 W184A mutation of ORF1 CATGCAG Sacll-Rev 1044-1021 tTTTTCTAGAGGTTACAACATCATGATGTACTGG Sequential cloning: Fl fragment Sacll-For 978-1000 AAAGGATCTTGAACCTCTAAGCC Sequential cloning: F2 fragment Nrul-Rev 2578-2559 Mil IUAGAICAGGUGCAAIGAI IACCC Sequential cloning: F2 fragment 64 Nrul-For 2525-2545 CI 1 ICACIAI IAGIAAGICGC Sequential cloning: F3 fragment ^lal-Rev 4472-4451 11 1 1 ICIAGAGTACI1IGACAIGGATCACCGG Sequential cloning: F3 fragment Clal-For 4384-4405 TTGTTTCCTCCTGGATACATCG Sequential cloning: F4 fragment i CCATTTACGAACGATAGCCCTGGCACTCATCTTTG RTLMUTl Delete Ncol site from pRTL2 GC i GCCAAAGATGAGTGCCAGGGCTATCGTTCGTAAA RTLMUT2 Delete Ncol site from pRTL2 fGG AGTGATTAATCCGCGTGGCAGGCCTTCTGCATTGT ProMutF 3741-3780 Delete putative cryptic promoter of ORF1 TTGAC GTCAAACAATGCAGAAGGCCTGCCACGCGGATTA ProMutR 3780-3741 Delete putative cryptic promoter of ORF1 ATCACT < < u 1- < 15 u u u U <5 15 15 15 u mRFPXbalF Z < < < < 1- U < < U H Amplify mRFP with no start codon and an Xba\ site mRFPXbaR NA III! IUAGAI IAGGLGCCGGIGGAGI Amplify mRFP with a stop codon and an Xba\ site 3 volumes of 95% ethanol was added to the reaction. Precipitation was at -20°C for 18 hours, after which the DNA was pelleted by centrifugation at 20,000 X g, 4°C, for 30 minutes. The pellet was washed with 200 uL of 70% ethanol and centrifuged as above. The pellet was then air dried at 37°C and re-suspended in 30 uL Tris-HCl (10 mM, pH 8.0). Successful precipitation was verified by running 3 uL on a 1% agarose gel. To prepare the GFP ORF insert, PCR was used to generate a GFP fragment flanked by the BamHl restriction site using the primers GFPAATG(Bam)-F and GFPAATG(Bam)-R (see Table 1). After purifying the PCR product, the GFP fragment was digested with BamHl and the entire reaction electrophoresed on a 1% agarose gel. The appropriate DNA band was gel extracted and ligated into the above precipitated plasmid by mixing approximately 300 ng of vector with 900 ng of the GFP insert, 3 units of T4 DNA ligase, IX reaction buffer, and sterile dFbO in a total reaction volume of 30 uL. The reaction was allowed to incubate at 4°C for 18 hours, after which 1.5 (iL was used to transform 40 uL of electrocompetent E. coli JM109 cells. The resulting colonies were cultured and screened via colony PCR. Recombinant plasmids were extracted from positive clones and the presence of GFP was confirmed by restriction analysis with BamHl. Clones testing positive for the GFP insert via restriction analysis were sequenced on an Applied Biosystems 377 Prism Sequencer (Advanced Analysis Centre, University of Guelph) to confirm there were no mutations and that the C-terminal GFP fusion was intact. Sequence analysis was performed using the DNAStar software package (DNAStar Inc., Madison WI, USA). Finally, 10 ng of sequence-verified pRSP- REP:GFP was used to transform 40 \\L of electrocompetent E. coli SURE2 cells, which are reported to be ideal for the storage and recovery of large plasmids (Yau et al, 2008). 65 2.12.2 Truncation Constructs (pMTRASD:GFP, pMTR.GFP, pMTR:mRFP, pSD:GFP, pPOL:GFP) All truncation constructs were based on the backbone vector pRTL2 (Figure 8), generously supplied by Dr. Baio Ding, Dept. of Biological Sciences, Ohio State University. This plant expression vector contains a multi-cloning site (MCS) flanked upstream by the 35S promoter of CaMV and the translation enhancer element of TEV (Carrington and Freed, 1990), and downstream by a nopaline synthase (NOS) terminator signal. Selection of transformed cells is via an ampicillin resistance gene. To construct the truncations used to locate the region of the replicase responsible for the formation of the puncta, four regions of the replicase were PCR amplified and cloned into pRTL2. For visualization during fluorescence microscopy, GFP tags were inserted immediately downstream of the regions to create a C-terminally tagged fusion construct. 2.12.2.1 Construction of pMTRASD:GFP and pMTR:GFP To clone the MTR regions required to make pMTRASD:GFP and pMTR:GFP, primers were designed (see Table 1) to PCR amplify nucleotides 62-455 (MTF1 and MTR3) and 62-683 (MTF1 and MTR4) respectively from ORF1 of the template vector pRSP28A. These primers also introduced flanking Ncol restriction sites and a 5' myc epitope tag. The vector backbone (pRTL2) was propagated in E. coli DH5a and prepared for cloning by extraction with a Qiagen MiniPrep Kit (Qiagen, Mississauga, Ontario, Canada) following the manufacturer's instructions. The vector and the above PCR 66 Fig. 8. Vector map of pRTL2. All truncation constructs were based on pRTL2. Expression of open reading frames cloned into the multiple cloning site are driven by the Cauliflower mosaic virus 35S promoter (35Sp). Translation is enhanced by a Tobacco etch virus translation enhancer element (TEVe), which allows for cap-independent translation of the expressed open reading frame. Transcription is terminated by a nopaline synthase terminator (NOSt)., Selection is via ampicillin resistance (Amp). Relative positions of useful restriction sites are shown. 67 A/col Sac\ Kpn\ Sma\ BamH\ Xba\ Xho\ EcoRI EcoRV H/ndlll NOSt 35Sp 3 17 bp TEVe 143 bp NOSt 2 44 bp 68 products were then digested with Ncol, and the digested vector was de-phosphorylated. Following de-phosphorylation, the digested PCR products and digested/de- phosphorylated pRTL2 were purified via gel extraction. Ligations were set up by mixing various ratios of insert and vector DNA. The ligation reaction was allowed to proceed at room temperature for 1 hour. Colony PCR screens were performed using appropriate primer sets to detect successful ligation of the MTR inserts into pRTL2 in the correct orientation. Recombinant plasmids were then recovered from positive clones and confirmed to contain the correct MTR inserts via restriction analysis. Plasmids confirmed to contain the proper inserts were then sequenced. These plasmids, containing the GRSPaV replicase amino acids 1-131 and 1- 207, were named pMTRASD and pMTR respectively. To fuse GFP to the C-termini of the MTR inserts, SDM was first used to insert the Xbal recognition sequence at the appropriate position so as to allow for the fusion of GFP in frame with the MTR insert. Table 1 shows the mutagenic primers employed for this purpose (SDMMT1 + SDMMT2 and SDMMT3 + SDMMT4). The resulting plasmids were confirmed to contain the Xbal restriction site by restriction analysis. Positive plasmids, termed pMTRASD-NX and pMTR -NX, were stored at -20°C until needed. To prepare the GFP insert for cloning into pMTRASD-NX and pMTR-NX, the GFP ORF was PCR amplified from pRTL2:GFP using primers GFPXbF and GFPXbR (see Table 1). The GFP insert, along with pMTRASD-NX and pMTR-NX, were then digested with Xbal and the vectors de-phosphorylated. Gel extraction was used to purify the appropriate digestion products, which were then ligated as mentioned previously. Potential pMTRASD:GFP and pMTR:GFP clones were screened via PCR using primers 69 specific to correctly orientated GFP. Plasmids from positive clones were isolated and confirmed to contain the GFP insert through restriction analysis. Positive plasmids were then sequenced to confirm the sequence of pMTRASD:GFP and pMTR:GFP. 2.12.2.2 Construction of pMTR:mRFP For co-expression of the MTR fusion product with GFP tagged constructs and markers, it was necessary to create a version of pMTR:GFP with a different colour fluorescent tag. Monomeric red fluorescent protein (mRFP) was chosen because: (1) It has a similar size to GFP and thus is not any more likely to interfere with peptide folding than GFP; (2) It has an excitation and emission spectrum that does not significantly overlap with that of GFP; and (3) It does not form tetramers like other common red fluorescent proteins such as dsRed, and thus should not influence localization. To create pMTR:mRFP, the mRFP ORF was PCR amplified from the plasmid pTJ*E2~:mRFP (Rebelo et al, 2008) using the primers mRFPXbalF and mRFPXbaR (see Table 1) which incorporate flanking Xbal sites and delete the start codon of mRFP. The PCR product was purified and digested, along with pMTR:GFP, with Xbal. The resulting digestion products were electrophoresed on a 1% agarose gel and the mRFP insert and pMTR:GFP vector (from which the GFP insert was released during digestion) were gel extracted. The digested vector was de-phosphorylated and various ligation ratios were performed. 5 uL of each ligation reaction was then used to transform 40 uL of chemically competent E. coli DH5a. The resulting colonies were screened for the correctly orientated mRFP insert, and plasmid DNA was extracted from positive clones. Positive plasmids were then confirmed to contain the mRFP insert via restriction analysis with Xbal. 70 2.12.2.3 Construction of pSD:GFP To construct pSD:GFP, PCR was used to amplify a 3' region of pMTR.GFP from nt 456 to the stop codon of the GFP tag with primers AmphNcoF and ctgfpSTOPSpeR (see Table 1). These primers incorporated an Ncol and Spel restriction site at the 5' and 3' ends of the amplicon respectively. The PCR product was purified using a PCR purification kit and digested with Ncol and Spel. Simultaneously, pRTL2 was digested with Ncol and Xbal, with Xbal generating identical cuts to that of Spel. The digestion products were electrophoresed on a 1% agarose gel and extracted via a gel extraction kit. The Ncol/Spel digested insert and Ncol/Xbal digested vector were then ligated and transformed into E. coli DH5a. The resulting colonies were screened for the presence of the insert and the plasmids from positive clones were isolated and sequenced. 2.12.2.4 Construction of pPOL-.GFP To create pPOL:GFP, PCR was used in conjunction with the primers POLF2 and POLR2 (see Table 1) to amplify nt 6282-6549 (the RdRp domain) of the GRSPaV replicase. These primers incorporated an Ncol and an Xbal restriction site at the 5' and 3' terminus respectively. The PCR product, along with pRTL2, was digested directly with Ncol and Xbal, and the digestion products were purified via gel extraction. The RdRp insert was then ligated into pRTL2 and transformed into E. coli DH5a. The resulting colonies were screened by colony PCR and plasmid DNA was extracted from a positive clone. This intermediate plasmid was termed pPOL-NX, but contained a stop codon for potential future experiments, so further modification was necessary prior to inserting the GFP tag. The mutagenic primers SDMREP1 and SDMREP2 (see Table 1) were used in 71 an SDM reaction for the purpose of deleting the stop codon of pPOL-NX and shifting the Xbal restriction site in-frame with the RdRp nucleotide sequence. The resulting plasmid, termed pPOL-NXmut, was transformed into E. coli DH5a. Since there was no way to screen for this particular mutation a single colony was simply cultured and the plasmid DNA extracted for further cloning. To prepare the GFP insert, the GFP ORF was amplified from pRTL2:GFP using the primers GFPXbF and GFPXbR (see Table 1). These primers introduce Xbal restriction sites at the 5' and 3' termini of the GFP insert. After the PCR product was purified via a PCR purification kit, it was digested, along with pPOL-NXmut, with Xbal. The insert and vector digestion products were purified via gel extraction and pPOL- NXmut was de-phosphorylated to prevent re-ligation. The GFP insert was then ligated into pPOL-NXmut and transformed into E. coli DH5a. The resulting colonies were screened for the presence of the correctly orientated GFP insert via colony PCR. Plasmid DNA was extracted from positive clones and analyzed by digestion with Ncol and Xbal. The nucleotide sequence of positive plasmids was then confirmed by sequencing. 2.12.3 Construction of pMTR:GFP (R173A), pMTR-.GFP (W184A), and pMTR:GFP (R173A,W184A) To make the single mutant constructs pMTR:GFP (R173A) and pMTR:GFP (W184A), SDM was used in conjunction with the appropriate mutagenic primers listed in Table 1 (REPMutR173A-F + REPMutR173A-R and REPMutW184A-F + REPMutW184A-R). pMTR:GFP (-20 ng) was used as the parental template, and thermocycling consisted of 95°C for 30 seconds followed by 12 cycles of {95°C for 30 72 seconds, 55°C for 1 minute, 70°C for 5 minutes}. The parental template was then digested with Dpnl at 37°C for 30 hours and 4 uL was used to transform 50 uL of chemically competent E. coli DH5a. Plasmid DNA was extracted from the resulting colonies and sequenced to confirm the appropriate R173A and W184A mutations. To create the double mutant construct pMTR:GFP (R173A,W184A), a second round of SDM was applied to the parental template pMTR:GFP (R173A) using the primers REPMutW184A-F and REPMutWl 84A-R (see Table 1), which introduced the W184A mutation. Identical SDM parameters were used as mentioned above. The resulting double mutant construct was again sequenced to confirm the presence of both mutations. 2.12.4 Construction of pER-mRFP Since the majority of the constructs used in these experiments were tagged with GFP, and a red fluorescent protein-based ER marker was not available, it was necessary to construct one by modifying pER-GFP (Haseloff et al, 1997). This plasmid expresses a peptide containing an ER signal sequence and an "HDEL" ER retention motif fused to GFP, which effectively accumulates in the ER lumen. Selection is via kanamycin resistance. To create a red fluorescent version of this marker, the GFP tag was simply replaced with mRFP. To accomplish this, the primers RFPEcoF and RFPSacR (see Table 1) were used to amplify the mRFP ORF from pRTL2:mRFP (kindly provided by Dr. L. Torrance, Scotish Crop Research Insititute) while introducing EcoRl and Sacl restriction sites at the 5' and 3' terminus respectively. After purification with a PCR purification kit, the PCR product, along with pER:GFP, was digested with EcoRI and Sacl. The digestion products were purified via gel extraction, ligated, and transformed into chemically 73 competent E. coli DH5a. The resulting colonies were screened for the presence of mRFP by colony PCR, and plasmid DNA from positive clones was extracted. Due to the presence of an EcoKL restriction site within the ER lumen protein ORE, part of the ER signal peptide was accidently excised during the above cloning procedure. It was therefore necessary to insert the ER signal sequence back into the above construct, termed pmRFPAER. PCR was used with primers ER-For and ER-Rev (see Table 1) to amplify the missing sequence (using pER-GFP as a template) flanked by EcoRl restriction sites. Since the amplicon was less than 200 bp, ten identical PCR reactions were performed and concentrated into one sample by centrifuging all reactions through a single PCR purification column. The entire concentrated sample (-25 uL), along with 1 ug of pmRFPAER, was then digested with 10 units of EcoRl in a total reaction volume of 40 uL. The digested products were gel extracted and pmRFPAER was de- phosphorylated. The ER signal was then ligatedjinto pmRFPAER and transformed into E. coli DH5a. The resulting colonies were screened with colony PCR using primers to detect the ER signal in the correct orientation. Sequencing was performed to confirm all components of pER-mRFP were correct. To confirm that the above cloning procedure did not change the ability of the expression product to localize to the ER lumen, pER- mRFP was co-expressed in tobacco BY-2 protoplasts with pER-GFP. The results clearly showed perfect co-localization of the two markers (data not shown), demonstrating the proper functionality of pER-mRFP. 74 CHAPTER 3: RESULTS 3.1 The GRSPaV replicase forms punctate structures in the cytoplasm of tobacco BY-2 cells and protoplasts Since the natural hosts of GRSPaV are grapevines (Vitis sp.), it would have been best to perform all of the following experiments in a grapevine cell system. However, attempts to develop such a system failed (see Appendix II), and so Nicotiana tabacum cv. Bright Yellow (BY-2) cells were used instead. pRSP28A, a GRSPaV cDNA clone previously constructed in our lab (Figure 9) (Meng et al, 2009), was transfected into tobacco BY-2 cells via biolistic bombardment. The BY-2 cells were then fixed and immunostained with an anti-RdRp polyclonal antibody to detect the RdRp domain of the replicase. The transfection efficiency using this method was approximately 1%, with at least 20 healthy cells being examined per experiment. Punctate structures (referred hereafter as "viral puncta" for simplicity) were observed in the cytoplasm of transfected cells at 24 hours post transfection (hpt) (Figure 10). The puncta appeared to be approximately 0.5 urn in diameter, with each cell containing approximately 200 puncta. These viral puncta were thought to be the replicase complexes residing in membranous vesicles derived from a host membrane. Since immunostaining involved the process of fixing the cells, there was the possibility of damaging/altering the cells and creating artefacts. Additionally, immunostaining allows one to examine only fixed cells, so any behaviour of the replicase due to living biological processes could not be observed. To facilitate future studies of the viral puncta and negate the need for immunostaining, GFP was fused to the C- terminus of the RdRp domain in pRSP28A, yielding the construct pRSP-REP:GFP 75 Fig. 9. Full-length and truncated replicase constructs used in this study. All constructs are expressed from the CaMV 35S promoter (35Sp). Truncation constructs also contain a Tobacco etch virus translational enhancer element (TEVe) downstream of the promoter. The sequence and relative position of the anti-RdRp epitope is shown for pRSP28A. Restriction sites and amino acids positions are also shown. For potential future experiments, a human myc epitope tag was fused to the N-terminus of the constructs pMTR:GFP, pMTR:mRFP, and pMTRASD:GFP, and a hemagglutinin (HA) epitope tag was fused to the C-terminus of the GFP tag of pRSP-REP:GFP. The shaded area represents the sub-domain hypothesized to contain a membrane association sequence. UTR, un-translated region; MTR, methyltransferase; O-Pro, ovarian tumour protease; P-Pro, papain-like cysteine protease; HEL, helicase; POL, RNA-dependent- RNA-polymerase; GFP, green fluorescent protein; mRFP, monomeric red fluorescent protein; pi, Triple Gene Block protein 1; p2, Triple Gene Block protein 2; p3, Triple Gene Block protein 3; CP, capsid protein. For simplicity, the highly variable region and AlkB-like domains of the replicase were not labelled. 76 PRSP28A 'GKGLDELNEWVRTRG' CEIy35 Sp >5'UTi R MTR O-Pro P-Pro HEL POL || Pi ifpTI pp p3 3'UTR pRSP-REP:GFP •• SamHI BamHI |35Sp >5'UTR [ MTR HEL POL GFP HA 1 % 2161 pMTR:GFP A/col *• • Xba\ Xba\ 35 Sp TEVe | myc | MTR "^-- — "1 GFP 207 pMTR:mRFP . A/col Xba\ Xba\ | 35Sp^|"TEVe j myc | MTR [-— — — — { mRFP * 1 207 • pMTRASD:GFP . Nco\ Xba\ Xba\ \ 35Sp^T^ve | myc | MTR \------\ GFP ^ 1 132 pSD:GFP A/col Xba\ Xba\ISpe\ | 35 Sp ^| TEVe Y> •-0--—QEL3 132 207 pPOL:GFP A/col Xba\ Xba\ { 35 Sp y TEVe |— POL GFP pRTL2:GFP 2094 2183 35 Sp TEVe GFP | 77 Fig. 10. Epifluorescent images of tobacco BY-2 cells and protoplasts expressing the full-length GRSPaV replicase. The native (i.e. not GFP tagged) replicase was expressed in BY-2 cells biolistically bombarded with pRSP28A and detected via immunostaining at 24 hours post transfection (hpt) (a and b). Immunostaining was performed with a primary anti-RdRp polyclonal antibody produced in rabbits injected with a synthetic peptide from within the RdRp domain of the replicase polyprotein, and a secondary polyclonal goat anti-rabbit antibody conjugated to the Allexafluor 488 fluorophore (Invitrogen). Alternatively, a GFP tagged version of the replicase was electroporated into BY-2 protoplasts and observed directly in live cells at 18 hpt (c and d). Bar = 20 urn 78 Brightfield Fluorescence pRSP28A C MTR HEL RdRp | pRSP-REP:GFP | MTR HEL RdRp | GFP | 79 (Figure 9). The expression product of pRSP-REP:GFP, a GFP tagged version of the full- length replicase polyprotein, could then be observed directly in live cells. Thus, pRSP- REP:GFP was transfected into tobacco BY-2 protoplasts via electroporation. The transfection efficiency using this method was approximately 1% for pRSP-REP:GFP, and at least 20 healthy cells were examined per experiment. Unless otherwise stated, all observations were performed at 18 hpt, as this time point was found to produce protoplasts of healthy quality (as judged by the visual appearance of the protoplasts) as well as high levels of expression (as judged by the expression of un-fused GFP) (data not shown). Upon expression of pRSP-REP:GFP, viral puncta were observed in the cytoplasm of transfected protoplasts (Figure 10), suggesting that the GFP tag is not interfering with the behaviour of the replicase with respect to the formation of the viral puncta. A background diffusion pattern, similar to that of GFP alone, can also be seen in transfected protoplasts. The number of viral puncta observed varied from five to several dozen with a majority of the cells having 25-50 per cell. The number of viral puncta per cell did not seem to vary with the incubation time post transfection. The size of the viral puncta was approximately 1 urn in diameter, though some larger punca (~5 um in diameter) were observed in approximately 20% of the cells examined. Cells that harboured the larger viral puncta usually had approximately 8 small puncta for every 1 large punctate structure. It should be noted that cells containing high numbers (several dozen) of viral puncta tended to produce many punctate structures that were smaller than 1 um in diameter. The viral puncta produced by pRSP-REP:GFP were visibly mobile within the cytoplasm, which was possible to observe directly since the protoplasts were still living during microscopic examination. 80 3.2 A sub-domain within the MTR domain is important for the formation of the viral puncta To determine which part of the replicase polyprotein is important for the formation of the puncta, in silico analysis of the GRSPaV replicase sequence was first performed to predict possible membrane association sequences. Though possibly separate phenomena, membrane association and the formation of the viral puncta may be interdependent, if the viral puncta do indeed represent replicase-derived vesicles in which GRSPaV replication would take place. It was therefore logical to speculate that any putative membrane association signal within the replicase may be responsible, or at least required, for the formation of the viral puncta observed in the above experiments. In silico analysis of the GRSPaV replicase amino acid sequence yielded no predicted transmembrane domains or significant hydrophobic regions (data not shown), suggesting membrane association functions of the replicase, if any, would be controlled by a method other than a transmembrane signal. For example, research on the model alphavirus SFV and more recently on BMV has shown that replicase proteins lacking transmembrane domains may utilise an amphipathic helix to associate with the target membrane. Such monotopic signals are typified by possessing alternating polar and non-polar amino acid residues, a critical arginine residue, and the propensity to form an amphipathic alpha- helix. Sequence analysis of the GRSPaV replicase using DNAStar software predicted the largest amphipathic region at aa 157-173 (Figure 11). The next largest predicted amphipathic regions were at aa 69-79 and aa 49-59. However, since most monotopic signals discovered to date contain approximately 18-20 (or more) amino acids 81 Fig. 11. Predicted amphipathic regions of the GRSPaV methyltransferase domain. Predicted locations of amphipathic alpha-helices (shaded boxes) are indicative of putative membrane binding domains. Numbers are in amino acids (aa). Circled region shows location of the largest amphipathic alpha-helix (aa 157-173). This region was hypothesized to contain a membrane-associaiion signal important for the formation of the viral puncta observed upon expression of the GRSPaV replicase in BY-2 cells. 82 a a I I I 1 I I I I I I I 1 I I T" 32 34 36 38 40 42 44 46 48 50 52 54 56 58 60 aa —i 1 1 r l i i i i i i i i i i 92 94 96 98 100 102 104 106 108 110 112 114 116 118 120 aa —i 1 1 1 1 1 1 1 1 1 1 1 1 1 1— 122 124 126 128 130 132 134 136 138 140 142 144 148 148 150 aa —i r T I I 1 I 1 I I I I 1 182 184 186 188 190 192 194 196 193 200 202 204 206 208 210 ftft —i 1 1 1 1 1 1 1 1 1 1 1 1 1 r~ 212 214 216 218 220 222 224 226 228 230 232 234 236 238 240 aa —i 1 1 1 1 1 1 1 1 1 1 1 1 1 1— 242 244 248 248 250 252 254 256 258 260 262 264 266 268 270 aa —i 1 1 1 1 1 1 1 1 1 1 1 1 1 1— 272 274 276 278 280 282 284 286 288 290 292 294 296 298 300 83 (Sharadadevi et al, 2005; Spuul et al, 2007; Liu et al, 2009; Sivakamasundari and Nagaraj, 2009), the predicted region at aa 157-173 was chosen as the most likely location of a monotopic signal, if such a signal indeed exists within the GRSPaV replicase. Since a monotopic signal was reported for SFV at the time, an amino acid sequence alignment was performed using the GRSPaV and SFV (nsPl) replicase sequences in order to identify any monotopic tell-tale residues in the GRSPaV replicase. ClustalW alignment using the Biology Workbench software (available at http://workbench.sdsc.edu) identified a series of residues at aa 165-207 within the MTR domain of the GRSPaV replicase that could potentially form an amphipathic a-helix similar to that of SFV (Figure 12). The amino acid residues included alternate polar and non-polar amino acids as well as potentially critical arginine (aa 173, R173) and tryptophan (aa 184, W184) residues, suggesting a monotopic membrane association signal may reside within aa 165-187. This region overlaps the aforementioned predicted amphipathic region (aa 157-173) (Figures 11 and 12). Taken together, these results suggest that if a monotopic signal is indeed present within the GRSPaV replicase, it is most likely present within the MTR domain, henceforth referred to as the "sub-domain". In order to encompass the previously mentioned amphipathic region (aa 157-173) as well as the region which aligns with the SFV monotopic sequence (aa 165-187), this "sub-domain" region refers to amino acids 132-207 (Figure 12). To test the hypothesis that the sub-domain is needed for the formation of the viral puncta, the construct pMTR:GFP (Figure 9) was created and expressed in tobacco BY-2 protoplasts. pMTR:GFP expresses the GRSPaV replicase amino acid sequence 1-207 fused at the C-terminus to GFP. Since the expression product of this construct contains 84 Fig. 12. Nucleotide alignment of the GRSPaV methyltransferase domain with the Semliki Forest virus nsPl replication protein and its relative position to a predicted amphipathic region within the "sub-domain". Alignment of the GRSPaV and Semliki Forest virus methyltransferase domains show a GRSPaV region (ORF1 aa 165-187) aligning with the monotopic signal sequence of SFV (SFV aa position 245-265, in red). Arrows indicate the critical arginine and tryptophan residues in SFV monotopic signal and counterpart putative critical residues in GRSPaV sequence. Bold amino acids are important hydrophobic residues required to form an amphipathic a-helix in SFV and putative counterpart residues in GRSPaV sequence. Position of the alignment relative to the GRSPaV predicted amphipathic region is also shown. Collectively, these regions make up the "sub-domain" (ORF1 aa 132-207, shaded in grey), a region of the GRSPaV MTR domain. Units are in GRSPaV amino acids. MTR, methyltransferase; O-Pro, ovarian tumour protease; P-Pro, papain-like cysteine protease; HEL, helicase; RdRp, RNA-dependent-RNA-polymerase. Alignment was performed using the ClustalW program of Biology Workbench (Bioinformatics and Computational Biology, University of California, San Diego) available at http://workbench.sdsc.edu/. 85 E-iCk CuEnV CM < •A ^s « J \ oI - Q. CM OP V cc JH \ "D KCO^ •• • s. • CC PH&<<- •• • >>H \ *• J<3 0 • • • ^1 COK •• PLI H • HI c •• CD • X COK CO • E Is- * PtJ J c •• Jfc O* • • • • • o "CO • J J •• ^ !^S > • Q. LL •• i CO • CO CM CO T— ' ^_! • • LL > DC nJ o ••ft >co > fcpc; CO coO Q_ CO CC (D 86 the sub-domain, it was expected to allow formation of the viral puncta. Indeed, an expression pattern similar to that of pRSP-REP:GFP was observed upon expression of pMTR:GFP in tobacco BY-2 protoplasts (Figure 13, c-d). The viral puncta appear similar in size, number, and distribution to those of the GFP-tagged full-length replicase. The transfection efficiency was approximately 4-fold higher than with pRSP-REP:GFP, which could be because pMTR:GFP is significantly smaller than pRSP-REP:GFP, which directly, and positively, would affect transformation efficiency on a per microgram basis (Yau et al, 2008). Also, pMTR:GFP contains a TEV translational enhancer element so that expression is consequently increased, rendering transfected protoplasts more readily observable compared to those transfected with pRSP-REP:GFP. These factors together explain the significant increase in observable transfected protoplasts noted with all truncation constructs compared to pRSP-REP:GFP. To determine if the sub-domain (aa 132-207) specifically is responsible for the formation of the viral puncta observed during pMTR:GFP expression, the sub-domain was removed from pMTR:GFP and mobilized into pRTL2 to create the constructs pMTRASD:GFP and pSD:GFP respectively (Figure 9). pMTRASD:GFP expresses the GRSPaV MTR domain with the sub-domain deleted (aa 1-131) fused at the C-terminus to GFP, and pSD:GFP expresses only the sub-domain (aa 132-207) fused at the C-terminus to GFP. It was expected, according to the original hypothesis, that the product of pMTRASD:GFP would not be able to form viral puncta, as this construct lacks the sub- domain. Conversely, it was expected that, during pSD:GFP expression, the sub-domain alone would alter the pattern of GFP resulting in the formation of the viral puncta. As predicted, protoplasts expressing pMTRASD:GFP (Figure 13, e-f) showed a general 87 Fig. 13. Epifluorescent images of tobacco BY-2 protoplasts expressing regions of the methyltransferase domain. Nicotiana tabacum cv. Bright Yellow (BY)-2 protoplasts (18 hpt) expressing GFP alone (a and b), the methyltransferase (MTR) domain (aa 1-207) of the GRSPaV replicase (c and d), the MTR domain with the sub-domain deleted (aa 1 - 131) (e and f), and the sub-domain alone (aa 132-207) (g and h). Expression of the sub- domain with the rest of the MTR domain or alone resulted in the formation of punctate structures in the cytoplasm (d and h respectively), while expression of the MTR domain without the sub-domain resulted in a uniform diffusion pattern (f) similar to GFP alone (b). Bar = 20 urn. 88 Brightfield Fluorescence pRTL2:GFP • | GFP | pMTR:GFP | MTR h"-| GFP | pMTRASD:GFP 1 GFP | pSD:GFP • l-Gm 89 fluorescent diffusion pattern similar to GFP alone (Figure 13, a-b), while viral puncta were observed in protoplasts expressing pSD:GFP (Figure 13, g-h). The viral puncta produced by pSD:GFP expression were similar in size, number, and distribution to those produced by pRSP-REP:GFP and pMTR:GFP. These results suggested that the sub- domain within the MTR of the GRSPaV replicase is indeed responsible for the formation of the viral puncta. 3.3 The GRSPaV replicase polyprotein residues R173 and W184 are not critical for the formation of the viral puncta The formation of the viral puncta may depend on the association of the replicase with a membrane, presumably via an amphipathic monotopic signal. Studies with SFV have demonstrated that an arginine and a tryptophan residue are both required for the monotopic signal to bind to a membrane. Alignment of the GRSPaV and SFV replicase amino acid sequences was used to locate the counterpart potentially critical residues (R173 and W184) within the GRSPaV putative monotopic signal (Figure 12), which were chosen as targets whose alteration would most likely result in a loss-of-function if the region in question is indeed a monotopic signal. Therefore, to test the hypothesis that a monotopic membrane association signal was present at aa 165-187 within the previously mentioned "sub-domain" (aa 132-207) of the GRSPaV replicase, site-directed mutagenesis (SDM) was used to replace these two potentially critical residues. As the formation of the viral puncta was hypothesized to require association with a membrane, the loss of the membrane binding mechanism was predicted to manifest as a failure of the viral puncta to form. SDM of pMTR:GFP was used to replace the putatively critical 90 arginine (R173A) and tryptophan residues (W184A), separately and together, with alanine. The resulting three mutant constructs, which expressed products with a single arginine mutation, a single tryptophan mutation, or both mutations, were termed pMTR:GFP (R173A), pMTR:GFP (W184A), and pMTR:GFP (R173A,W184A) respectively (Figure 14). These constructs were sequenced to confirm the mutations. Expression of these constructs in tobacco BY-2 protoplasts resulted in the formation of viral puncta in each case (Figure 15). The viral puncta produced by the mutant constructs in each case appeared similar in size, distribution, and number compared to the viral puncta produced by the wildtype MTR domain. This data suggests that the R173 and W184 residues are not critical with respect to the formation of the viral puncta. 3.4 The viral puncta appear to localize to actin microfilaments and adjacent to the ER lumen To deduce the sub-cellular location of the viral puncta, pMTR:GFP was co- expressed with plasmids expressing organelle markers fused to red fluorescent proteins. Unless otherwise stated, all organelle markers were generously provided by Dr. E. Blancaflor, The Samuel Roberts Noble Foundation. Organelle markers used included those for the ER (pER:mRFP), Golgi stacks (pRatST-dsRed) (Wee et al, 1998), peroxisomes (pRTL2:MFP:RFP, mRFP tag, generously provided by Dr. Robert Mullen, University of Guelph), and actin microfilaments (pmTalin-dsRed) (Kost et al, 1998; Liu et al, 2005). pER-mRFP was created by altering pER:GFP (see section 2.12.4), and tested by co-expressing with pER:GFP prior to use. The ER was chosen as the most likely target of the GRSPaV replicase simply because the ER is utilized by many 91 Fig. 14. Mutant truncation constructs used in this study. Schematic representation of GRSPaV methyltransferase (MTR) mutant constructs used to determine the significance of the amino acids R173 and W184 with regards to the formation of the viral puncta. Mutant constructs are expressed from the Cauliflower mosaic virus 35S promoter (35 Sp). A Tobacco etch virus translational enhancer (TEVe) is downstream of the promoter. A human myc epitope tag was fused to the N-terminus of each construct for possible future experiments. Enhanced green fluorescent protein (GFP) was fused to the C- terminus of each mutant polypeptide for direct visualization. Point mutations within each mutant construct are summarized in a table. R, arginine; W, tryptophan. 92 pMTR:Amph:GFP(R173A) R173A | 35Sp^| TEVe I myc | MTR h""4" GFP pMTR:Amph:GFP (W184A) W184A L | 35 Spi yj TEVe I myc | MTR H —^ GFP pMTR:Amph:GFP (R173A,W184A) R173A W184A | 35Sp^>| TEVe | myc | MTR ^----^ GFP Amino acid: 173 184 Wildtype R w R173A A w W184A R A R173A,W184A A A 93 Fig. 15. Epifluorescent images of tobacco BY-2 protoplasts expressing mutant truncation constructs. Nicotiana tabacum cv. Bright Yellow (BY)-2 protoplasts (18 hpt) expressing the methyltransferase domain (aa 1-207 of the GRSPaV replicase) fused to GFP containing the alanine substitutions R173A (a and b), W184A (c and d), or R173A and W184A (e and f). Each mutation either individually or simultaneously did not abolish the formation of the viral puncta, suggesting that these residues are not involved in punctal formation. Bar = 20 urn. 94 Brightfield Fluorescence R173A | MTR h"H GFP I W184A | MTR h"| GFP | R173A,W184A | MTR h"-| GFP | 95 alphavirus-like viruses (Chen and Ahlquist, 2000; den Boon et al, 2001; Wei and Wang, 2008). The Golgi stacks and peroxisomes were examined as possible targets due to the similarity of the viral puncta, in size and distribution, with these organelles. Lastly, the actin microfilaments were examined as possible targets of the GRSPaV replicase because the replicase of TMV, which is also a member of the alphavirus-like superfamily, associates with actin microfilaments (Wright et al, 2007; Guenoune-Gelbart et al, 2008; Sambade et al, 2008). The viral puncta produced by expression of pMTR:GFP were observed in close proximity to the ER network, as judged by co-expression of pMTR:GFP with an ER lumen marker (i.e. pER-mRFP) (Figure 16). This data suggested that the replicase may be associating with Golgi stacks, which are also in close proximity to the ER network and appear similar in size to the viral puncta. However, the viral puncta were found not to co- localize with the Golgi stacks (Figure 17 a-c). Similarly, the peroxisomes were also found not to co-localize with the viral puncta (Figure 17 d-f). These findings are not surprising as there have been no reports of a plant virus replicating solely on the Golgi stacks, and only some members of Tombusviridae utilize peroxisomes for replication. When pMTR:GFP was co-expressed with pmTalin-dsRed, an actin marker, the viral puncta were observed to align along the actin microfilaments (Figure 18). The viral puncta were mobile, appearing to move through the cytoplasm; however, it was not possible to conclude that the puncta were trafficing along the actin filaments using epifluorescence microscopy, as the actin filaments and viral puncta could not be observed simultaneously. Efforts to obtain high resolution images using confocal microscopy were not successful (data not shown). Regardless, the viral puncta are apparently associating 96 Fig. 16. Epifluorescent images of tobacco BY-2 protoplasts co-expressing the methyltransferase domain with an ER marker protein. Nicotiana tabacum cv. Bright Yellow (BY)-2 protoplasts (18 hpt) co-expressing the methyltransferase (MTR) domain of the GRSPaV replicase with an ER lumen marker. The MTR domain localizes adjacent to, but not within or in contact with, the ER. Panels a-d represent a single protoplast while panels e-h represent another protoplast from a duplicate experiment. To better show localization of the puncta adjacent to the ER, a region of panel c is magnified in d, and a region of panel g is magnified in h. Images were processed using ImageJ v. 1.40 (National Institutes of Health, USA). Bar = 20 urn. 97 pMTR:GFP GFP -|- ER-mRFP MTR I ' ' Green Fluor. Red Fluor. Merged 98 Fig. 17. Epifluorescent images of tobacco BY-2 protoplasts co-expressing the methyltransferase domain with Golgi and peroxisome marker proteins. Nicotiana tabacum cv. Bright Yellow (BY)-2 protoplasts (18 hpt) co-expressing the methyltransferase (MTR) domain of the GRSPaV replicase with a Golgi marker (a-c), or a peroxisome marker (d-f). The MTR domain does not co-localize with the Golgi stacks or peroxisomes. Images were processed using Image J v. 1.40 (National Institutes of Health, USA). Bar = 20 urn. 99 pMTR:GFP ^ Golgi-dsRed | MTR H — Ml GFP | Green Fluor. Red Fluor. Merged pMTR:GFP «^ Peroxisomes-mRFP | MTR HMMI GFP | Green Fluor. Red Fluor. Merged 100 Fig. 18. Epifluorescent images of tobacco BY-2 protoplasts co-expressing the methyltransferase domain with an actin marker protein. Nicotiana tabacum cv. Bright Yellow (BY)-2 protoplasts (18 hpt) co-expressing the methyltransferase domain of the GRSPaV replicase with an actin marker. The MTR domain forms punctate structures which align along actin microfilaments. Panels a-d represent a single protoplast while panels e-h represent another protoplast from a duplicate experiment. To better show actin localization, a region of panel c is magnified in d, and a region of panel g is magnified in h. Images were processed using ImageJ v. 1.40 (National Institutes of Health, USA). Bar = 20 fim. 101 pMTR:GFP J_ Actin-dsRed | MTR h"| GFP~*n Green Fluor. Red Fluor. Merged 102 with the actin microfilaments as indicated by the physical location of almost every punctate structure next to, or on top of, a strand of actin. 3.5 The RdRp domain of the GRSPaV replicase forms aggregate structures in the cytoplasm when expressed alone Since the RdRp domain was the region of the replicase targeted for fluorescence immunostaining following expression of pRSP28A, as well as the region to which GFP was fused in pRSP-REP:GFP, a truncation construct was created to express only the RdRp domain (aa 2094-2183) fused at it's C-terminus to GFP (pPOL:GFP) (Figure 9). Expression of pPOL:GFP in BY-2 protoplasts yielded aggregate structures in the cytoplasm (Figure 19 A). These structures were larger (3-5 urn in diameter) than the previously mentioned viral puncta (1 urn in diameter), and appeared to predominantly surround the nucleus of the cells, though aggregates were also observed elsewhere in the cytoplasm and at the cell periphery. The RdRp aggregate structures were also not visibly mobile. A time-lapse video over an eight hour period of an individual protoplast transfected with pPOL:GFP showed that the aggregates appeared at approximately 11 hpt and reached full size by 13 hpt (data not shown). The aggregates then remained stationary until the protoplast died. When pPOL:GFP was co-expressed with pMTR:mRFP, the RdRp expression pattern changed from larger aggregates to.smaller, cytoplasmic punctate structures which were visibly mobile (Figure 19 B, panel b). Additionally, the viral puncta produced by both constructs co-localized (Figure 19 B, panel d), suggesting that the RdRp domain was localizing to the same sub-cellular location as the MTR domain. 103 Fig. 19. Epifluorescent images of tobacco BY-2 protoplasts expressing the GRSPaV RNA-dependent-RNA-polymerase domain alone or with the GRSPaV methyltransferase domain. Nicotiana tabacum cv. Bright Yellow (BY)-2 protoplasts (18 hpt) expressing the RNA-dependent-RNA-polymerase domain (RdRp, aa 2094-2183) of the GRSPaV replicase fused to GFP either alone (A) or co-expressed with the methyltransferase domain (MTR, aa 1-207) of the replicase fused to mRFP (B). Expressed alone, the RdRp domain forms aggregates in the cytoplasm (Part A, panel b). When co-expressed with the MTR domain (pMTR:mRFP), the RdRp forms smaller puncta that co-localize with the viral puncta (Part B, panels a-d). Bar = 20 urn. 104 (A) Brightfield Fluorescence pPOLGFP — —— —^ RdRp | GFP | (B) Brightfield Green Fluor. pPOLGFP —— —— ^ RdRp | GFP | Red Fluor. Merged pMTRimRFP | MTR h"| mRFP | 105 CHAPTER 4: DISCUSSION AND CONCLUSIONS 4.1 The GRSPaV replicase forms punctate structures when expressed in tobacco BY- 2 cells and protoplasts All (+) RNA viruses examined in detail to date replicate in association with a host membrane. Explanations for this necessity range from a method to increase viral replication efficiencies to evasion of host antiviral responses. The specific membrane with which a (+)RNA virus associates varies depending on the viral species and is determined by the viral replicase protein(s). GRSPaV is a plant virus belonging to the alphavirus-like superfamily of (+)RNA viruses, however the sub-cellular location of viral replication has not been determined. In an attempt to address this matter, we transiently expressed the GRSPaV replicase polyprotein with and without a GFP tag in Nicotiana tabacum cv. Bright Yellow (BY-2) cells and protoplasts. Small punctate structures, referred to here as "viral puncta", were observed in the cytoplasm of transfected cells. It was hypothesized that these structures represented replicase complexes within membranous vesicles derived from a host membrane (i.e. progenitors of what would become "virus factories" in the context of a viral infection). Indeed, transient expression of alphavirus-like replicases alone is often enough to cause membrane restructuring leading to the formation of viral replication vesicles (den Boon et al, 2001; Prod'homme et al, 2001; Spuul et al, 2007; Nagy and Pogany, 2008). The presence of a GFP tag at the C-terminus of the replicase did not affect the appearance of the viral puncta, however significantly fewer puncta were observed in cells transfected with the GFP-tagged replicase. The native (i.e. non-GFP tagged) version of the replicase was detected via immunostaining using a polyclonal antibody raised against 106 the RdRp domain of the GRSPaV replicase followed by fluorescent microscopy. The increased number of punctate structures observed after expression of the native replicase compared to the GFP tagged replicase (-200 puncta vs. -50 puncta respectively) probably reflects a technical difference in the way that BY-2 cells are prepared for microscopy compared to BY-2 protoplasts. BY-2 cells, which are rigid due to the presence of a cell wall, must be compressed on the slide prior to microscopic examination (see section 2.11.1). As a consequence, this step may have artificially increased the number of punctate structures visible during microscopy. After compression, the lower, central, and upper sides of the BY-2 cells are all forced into the same visual plane. Protoplasts, on the other hand, do not need to be compressed as they lack cell walls and thus maintain a three dimensional shape during microscopy. Thus, it was not possible to observe all of the punctate structures in a single focal plane, or capture all of the puncta in a single image. Though protoplast immunostaining was unsuccessfully attempted (see Appendix III), expression and immunostaining of the native replicase in BY-2 protoplasts instead of BY-2 cells would be prudent and likely yield a similar number of puncta per fluorescent image as the GFP tagged version reported here. 4.2 Amino acid residues 132-207 of the MTR domain of the GRSPaV replicase are responsible for the formation of viral puncta If the viral puncta reported above represent the GRSPaV replicase associating with and re-arranging a host membrane, then the replicase must contain a membrane association signal itself, or else interact with a host membrane-associating protein. The latter possibility was not examined in this project, as evidence was found which 107 suggested that the GRSPaV replicase contains a membrane associating signal sequence. While the GRSPaV replicase does not contain any predicted transmembrane domains, it may contain another type of membrane association signal. The alphavirus-like viruses SFV and BMV each contain an amphipathic a-helix in the MTR domain of their respective replicase proteins, which mediates membrane association in a monotopic fashion (Spuul et al, 2007; Liu et al, 2009). Such signals are characterized by alternate polar and non-polar amino acid residues with the propensity to form amphipathic helices. In silico analysis of the GRSPaV replicase predicted an amphipathic region (aa 165-187) within the MTR domain of the replicase. In addition, the monotopic signal sequence of SFV (which was the only pertinent sequence available at the time) aligned with a region of the GRSPaV MTR domain (aa 157-173). From the alignment, counterpart residues important towards forming a monotopic signal were identified in the GRSPaV replicase. Taken together, these results strongly suggested that if an amphipathic monotopic signal is present in the GRSPaV replicase, it likely resides within the MTR domain between amino acids 132-207, which was dubbed the "sub-domain". To test if the sub-domain was important for the formation of the viral puncta observed during expression of the full-length replicase, three truncations were created surrounding the sub-domain, each with a GFP tag fused to the C-terminus. One truncation contained the complete MTR domain (pMTR:GFP). The second truncation contained the MTR domain with the sub-domain deleted (pMTRASD:GFP), and the third truncation contained only the sub-domain fused to GFP (pSD:GFP). Expression of the entire MTR domain (pMTR:GFP) or the sub-domain alone (pSD:GFP) resulted in the formation of the viral puncta. Conversely, when the sub-domain was removed from the 108 MTR domain (pMTRASD:GFP), the remaining MTR peptide was no longer able to form punctate structures. The results suggested that the hypothesis that the sub-domain is important for the formation of the viral puncta was correct. It is unlikely that the formation of viral puncta was due to hydrophobic aggregation, as the sub-domain contains no significant hydrophobic regions (data not shown). The formation of cytoplasmic punctate structures due to a region of a MTR domain has also been reported for the alphavirus-like viruses SFV and BMV. When fused to GFP, the C-terminal region (aa 392-409) of the SFV MTR domain forms puncta which localize to endosomes, lysosomes, mitochondria, and the ER in HeLa cells (Spuul et al, 2007). Similarly, a region of the la protein of BMV, which maps to a C-terminal region of the MTR domain (aa 245-264), forms cytoplasmic punctate structures when expressed in Saccharomyces cerevisae (den Boon et al, 2001). Unlike the puncta produced by the SFV MTR truncation, those produced by the BMV MTR truncation did not associate with any membrane examined (i.e. the ER and Golgi). Regardless, a common finding among these two alphavirus-like replicases is that the MTR domain appears to determine, or at least affect, the sub-cellular localization of the replicase. The same appears to be true for GRSPaV, suggesting that the MTR domain may serve multiple functions during viral infection (i.e. directing the replication complex to the site of replication in addition to viral RNA capping). 4.3 The GRSPaV replicase polyprotein residues R173 and W184 are not critical for the formation of viral puncta The "sub-domain", a region of the GRSPaV MTR domain, was shown to be 109 involved in the formation of viral puncta. It was hypothesized that the viral puncta represented replicase compexes in association with a host membrane. A corollary of this initial hypothesis was the assumption that the replicase must associate with a membrane to contribute to the formation of the replication vesicles. Specifically, the replicase was thought to associate with a membrane via a monotopic signal located within the sub- domain. The most well characterized monotopic signal at the time was that which is found in the MTR domain of the SFV replicase protein nsPl (Spuul et al, 2007). Alignment of the GRSPaV and SFV MTR domains identified two residues, R173 and W184, within the GRSPaV sub-domain which may have been required for the putative GRSPaV monotopic signal to function, based on the fact that their SFV counterparts are critical for membrane association (Spuul et al, 2007). Destruction of these domains was therefore expected to manifest as the inability of the mutant constructs to associate with membranes and in turn form viral puncta. However, substitution of the arginine residue alone (R173A), the tryptophan residue alone (W184A), or both residues together (R173A,W184A) had no effect on the formation of the viral puncta. From these results it was concluded that the residues R173 and W184 are not required for the formation of the viral puncta. There are several explanations for the failure of the R173A and W184A mutations to inhibit puncta formation. Firstly, puncta formation may not require membrane association. It was hypothesized that puncta formation involved the GRSPaV replicase interacting with, and re-organizing, a host membrane. If membrane association is not required for the formation of viral puncta, then the destruction of the putative membrane association signal would have no phenotypic effect during the experiments reported here. 110 Secondly, it is possible that association of the replicase with a membrane is required for the formation of viral puncta, but the introduced mutations (i.e. R173A, W184A, and R173A,W184A) did not abolish the ability of the sub-domain to bind to the membrane. In this scenario, the residues R173 and W184 may not be critical with regards to the proper functioning of the putative monotopic signal. Instead, other amino acids in the vicinity may be fulfilling the critical functions (interaction with phospholipid head groups and penetration into the bilayer core for enhanced binding strength). While the functions of the critical arginine and tryptophan residues in the SFV membrane association signal are likely a common requirement of all viral monotopic signals, the specific amino acid residues which fulfill those functions might vary. For example, the function of the SFV critical tryptophan could possibly be performed by another amino acid with a similar bulky and hydrophobic side chain, such as phenylalanine. Likewise, a lysine residue could probably replace the critical arginine residue since both amino acids are positively charged and polar (and thus likely to interact with phospholipid head groups). Until more viral monotopic signals are discovered and characterized, the extent to which the amino acid sequences of different signals can vary will remain unknown. It should also be noted that, while the region surrounding aa 157-173 was predicted to be the most probable location of a monotopic signal within the GRSPaV MTR domain (based on the SFV alignment), the region important for viral puncta formation (i.e. sub- domain) encompasses aa 132-207, so a monotopic signal could theoretically reside in other locations. It is also possible that membrane association signals reside in other regions of the replicase polyprotein. Two large amphipathic regions have recently been identified at amino acid positions 325-410 (data not shown). Further truncations or 111 alanine scanning mutagenesis of the sub-domain and/or replicase is required to analyze this possibility. Another explanation as to why the R173A, W184A, and R173A,W184A mutations did not abolish puncta formation is that there may not be a monotopic signal, or any type of membrane association signal, within the GRSPaV replicase. If this were true, the replicase could be associating with a membrane to form the viral puncta via interaction with a host protein. If the interacting domain is within the sub-domain of the GRSPaV replicase, the results of the truncation constructs as well as the results of the mutagenesis of R173 and W184 would be expected. Unless the mutations (i.e. R173A, W184A, and R173A/W184A) interfered with the ability of the sub-domain to interact with the necessary host protein(s), a phenotypic effect would not be observed. To identify a possible interacting protein, the use of a yeast-two-hybrid assay or co- immunoprecipitation of the GRSPaV replicase could be performed. However, these methods require either the construction of a N. tabacum cv. BY-2 cDNA library or the ability to detect the suspected interacting host protein. The most practical method may be to start with a plant virus model known to utilize this method of membrane association. The replicase of TMV interacts with TOM1 and TOM3 from A. thaliana and a TOM1 homologue from N. benthamiana in order to associate with the ER (Yamanaka et al, 2002; Chen et al, 2007). A TOM1 homologue, ntTOMl (Genbank # AB193039), exists in N. tabacum (Chen et al, 2007), so a yeast-two-hybrid screen focussing on ntTOMl may be a prudent first step in discovering one or more interacting proteins. Prior to the above experiments, co-immunoprecipitation of the GRSPaV replicase followed by two- dimensional sodium-dodecyl sulphate polyacrylamide gel electrophoresis may be useful 112 to determine if a second protein co-precipitates with the GRSPaV replicase. Regardless, the possibility that the GRSPaV replicase associates with a host membrane via interaction with a host protein must be considered. 4.4 The viral puncta localize adjacent to the ER lumen and appear to align along actin microfilaments The replicase of all (+)RNA viruses must associate with a host membrane in order for replication to occur, and GRSPaV is not expected to be an exception. The sub- domain, located within the MTR domain, was found to be responsible for the formation of punctate structures in the cytoplasm of N. tabacum cv. BY-2 cells, though the question of how the sub-domain stimulates puncta formation was not answered. To determine where the viral puncta were forming, the MTR domain was co-expressed with fluorescent organelle-specific proteins. Since many (+)RNA plant virus replicases associate with the ER, this was the first organelle tested via co-expression of the MTR domain, fused to GFP, with an ER lumen marker protein. While the majority of viral puncta did not co- localize exactly with the ER lumen marker, most of the puncta localized immediately adjacent to the ER, suggesting that the GRSPaV replicase may be deriving membrane from the ER to form replication vesicles. Since the ER marker used in these experiments marked only the ER lumen, and not the ER membrane, any membranous material disassociated from the ER network would not be visible during fluorescence microscopy. This would explain why the viral puncta, which were observed adjacent to the ER, were not marked by the ER lumen marker. Futher experiments are required in which the MTR domain will be co-expressed with an ER membrane marker. 113 Many (+)RNA viruses replicate in association with the ER membrane, including the well characterized plant virus BMV (den Boon et al, 2001). The MTR domain of the BMV replicase also localizes adjacent to the ER when expressed transiently (den Boon et al, 2001). Since the full-length BMV replicase localizes directly over the ER, not adjacent to the ER, the authors concluded that another region of the replicase, in addition to the MTR domain, is required for normal ER localization (see Liu et al, 2009). The full-length GRSPaV replicase may behave in a similar fashion (i.e. it would co-localize perfectly with the ER lumen marker). Due to technical issues encountered when manipulating the full-length replicase (see Appendix I), it was not possible to determine the sub-cellular location of the full-length replicase during the course of this project. To overcome this issue, the full-length ORF1-GFP fusion is currently being cloned into pRTL2, which will be used for further localization experiments. Due to recent reports of the TMV replicase associating with and trafficking along actin microfilaments (Liu et al, 2005; Wright et al, 2007; Guenoune-Gelbart et al, 2008; Sambade et al, 2008), the MTR domain of the GRSPaV replicase was co-expressed with an actin marker. The viral puncta were observed aligning along the actin filaments, which may account for the movement of the puncta noted when the full-length or truncation constructs were expressed alone. Since the actin network and viral puncta could not be observed simultaneously, it was not possible to determine if the puncta were indeed moving along the filaments. However, the physical contact of almost every punctate structure with the actin marker in each cell examined strongly suggests an interaction between the viral puncta and actin microfilaments. A new theory, based on the above localization findings, is that the GRSPaV replicase may be harvesting ER 114 membrane in order to form replication vesicles, which then associate with actin microfilaments presumably for intracellular transport. Further experimentation is required to test this theory. 4.5 The GRSPaV RdRp domain forms aggregate structures in tobacco BY-2 protoplasts when expressed alone and may interact with the MTR domain When the full-length GRSPaV replicase was expressed without a GFP tag and observed via immunostaining, the primary antibody used specifically detected the RdRp domain of the replicase. Similarly, when the full-length GRSPaV replicase was observed directly in protoplasts, the GFP tag used to visualize the replicase during fluorescence microscopy was fused to the C-terminus of the replicase (i.e. RdRp domain). Also, the RdRp domain was hypothesized to be auto-catalytically cleaved from the rest of the replicase polyprotein. It was therefore prudent to express the RdRp domain alone to discount the possibility that the RdRp domain was forming viral puncta independently of the rest of the replicase. The RdRp domain was not expected to alter the sub-cellular distribution of GFP, as the RdRp domain does not contain identifiable localization signals. However, when GFP was fused to the C-terminus of the RdRp domain and expressed in protoplasts, aggregate structures were observed in the cytoplasm. Time lapse video of an individual protoplast, starting at approximately 10 hpt and ending at approximately 18 hpt, showed the aggregate structures to appear at 11 hpt, grow to approximately 4 urn in diameter by 13 hpt, and remain stationary until the end of the time course experiment. The structures observed in this experiment are probably due to hydrophobic aggregation of the RdRp:GFP fusion protein because: (1) the RdRp domain 115 does not contain a known signal sequence and should therefore diffuse evenly throughout the cytosol; and (2) The RdRp domain of two other related plant viruses, BMV and TYMV, has also been reported to form cytosolic aggregates under certain conditions (Chen and Ahlquist, 2000; Prod'homme et al, 2003). When the BMV RdRp domain was fused to GFP and expressed in yeast, aggregate structures formed in the cytoplasm (Chen and Ahlquist, 2000). While these authors did not determine the sub-cellular location of the aggregates, they demonstrated that the RdRp was not associating with the ER (site of BMV replication) or Golgi. The RdRp domain of TYMV, when fused to GFP, formed aggregate structures in the cytosol that did not localize to the site of TYMV replication (i.e. chloroplast outer membrane) (Prod'homme et al, 2001; Prod'homme, et al, 2003). When the TYMV RdRp domain was expressed without a GFP tag and detected via fluorescence immunostaining, the RdRp did not aggregate and a uniform cytosolic pattern was observed. The RdRp aggregates reported here, as well as for BMV and TYMV, may therefore represent aggregation stimulated by over-expression and the large GFP tag. The three dimensional structure of the RdRp:GFP fusion protein may contain significant hydrophobic surfaces and thus have a tendency to aggregate. The conservation of plant viral RdRp domains (Koonin, 1993) may explain why similar behaviour was observed for a bromovirus (BMV), a tymovirus (TYMV) and a foveavirus (GRSPaV). Fluorescent immunostaining to detect the GRSPaV RdRp domain when expressed alone (see Appendix III) would be prudent in future experiments to avoid possible hydrophobic aggregation. When the aforementioned BMV RdRp:GFP fusion protein was expressed with the BMV la protein, which contains the rest of the BMV replicase, the RdRp was found to 116 co-localize with la at the ER instead of forming cytosolic aggregates (Chen and Ahlquist, 2000). These authors concluded that the BMV RdRp domain is directed to the ER by the la protein, but forms unidentified aggregates in the cytoplasm in the absence of la. Similarly, when the rest of the TYMV replicase was supplied, the TYMV RdRp:GFP fusion protein was able to localize to the chloroplasts instead of forming cytosolic aggregates (Prod'homme et al, 2003). To determine if the GRSPaV RdRp domain behaves in a similar manner, the RdRp domain, fused to GFP, was co-expressed with the GRSPaV MTR domain fused to mRFP. In the presence of the MTR fusion, the RdRp domain formed small punctate structures that appeared similar in size, distribution, and mobility to the viral puncta formed by the MTR fusion. Indeed, when the two fluorescent images were merged, the puncta produced by the RdRp domain and MTR domain were shown to co-localize. While it can be concluded only that the MTR domain affects the sub-cellular localization of the RdRp domain, the MTR and RdRp domains may be interacting for this effect to occur. Further yeast-two-hybrid or co-immunoprecipitation experiments would help answer this question. 4.6 Conclusions Positive RNA viruses are obliged to replicate in association with a host membrane. The choice of membrane depends on the specific virus. GRSPaV is a member of the virus family Flexiviridae, of which little is known. As a (+)RNA virus, the replicase of GRSPaV is expected to associate with a host membrane in order to function. The purpose of this study was to elucidate details about the sub-cellular localization of the GRSPaV replicase. Questions which were addressed included: (1) 117 Does the GRSPaV replicase form viral replication structures when expressed transiently in Nicotiana tabacum cv. BY-2 cells; (2) How does the GRSPaV replicase form such structures (i.e. which region of the replicase is important for this function); and (3) Where does the GRSPaV replicase localize within BY-2 cells. The findings that the GRSPaV replicase formed punctate structures in the cytoplasm re-enforced the hypothesis that the replicase would concentrate in distinct membrane-derived vesicles, though direct evidence that a membrane association was occurring was not found. The region required for the localization of the replicase into "viral puncta" was traced to a region of the MTR domain, termed the "sub-domain" (aa 132-207). Further analysis of the sub-domain revealed that the residues R173 and W184 were not critical with respect to the ability of the sub-domain to form viral puncta. This finding weakened the hypothesis that a putative monotopic signal, located within the sub- domain at aa 165-187, was responsible for membrane association and the subsequent formation of viral puncta. However, the sub-domain may still harbour a membrane association signal elsewhere, or possibly a host protein interaction domain, in order to associate with a host membrane and form viral puncta. To deduce the sub-cellular location of the sub-domain induced puncta, the MTR domain, fused to GFP was co-expressed with dsRed or mRFP tagged marker proteins that label the ER, actin microfilaments, Golgi stacks, or peroxisomes. The ER and actin microfilaments were tested as possible sites of viral puncta formation because related plant virus replicases localize to these organelles. Golgi stacks and peroxisomes were also examined as possible locations of the viral puncta because the size and distribution of the viral puncta resembled those of these organelles. The viral puncta, produced by the 118 sub-domain, were found to localize adjacent to the ER lumen, and appeared to align along actin microfilaments. Based on these findings as well as literature on related plant virus replicase studies, a new theory was developed: The GRSPaV replicase may be deriving membrane from the ER from which replication vesicles are formed (and observed as viral puncta). Membrane association is probably accomplished by either the interaction of the GRSPaV replicase with an integral membrane host protein (similar to TMV), or by a currently unidentified monotopic signal (simiar to SFV and BMV). The replication vesicles may then associate with actin microfilaments via the MTR domain, presumably for translocation to another region of the cell. Indeed, the N-terminal domain of TMV has recently been found to associate with actin microfilaments when expressed in the absence of the rest of the replicase (Harries et al, 2009). It should be noted that no direct evidence was found to support the trafficking of the GRSPaV viral puncta along actin microfilaments. Rather, it would be reasonable to presume that the replicase would associate with actin filaments only for the purpose of intracellular transport. However, validation of the above theory is lacking. The above theory can also be incorporated into the present model of the behaviour of the GRSPaV TGB movement proteins (see section 1.6.4). Briefly, TGBpl binds the viral ribonucleoprotein complex and associates it with ER-derived vesicles containing TGBp2 (Rebelo et al, 2008). These vesicles are then expected to traffic along actin microfilaments (requiring TGBp3) to the cell periphery. Given that small RNA viruses such as GRSPaV have evolved to be very efficient and generally do not utilize redundant tactics, it is tempting to speculate that the ER-derived vesicles in the TGB model are the 119 same as those theorized to be derived by the GRSPaV replicase. It would bot be very efficient for the virus to utilize one vesicle for replication and then have to move all of the viral cargo to a second vesicle for movement. Rather, it would be much more cost- effective for the virus to use one vesicle for multiple purposes. In such a model, the GRSPaV replicase may derive vesicles from the ER from within which replication occurs. At the same time, TGBpl binds to genomic viral RNA and TGBp2 and TGBp3 associate with the vesicle membrane. The MTR domain of the replicase then mediates the association of the vesicles with actin microfilaments. Through normal cellular trafficking, and perhaps the actions of the TGB proteins, the vesicles are then transported to the cell periphery, where one or multiple TGB proteins recognize and manipulate the plasmodesmata for egress of the newly replicated viral RNA. In this new model, viral replication and movement, typically thought of as separate events, are intertwined as a single complex process. Indeed, the TGBp3 movement protein of PVX has recently been shown to co-localize with the PVX replicase in ER-derived vesicles (Bamunusinghe et al, 2009). Further research is required to determine if this model is accurate, not only for GRSPaV, but also for members of Flexiviridae and plant viruses in general. The GRSPaV RdRp domain was studied independently from the rest of the replicase via transient expression in BY-2 protoplasts. The RdRp domain, when fused to GFP and expressed alone, formed cytosolic aggregate structures. The aggregate structures differed from the viral puncta produced by the sub-domain in that they were larger (approximately 4 um in diameter compared to the viral puncta diameter of 1 urn), and the aggregates were immobile as shown by time-lapse video. Based on similar findings from expression of other plant virus RdRp domains, as well as the fact that the 120 GRSPaV RdRp domain contains no known membrane association domains, it was concluded that the RdRp:GFP fusion protein was likely forming hydrophobic aggregates in the cytosol of transfected cells due to over-expression. Studies on similar plant viruses suggested that co-expression of the RdRp domain with the rest of the viral replicase would result in the re-distribution of the RdRp to the site of viral replication. When the GRSPaV RdRp domain, fused to GFP, was co-expressed with the GRSPaV MTR domain fused to mRFP, the RdRp:GFP fusion protein co-localized with the viral puncta produced by the MTR:mRFP fusion protein. Additionally, the punctate structures resulting from the above co-expression were very similar to those produced by the MTR domain alone (e.g. the RdRp puncta were approximately 1 um in diameter and mobile). These results suggested that the MTR domain is able to change the localization of the RdRp:GFP fusion protein from large immobile aggregate structures to smaller, mobile punctate structures. This effect is presumably dependent on a MTR:RdRp interaction, though direct evidence of such will require further experimentation. To summarize, the following can be concluded from the experiments described here, though it should be noted that these conclusions are based on experiments conducted on a grapevine virus in tobacco cells, so any results may theoretically represent artefacts due to the non-native system: 1. The GRSPaV replicase, whether detected via immunostaining or a GFP tag, forms punctate structures in the cytoplasm when transiently expressed in Nicotiana tabacum cv. BY-2 cells. 2. A region located within the MTR domain (aa 132-207) is responsible for the formation of punctate structures. 121 3. The amino acid residues R173 and W184, which were thought to be part of a monotopic signal, are not required for the formation of the above punctate structures. 4. These punctate structures localize adjacent to the ER lumen and co-localize with actin microfilaments but not Golgi stacks or peroxisomes. 5. The GRSPaV RdRp domain forms aggregate structures in the cytoplasm of BY-2 cells when transiently expressed as a GFP fusion, but re-distributes upon co-expression with the MTR domain to form puncta similar in size and position to the MTR punctate structures. While many studies have been conducted on plant virus movement and replication, the unanswered questions still vastly outweigh what we know about these elusive pathogens. Plant viruses employ a broad spectrum of strategies in every aspect of the infection cycle. Replication and movement models based on research conducted on one virus may not necessarily explain the behaviour of another virus. Thus, a full understanding of plant virus molecular biology, indispensible for the control of economically important pathogens, will require more in-depth studies on representative members of each family. Reported here are the first steps towards understanding the replication machinery of GRSPaV and the sub-cellular localization of the replicase of a member of the virus family Flexiviridae. 122 REFERENCES Aas PA, Otterlei M, Falnes PO, Vagbo CB, Skorpen F, Akbari M, Sundheim O, Bjoras M, Slupphaug G, Seeberg E, Krokan HE (2003) Human and bacterial oxidative demethylases repair alkylation damage in both RNA and DNA. Nature 421: 859-863 Adams MJ, Antoniw JF, Bar-Joseph M, Brunt AA, Candresse T, Foster GD, Martelli GP, Milne RG, Zavriev SK, Fauquet CM (2004) The new plant virus family Flexiviridae and assessment of molecular criteria for species demarcation. Arch Virol 149: 1045-1060 Ahola T, Kujala P, Tuittila M, Blom T, Laakkonen P, Hinkkanen A, Auvinen P (2000) Effects of palmitoylation of replicase protein nsPl on alphavirus infection. J Virol 74: 6725-6733 Bamunusinghe D, Hemenway C, Nelson R, Sanderfoot A, Ye C, Silva M, Payton M, Verchot-Lubicz J (2009) Analysis of the Potato virus X replicase and TGBp3 subcellular locations. Virology 393: 272-285 Banjoko A, Trelease RN (1995) Development and application of an in vivo plant peroxisome import system. Plant Physiol 107: 1201-1208 Boscia D, Savino V, Minafra A, Namba S, Elicio V, Castellano MA, Gonsalves D, Martelli GP (1993) Properties of a filamentous virus isolated from grapevines affected by corky bark. Arch Virol 130: 109-120 Bristow PR, Martin RR, Windom GE (2000) Transmission, Field Spread, Cultivar Response, and Impact on Yield in Highbush Blueberry Infected with Blueberry scorch virus. Phytopathology 90: 474-479 Buck KW (1999) Replication of Tobacco mosaic virus RNA. Philos Trans R Soc Lond B Biol Sci 354: 613-627 Buck KW (1996) Comparison of the replication of positive-stranded RNA viruses of plants and animals. Adv Virus Res 47: 159-251 Bushell M, Sarnow P (2002) Hijacking the translation apparatus by RNA viruses. J Cell Biol 158: 395-399 Carrington J, Freed D (1990) Cap-independent enhancement of translation by a plant potyvirus 5' nontranslated region. J Virol 64: 1590-1597 Carstens E (2009) Report from the 40th meeting of the Executive Committee of the International Committee of Taxonomy of Viruses. Arch Virol July 19 123 Chen B, Jiang JH, Zhou XP (2007) A TOM1 homologue is required for multiplication of Tobacco mosaic virus in Nicotiana benthamiana. J Zhejiang Univ Sci B 8: 256-259 Chen J, Ahlquist P (2000) Brome mosaic virus polymerase-like protein 2a is directed to the endoplasmic reticulum by helicase-like viral protein la. J Virol 74: 4310-4318 Chevalier S, Greif C, Clauzel J, Walter B, Fritsch C (1995) Use of an immunocapture- polymerase chain reaction procedure for the detection of Grapevine virus A in Kober stem grooving-infected grapevines. J Phytopathol 143: 368-373 David C, Gargouri-Bouzid R, Haenni AL (1992) RNA replication of plant viruses containing an RNA genome. Prog Nucleic Acid Res Mol Biol 42: 157-227 den Boon JA, Chen J, Ahlquist P (2001) Identification of sequences in Brome mosaic virus replicase protein 1 a that mediate association with endoplasmic reticulum membranes. J Virol 75: 12370-12381 Dolja VV, Kreuze JF, Valkonen JP (2006) Comparative and functional genomics of closteroviruses. Virus Res 117: 38-51 dos Reis Figueira A, Golem S, Goregaoker SP, Culver JN (2002) A nuclear localization signal and a membrane association domain contribute to the cellular localization of the Tobacco mosaic virus 126-kDa replicase protein. Virology 301: 81-89 Escoubas JM, Lane D, Chandler M (1994) Is the IS1 transposase, InsAB', the only IS1- encoded protein required for efficient transposition? J Bacteriol 176: 5864-5867 Gentit P, Foissac X, Svanella-Dumas L, Peypelut M, Candresse T (2001) Characterization of two different Apricot latent virus variants associated with peach asteroid spot and peach sooty ringspot diseases. Arch Virol 146: 1453-1464 Gorbalenya AE, Koonin EV, Donchenko AP, Blinov VM (1988) A novel superfamily of nucleoside triphosphate-binding motif containing proteins which are probably involved in duplex unwinding in DNA and RNA replication and recombination. FEBS Lett 235: 16-24 Goregaoker SP, Culver JN (2003) Oligomerization and activity of the helicase domain of the Tobacco mosaic virus 126- and 183-kilodalton replicase proteins. J Virol 77: 3549- 3556 Granett J, Walker MA, Kocsis L, Omer AD (2001) Biology and management of grape phylloxera. Annu Rev Entomol 46: 387-412 Guenoune-Gelbart D, Elbaum M, Sagi G, Levy A, Epel BL (2008) Tobacco mosaic virus (TMV) replicase and movement protein function synergistically in facilitating TMV 124 spread by lateral diffusion in the plasmodesmal desmotubule of Nicotiana benthamiana. Mol Plant Microbe Interact 21: 335-345 Hall BG (1999) Transposable elements as activators of cryptic genes in E. coli. Genetica 107: 181-187 Harries P, Park J, Sasaki N, Ballard K, Maule A, Nelson R (2009) Differing requirements for actin and myosin by plant viruses for sustained intercellular movement. PNAS 106: 17594-17599 Haseloff J, Siemering K, Prasher D, Hodge S (1997) Removal of a cryptic intron and subcellular localization of green fluorescent protein are required to mark transgenic Arabidopsis plants brightly. PNAS 94: 2122-2127 Hu M, Deonier RC (1981) Comparison of IS1, IS2 and IS3 copy number in Escherichia coli strains K-12, B and C. Gene 16: 161-170 Jakubiec A, Jupin I (2007) Regulation of positive-strand RNA virus replication: the emerging role of phosphorylation. Virus Res 129: 73-79 Jakubiec A, Notaise J, Tournier V, Hericourt F, Block MA, Drugeon G, van Aelst L, Jupin I (2004) Assembly of Turnip yellow mosaic virus replication complexes: interaction between the proteinase and polymerase domains of the replication proteins. J Virol 78: 7945-7957 Jakubiec A, Tournier V, Drugeon G, Pflieger S, Camborde L, Vinh J, Hericourt F, Redeker V, Jupin I (2006) Phosphorylation of viral RNA-dependent RNA polymerase and its role in replication of plus-strand RNA virus. J Biol Chem 281: 21236-21249 Jonczyk M, Pathak KB, Sharma M, Nagy PD (2007) Exploiting alternative subcellular location for replication: tombusvirus replication switches to the endoplasmic reticulum in the absence of peroxisomes. Virology 362: 320-330 Kharat AS, Coursange E, Noirclerc-Savoye M, Lacoste J, Blot M (2006) IS1 transposition is enhanced by translation errors and by bacterial growth at extreme glucose levels. Acta Biochim Pol 53: 729-738 Koonin EV (1991) The phylogeny of RNA-dependent RNA polymerases of positive- strand RNA viruses. J Gen Virol 72: 2197-2206 Koonin EV, Dolja VV (1993) Evolution and taxonomy of positive-strand RNA viruses: implications of comparative analysis of amino acid sequences. Crit Rev Biochem Mol Biol 28: 375-430 125 Kost B, Spielhofer P, Chua N (1998) A GFP-mouse talin fusion protein labels plant actin filaments in vivo and visualizes the actin cytoskeleton in growing pollen tubes. Plant J 16:393-401 Laakkonen P, Ahola T, Kaariainen L (1996) The effects of palmitoylation on membrane association of Semliki forest virus RNA capping enzyme. J Biol Chem 271: 28567-28571 Lampio A, Kilpelainen I, Pesonen S, Karhi K, Auvinen P, Somerharju P, Kaariainen L (2000) Membrane binding mechanism of an RNA virus-capping enzyme. J Biol Chem 275: 37853-37859 Lawrence DM, Rozanov MN, Hillman BI (1995) Autocatalytic processing of the 223- kDa protein of Blueberry scorch carlavirus by a papain-like proteinase. Virology 207: 127-135 Lima MF, Alkowni R, Uyemoto JK, Golino D, Osman F, Rowhani A (2006) Molecular analysis of a California strain of Rupestris stem pitting-associated virus isolated from declining Syrah grapevines. Arch Virol 151: 1889-1894 Liu JZ, Blancafior EB, Nelson RS (2005) The Tobacco mosaic virus 126-kilodalton protein, a constituent of the virus replication complex, alone or within the complex aligns with and traffics along microfilaments. Plant Physiol 138: 1853-1865 Liu L, Westler W, den Boon J, Wang X, Diaz A, Steinberg H, Ahlquist P (2009) An amphipathic alpha-helix controls multiple roles ofBrome mosaic virus protein la in RNA replication complex assmebly and function. PLoS Pathog 5: el000351 Martelli GP (1993) Rugose wood complex. Graft-transmissible diseases of grapevines: Handbook for detection and diagnosis. Food and Agriculture Organization of the United Nations, Rome, Italy Martelli GP, Adams MJ, Kreuze JF, Dolja VV (2007) Family Flexiviridae: a case study in virion and genome plasticity. Annu Rev Phytopathol 45: 73-100 Martelli GP, Jelkmann W (1998) Foveavirus, a new plant virus genus. Arch Virol 143: 1245-1249 McCartney AW, Greenwood JS, Fabian MR, White KA, Mullen RT (2005) Localization of the Tomato bushy stunt virus replication protein p33 reveals a peroxisome-to-endoplasmic reticulum sorting pathway. Plant Cell 17: 3513-3531 Meng B, Gonsalves D (2008) Grapevine rupestris stem pitting-associated virus. In G Rao, A Myrta, K Ling, eds, Characterization, Diagnosis, and Management of Plant Viruses Vol 2. Studium Press LCC, Texas, USA, pp 201-222 126 Meng B, Gonsalves D (2007) Grapevine rupestris stem pitting-associated virus: A decade of research and future perspectives. Plant Viruses 1: 52-62 Meng B, Li C, Wang W (2009) Towards development of Grapevine rupestris stem pitting-associated virus into a VIGS vector. Extended Abstracts of the 16th Meeting of the ICVG,pp 331-332. Meng B, Li C, Wang W, Goszczynski D, Gonsalves D (2005) Complete genome sequences of two new variants of Grapevine rupestris stem pitting-associated virus and comparative analyses. J Gen Virol 86: 1555-1560 Meng B, Pang SZ, Forsline PL, McFerson JR, Gonsalves D (1998) Nucleotide sequence and genome structure of Grapevine rupestris stem pitting associated virus-l reveal similarities to Apple stem pitting virus. J Gen Virol 79: 2059-2069 Meng B, Rebelo AR, Fisher H (2006) Genetic diversity analyses of grapevine Rupestris stem pitting-associated virus reveal distinct population structures in scion versus rootstock varieties. J Gen Virol 87: 1725-1733 Meng B, Zhu HY, Gonsalves D (1999) Rupestris stem pitting associated virus-\ consists of a family of sequence variants. Arch Virol 144: 2071-2085 Molnar A, Csorba T, Lakatos L, Varallyay E, Lacomme C, Burgyan J (2005) Plant virus-derived small interfering RNAs originate predominantly from highly structured single-stranded viral RNAs. J Virol 79: 7812-7818 Mullins MG, Bouquet A,. Williams LE (1992) Mandahar (ed), Biology of the grapevine, Ed 1. Cambridge Iniveristy Press, Cambridge, England Nagy PD, Pogany J (2008) Multiple roles of viral replication proteins in plant RNA virus replication. Methods Mol Biol 451: 55-68 Osman F, Rowhani A (2008) Real-time RT-PCR (TaqMan) assays for the detection of viruses associated with Rugose wood complex of grapevine. J Virol Methods 154: 69-75 Pathak KB, Sasvari Z, Nagy PD (2008) The host Pexl9p plays a role in peroxisomal localization of tombusvirus replication proteins. Virology 379: 294-305 Peng CW, Dolja VV (2000) Leader proteinase of the Beet yellows closterovirus: mutation analysis of the function in genome amplification. J Virol 74: 9766-9770 Petrovic N, Meng B, Ravnikar M, Mavric I, Gonsalves D (2003) First detection of Rupestris stem pitting-associated virus particles in grapevine using the antibody to the recombinant coat protein. Plant Disease 87: 510-514 127 Pierce M, Essenberg M (1987) Localization of phytoalexins in fluorescent mesophyll cells isolated from bacterial blight-infected cotton cotyledons and separated from other cells by fluorescence-activated cell sorting. Physiol Mol Plant Path 31: 273-290 Pogany J, White KA, Nagy PD (2005) Specific binding of tombusvirus replication protein p33 to an internal replication element in the viral RNA is essential for replication. J Virol 79: 4859-4869 Prod'homme D, Jakubiec A, Tournier V, Drugeon G, Jupin I (2003) Targeting of the Turnip yellow mosaic virus 66K replication protein to the chloroplast envelope is mediated by the 140K protein. J Virol 77: 9124-9135 Prod'homme D, Le Panse S, Drugeon G, Jupin I (2001) Detection and subcellular localization of the Turnip yellow mosaic virus 66K replication protein in infected cells. Virology 281: 88-101 Rebelo AR, Niewiadomski S, Prosser SW, Krell P, Meng B (2008) Subcellular localization of the triple gene block proteins encoded by a Foveavirus infecting grapevines. Virus Res 138: 57-69 Restrepo-Hartwig MA, Ahlquist P (1996) Brome mosaic virus helicase- and polymerase-like proteins colocalize on the endoplasmic reticulum at sites of viral RNA synthesis. J Virol 70: 8908-8916 Rozanov MN, Koonin EV, Gorbalenya AE (1992) Conservation of the putative methyltransferase domain: a hallmark of the 'Sindbis-like' supergroup of positive-strand RNA viruses. J Gen Virol 73: 2129-2134 Salonen A, Ahola T, Kaariainen L (2005) Viral RNA replication in association with cellular membranes. Curr Top Microbiol Immunol 285: 139-173 Sambade A, Brandner K, Hofmann C, Seemanpillai M, Mutterer J, Heinlein M (2008) Transport of TMV movement protein particles associated with the targeting of RNA to plasmodesmata. Traffic 9: 2073-2088 Sekino N, Sekine Y, Ohtsubo E (1995) IS 1-encoded proteins, Ins A and the InsA-B'- InsB transframe protein (transposase): functions deduced from their DNA-binding ability. AdvBiophys 31: 209-222 Serva S, Nagy PD (2006) Proteomics analysis of the tombusvirus replicase: Hsp70 molecular chaperone is associated with the replicase and enhances viral RNA replication. J Virol 80: 2162-2169 Shapka N, Stork J, Nagy P (2005) Phosphorylation of the p33 replication protein of Cucumber necrosis tombusvirus adjacent to the RNA binding site affects viral RNA replication. Virology 343: 65-78 128 Sharadadevi A, Sivakamasundari C, Nagaraj R (2005) Amphipathic alpha-helices in proteins: results from analysis of protein structures. Proteins 59: 791-801 Shin HI, Tzanetakis IE, Dreher TW, Cho TJ (2009) The 5'-UTR of Turnip yellow mosaic virus does not include a critical encapsidation signal. Virology 387: 427-435 Sivakamasundari C, Nagaraj R (2009) Interaction of 18-residue peptides derived from the amphipathic helical segments of globular proteins with model membranes. J Biosci 34: 239-250 Stork J, Panaviene Z, Nagy P (2005) Inhibition of in vitro RNA binding and replicase activity by phosphorylation of the p33 replication protein of Cucumber necrosis tombusvirus. Virology 343: 79-92 Spuul P, Salonen A, Merits A, Jokitalo E, Kaariainen L, Ahola T (2007) Role of the amphipathic peptide oiSemliki Forest virus replicase protein nsPl in membrane association and virus replication. J Virol 81: 872-883 Sztuba-Solinska J, Bujarski JJ (2008) Insights into the single-cell reproduction cycle of members of the family Bromoviridae: lessons from the use of protoplast systems. J Virol 82: 10330-10340 This P, Lacombe T, Thomas MR (2006) Historical origins and genetic diversity of wine grapes. Trends Genet 22: 511-519 Torrance L, Andreev IA, Gabrenaite-Verhovskaya R, Cowan G, Makinen K, Taliansky ME (2006) An unusual structure at one end of potato potyvirus particles. J Mol Biol 357: 1-8 Umeda M, Ohtsubo E (1991) Four types of IS1 with differences in nucleotide sequence reside in the Escherichia coli K-12 chromosome. Gene 98: 1-5 van den Born E, Omelchenko MV, Bekkelund A, Leihne V, Koonin EV, Dolja VV, Falnes PO (2008) Viral AlkB proteins repair RNA damage by oxidative demethylation. Nucleic Acids Res 36: 5451-5461 van der Heijden MW, Bol JF (2002) Composition of alphavirus-like replication complexes: involvement of virus and host encoded proteins. Arch Virol 147: 875-898 van Der Heijden MW, Carette JE, Reinhoud PJ, Haegi A, Bol JF (2001) Alfalfa mosaic virus replicase proteins P1 and P2 interact and colocalize at the vacuolar membrane. J Virol 75: 1879-1887 van Kammen A (1985) The replication of plant virus RNA. Microbiol Sci 2: 170-174 129 Verchot-Lubicz J, Ye CM, Bamunusinghe D (2007) Molecular biology of potexviruses: recent advances. J Gen Virol 88: 1643-1655 Wang RY, Stork J, Nagy PD (2009) A key role for heat shock protein 70 in the localization and insertion of tombusvirus replication proteins to intracellular membranes. J Virol 83: 3276-3287 Watanabe T, Honda A, Iwata A, Ueda S, Hibi T, Ishihama A (1999) Isolation from Tobacco mosaic vzVws-infected tobacco of a solubilized template-specific RNA- dependent RNA polymerase containing a 126K/183K protein heterodimer. J Virol 73: 2633-2640 Wee E, Sherrier D, Prime T, Dupree P (1998) Targeting of active sialyltransferase to the plant Golgi apparatus. Plant Cell 10: 1759-1768 Wei T, Wang A (2008) Biogenesis of cytoplasmic membranous vesicles for plant potyvirus replication occurs at endoplasmic reticulum exit sites in a COPI- and COPII- dependent manner. J Virol 82: 12252-12264 Wright KM, Wood NT, Roberts AG, Chapman S, Boevink P, Mackenzie KM, Oparka KJ (2007) Targeting of TMV movement protein to plasmodesmata requires the actin/ER network: evidence from FRAP. Traffic 8: 21-31 Yamanaka T, Imai T, Satoh R, Kawashima A, Takahashi M, Tomita K, Kubota K, Meshi T, Naito S, Ishikawa M (2002) Complete inhibition of tobamovirus multiplication by simultaneous mutations in two homologous host genes. J Virol 76: 2491-2497 Yamanaka T, Ohta T, Takahashi M, Meshi T, Schmidt R, Dean C, Naito S, Ishikawa M (2000) TOM1, an Arabidopsis gene required for efficient multiplication of a tobamovirus, encodes a putative transmembrane protein. Proc Natl Acad Sci U S A 97: 10107-10112 Yau SY, Keshavarz-Moore E, Ward J (2008) Host strain influences on supercoiled plasmid DNA production in Escherichia coli: Implications for efficient design of large- scale processes. Biotechnol Bioeng 101: 529-544 Zhang Y, Uyemoto JK, Golino D, Rowhani A (1998) Nucleotide sequence and RT- PCR detection of a virus associated with grapevine rupestris stem-pitting disease. Phytopathology 88: 1231-1237 Zhong X, Itaya A, Ding B (2005) Transfecting protoplasts by electroporation to study viroid replication. Curr Protoc Microbiol Chapter 16: Unit 16D.4 130 APPENDIX I: INSTABILITY OF pREP:GFP IN E. coli DH5a 1.1 Introduction Prior to the construction of pRSP-REP:GFP for expression of a GFP tagged full- length replicase polyprotein, attempts were made to clone the GRSPaV ORF1 sequence into a modified pRTL2 vector. This would allow for the transient over-expression of the full-length replicase polyprotein at levels higher than that resulting from the expression of pRSP-REP:GFP. The construct, termed pREP:GFP, was created and contained ORF1 fused at the C-terminus to GFP. However, pREP:GFP proved very difficult to recover from E. coli DH5a stocks, which showed symptoms of plasmid instability including failure of overnight cultures to grow, significantly low plasmid yields, and plasmids which appeared upon restriction analysis to contain extensive deletions. 1.2 Hypothesis The above phenomena led to the hypothesis that a cryptic prokaryotic promoter, located at nucleotide positions 3797-3824 of the GRSPaV genome, allowed for E. coli expression of the C-terminal region of the replicase polyprotein. If this expression product was also toxic to the E .coli cells, they would either die, which would account for the failure of some overnight cultures to grow, or surviving cells would be forced to alter the plasmid in some way in order to terminate the toxic expression, which would explain the deletion mutant plasmids often recovered from cultures that did grow. 1.3 Results Before construction of pREP:GFP could begin, pRTL2 had to be altered. pRTL2 131 contains an Ncol restriction site at the 5' end of the multi-cloning site which is typically utilized for cloning due to the presence of an optimal ATG translational start codon within its core sequence. This useful start codon does, however, preclude the possibility of using any of the downstream restriction sites for cloning, as translation of such would rely on "leaky scanning" by the ribosome. Also, since ORF1 of GRSPaV contains two naturally occurring Ncol restriction sites, it was not possible to use Ncol to clone ORJF1 into pRTL2. To overcome this problem, SDM was used with the primers RTLMUT1 and RTLMUT2 (see Table 1) to delete the Ncol site in pRTL2, with the resulting plasmid being re-named to pRTL3. ORF1 could then be cloned into pRTL3 using Kpnl and Xbal. When constructing pREP:GFP, numerous attempts to clone ORF1 as a single full- length 6.6 kb fragment with primers MTF1 and POLR2 failed. Thus, sequential cloning was used to clone smaller fragments one at a time, from the 5' terminus of ORF1 to the 3' terminus. ORF1 was cloned into pRTL3 in four fragments (Figure 20 A), with the intermediate plasmids being labelled pREPFl, pREPF2, and pREPF3. The primers used to amplify and clone each fragment (MTF1 + SacII-Rev, SacII-For + Nrul-Rev, Nrul-For + Clal-Rev, and Clal-For + POLR2) can be found in Table 1 and a diagram outlining the sequential cloning process can be found in Figure 20 A. Upon insertion of the final fragment, which contained approximately the last 2 kb of the GRSPaV genome (Figure 20 A), the above mentioned instability issues first appeared. The cryptic prokaryotic promoter hypothesis was supported not only by the nature of the instability issues, but also by in silico analysis of ORF1 and the fact that the appearance of the issues coincided with the insertion of the fourth fragment. Alignment of ORF 1 with the most common 132 Fig. 20. Strategy used to sequentially clone the GRSPaV replicase into pRTL3 and position of a putative prokaryotic cryptic promoter. Strategy used to clone the full- length GRSPaV replicase ORF (A) and nucleotide alignment of a portion of the GRSPaV replicase with a typical prokaryotic promoter sequence (B). Small horizontal arrows indicate primers used to amplify each fragment (solid lines) of ORF 1 which were cloned sequentially, yielding the intermediate constructs pREPFl, pREPF2, and pREPF3. The approximate size of each insert is shown, measured in base-pairs. The approximate location of the putative cryptic promoter in pREPF3 is represented by the large arrow. The relative positions of restriction sites used for sequential cloning are shown in (A). Mutations introduced into the -10 element of the putative cryptic promoter are shown in (B). These mutations had no effect on the issues attributed, to the putative cryptic promoter. 133 (A) -lOQQbp -1900 bp ORF1: myc MT HEL POL -1800 bp '* -21Q06p *™ pREPFl (1-1000) l§8* " Jgye^ MT HEL TTGACA 1ATAAT (iCiArrAaArra^A»T^v^TAATi^X5c^nx;KX-vvAATn'("rar,vrr'.'rr V pREPF2 (1-2600) ^^Y^ssjHa^weasB&afiij TyT^ MT HEL —v&tt^h ^*1—.*.* * •* A« »*&.<&* •• pREPR (1-4500) -"v ?"*%%,t>,;"t "r,™';?%l?%?fip?ss>~- —^*8§W5™<*| ^^ffl^s^w^^^^ 35S myc MT HEL' pREP (1-6500) (full length) 3B8~vriwe MT HEL. Kpnl Sad I Nru\ Cla\ Xba\ (B) -35 -10 E. coli promoter: TTGACA TATAAT GRSPaV sequence: CMTTGAAGTGATmATCCGCGTGGTAAAATTTCam T Mutations: C GGCC 134 prokaryotic promoter sequence showed a putative cryptic promoter located within the third fragment at nt 3797-3824 of ORF1 (Figure 20 B), which would only cause the expression of a significant polypeptide upon insertion of the forth fragment during the sequential cloning process. The position of the putative prokaryotic cryptic promoter was therefore consistent with the observed instability issues. To test the cryptic promoter hypothesis and resolve the instability issue thought to hinder cloning efforts, SDM was used in conjunction with the primers ProMutF and ProMutR (see Table 1) to delete critical residues within the -10 element (TATA box) of the cryptic promoter (Figure 20 B). Unfortunately, alteration of the TATA box had no effect on the instability issues, as the mutated construct, pREP:GFPAPro, displayed instability symptoms similar to the wildtype construct. It was concluded that the putative cryptic promoter is not responsible for the instability issues reported here. In a possibly related phenomenon, the full-length replicase construct pRSP- REP:GFP (see section 2.12.1) was also found to be unstable in E. coli DH5a. While amplification of the plasmid DNA in E. coli SURE2 cells alleviated the issues, a truncated plasmid mutant recovered from DH5a cells was analyzed in an effort to understand the instability problems. Restriction analysis was performed using the enzymes Sacll, Spel, Ncol, Nrul, Seal, and Smal, which are known to cut the GRSPaV genome in various locations with the exception of Smal, which cuts the vector backbone downstream of the multi-cloning site. Surprisingly, only Smal was able to linearize the mutant plasmid, suggesting that the GRSPaV genome was no longer present. PCR analysis using various primer sets to detect regions of the GRSPaV genome confirmed the absence of the majority of the genome. However, the 5" UTR was detected using 135 primers that bind to the 5' UTR and the vector backbone downstream of the multi-cloning site. These primers would therefore amplify the 5' UTR and any downstream sequence. The size of this amplicon suggested that an additional 750-800 bp of DNA was still present downstream of the viral 5' UTR region. Interestingly, sequencing revealed that an E .coli transposon, IS1, had replaced the viral genome immediately downstream of the viral 5' UTR (Figure 21). The IS1 transposon, which is 803 nucleotides long, is found in many different bacteria and is present in multiple copies in most instances. E. coli K12, the strain from which DH5a was derived, contains seven copies of IS1 (isA to isG) (Hu and Deonier, 1981; Umeda and Ohtsubo, 1991). The IS1 transposon contains two overlapping ORFs, termed insA and insB, which are flanked by non-identical inverted terminal repeats (Sekino et al, 1995). The product of insA contains a DNA-binding motif but is non-functional with respect to transposition. A "-1" ribosomal frameshift at the 3' terminus of insA allows for the continued translation of insB, resulting in the transframe protein insA-B. InsA-B possesses transposase activity and has been shown to bind the inverted terminal repeats of IS1, via the DNA-binding domain of insA, and mediate transposition of IS1 (Sekino et al, 1995). The frameshift required to activate the IS1 transposon is thought to occur naturally in only 1 out of 100 insA transcription events (Escoubas et al, 1994). Sequence alignments of the pRSP-REP:GFP deletion mutant with all seven E. coli K12 IS1 transposon copies showed that the inserted transposon is either is A or isE, which are both identical in nucleotide sequence (Umeda and Ohtsubo, 1991). It is unknown why isA or isE mobilized and replaced the GRSPaV genome during the cloning process while using DH5a, however there are two possibilities: (1) The transposon was activated 136 Fig. 21. Schematic diagram showing the insertion of the E. coli IS1 transposon into the GRSPaV genome during cloning procedures using E. coli DH5a as a host strain. The GRSPaV 5' un-translated region (UTR) is shown with the IS1 transposon inserted immediately downstream, replacing ORF1 starting from the ATG start codon. The relative GRSPaV nucleotide (nt) positions of each DNA element are shown, as well as the nucleotide sequence of each element. Sequence in bold is the IS1 5' terminal repeat region required for transposon integration. 137 DNA origin: GRSPaV Position (nt): (1-61) 5'...TATAATTGCAACGCA.GGTAACGACTCCAACTTATTGATA., i •c 1 5i 1 " GRSPaV 5' UTR E coli IS1 Transposon 138 naturally and through coincidence and random chance, inserted into pRSP-REP:GFP at the exact start codon of ORF1; or (2) The transposon was somehow stimulated into action, and due to the increased mobile activity, inserted in many target locations of the E. coli genome and also pRSP-REP:GFP. In support of the latter theory, IS1 transposon activity increases when the cell is under stress (Hall, 1999). Perhaps a prokaryotic cryptic promoter is indeed allowing expression of a toxic portion of the GRSPaV replicase polyprotein, which is in turn causing a stress response leading to increased mobile genetic element activity. Another explanation is simply that the larger size (10.5 kilo-base pairs) of the plasmid being housed in DH5a is causing a substantial increase in the metabolic demand of the cells, with the end result again being a stress response. In support of this latter theory, E. coli DH5a has been shown to grow significantly faster than the SURE2 strain, which may explain the stability of larger plasmids in SURE2 cells compared to DH5a (Yau et al, 2008). Additionally, low glucose levels have been shown to stimulate IS1 transposon mobility in E. coli K12 (Kharat et al, 2006). Undoubtedly, the combination of supporting a large plasmid while enduring rapid growth conditions would likely lead to glucose starvation quite quickly. In conclusion, the E. coli DH5a IS1 transposon, specifically isA or isE, inserted into pRSP-REP:GFP, replacing most of the viral sequence, likely due to a combination of factors during the cloning procedure. The combined effects of the large size of the plasmid and the rapid growth conditions of the DH5a strain most likely led to rapid glucose starvation, which in turn led to a general stress response including IS1 stimulation. The resulting insertion of isA/isE into pRSP-REP:GFP was likely random, however it is possible that the GRSPaV nucleotide sequence possesses some form of 139 favourable target sequence, though no such sequence as been reported for the IS1 transposon family. Regardless, this is the first known report of an E. coli transposable element inserting and replacing viral sequence during a typical cloning procedure. 140 APPENDIX II: GENERATION OF GRAPEVINE PROTOPLASTS 2.1 Introduction To date, GRSPaV is known to infect only Vitis species, and attempts to infect herbaceous hosts have all failed. While little is known about the replication of woody plant viruses in herbaceous plants, these failures could represent a need of GRSPaV for a Vitis specific host factor. GRSPaV may have co-evolved with Vitis for such a long period of time that it is now impossible for the virus to replicate outside its natural host. Regardless of the GRSPaV host range, the experiments reported here would better represent the wildtype behaviour of the replicase if transient expression could be performed in grapevine protoplasts as opposed to tobacco BY-2 cells. Consequently, initial attempts were done using grapevine material. Vitis rupestris cells in suspension culture and as callus tissue were generously donated by Dr. Jay Subramanian, Plant Agriculture Department, Vineland Station. As the liquid culture was found to contain approximately 2.5X105 cells/mL, four fold less than required for one BY-2 transfection, only a single transfection was attempted. Since the liquid culture and callus tissue samples were prepared differently, they will be discussed separately. 2.2 Generation of Grapevine Protoplasts from Suspension Cell Culture The 50 mL V. rupestris suspension cell culture arrived in a 125 mL tissue culture flask, and was cultured upon arrival in the dark at 25°C with shaking (150 RPM). Since an adequate grapevine protoplast isolation protocol could not be found in the literature, the tobacco BY-2 protoplast protocol (see section 2.8) was applied to the grapevine cells 141 with the following modifications: The total 50 mL culture was poured into a 50 mL conical tube and the cells were pelleted via centrifugation at 50 X g for 10 minutes, yielding a packed-cell-volume of approximately 5 mL. However, it should be noted that the majority of the packed-cell-volume was composed of unwanted plant debris. Digestion was performed using the BY-2 digestion buffer, however incubation was at room temperature for 3 hours. Room temperature incubation appeared to have no effect on the grapevine cells, so further incubation at 37°C for 2 hours was performed, resulting in 30% of the cells being converted to protoplasts. The remaining 70% of the cells remained clumped and, in that state, appeared recalcitrant to digestion even after two more hours of incubation. Digestion was therefore halted, and the protoplasts were washed the same as BY-2 protoplasts and finally re-suspended in 1 mL of electroporation buffer. pRTL2:GFP, which expresses GFP alone, was used to test transfection of the grapevine protoplasts. At this point, the grapevine protoplasts were individualized and free-floating among large clumps of unreleased grapevine cells and plant debris. The protoplasts were incubated on ice for 15 minutes, then mixed with 30 ug of pRTL2:GFP in a 4 mm electroporation cuvette (Bio-Rad). Electroporation parameters included 1 pulse at 200 V, 150 uF, and 100 Q. After electroporation, the BY-2 protocol was followed and the grapevine protoplasts were re-suspended in 1 mL of protoplast culture media and incubated at 25°C in the dark for 18 hours. Upon epifluorescence microscopy, only 10 transformants were observed in the entire sample, too few to be practical. However, despite the extremely low transfection efficiency, grapevine protoplasts were indeed generated and successfully transfected. Further modification of the above protocol may, in the future, increase the transfection efficiency so that grapevine 142 protoplasts may be used routinely for GRSPaV based experiments. Some future suggestions to this effect include starting with a higher quality culture by filtering out the unwanted plant debris. Alternately, the plant debris and non-digested cell aggregates could be filtered out of the culture after protoplast isolation but prior to electroporation. If multiple cultures could also be combined to get the optimal 1.0X10 cells/mL, transfection efficiency would likely be greatly enhanced. 2.3 Generation of Grapevine Protoplasts from Callus Tissue The grapevine callus tissue samples were cultured on solid media sealed in a sterile Petri dish. A total of three callus "colonies" were formed, each measuring approximately 3 cm X 3 cm X 2 cm in length, width, and height respectively. To isolate protoplasts, all of the callus tissue was finely chopped with a sterile razor blade and mixed, in a 50 mL conical tube, with 20 mL of protoplast digestion buffer. Digestion was for 2 hours and 20 minutes in the dark at 37°C on a nutating mixer. The cells were then pelleted by centrifugation at 150 X g for 8 minutes, resulting in a 15 mL packed- cell-volume. As with the suspension cell culture, the vast majority of the pellet was composed of unwanted plant cell debris. Since the cell pellet was so large (-15 mL), 10 mL of electroporation buffer was used to re-suspend the protoplasts. Following the B Y-2 protocol, 1 mL of the grapevine protoplasts was then mixed with 30 ug of DNA, and electroporation was performed the same as for the suspension cells. The protoplasts were washed and re-suspended in 1 mL of protoplast culture media. After culturing for 18 hours, no transformants could be detected in the sample, as the plant cell debris dominated the preparations and very few healthy cells were found. Clearly another 143 method must be employed for the purpose of transfecting grapevine callus tissue. Callus tissue, which can survive for long durations before use, may be useful for long term storage of BY-2 cells. In conclusion, the greatest success was obtained using grapevine suspension cells, and so it is recommended that future experiments use this type of culture and focus first on removing the large quantity of plant cell debris which greatly hindered the above experiments. 144 APPENDIX III: IMMUNOSTAINING OF TOBACCO BY-2 PROTOPLASTS 3.1 Introduction Attempts were made to observe the sub-cellular location of the RdRp domain when expressed without a GFP tag (pPOL) via fluorescent immunostaining using an anti- RdRp antibody. While several different methods were tested, all procedures required the protoplasts to be fixed, permeabilized, and finally immunostained. These three steps will be discussed separately with a focus on common problems, what was performed to overcome the problems, and ultimately what method produced the most achievable results, though it should be noted that immunostaining of the quality required to clearly distinguish and localize the RdRp was never achieved. 3.2 Fixation of Tobacco BY-2 Protoplasts Prior to fixation, protoplasts were generated, transfected, and cultured as usual. The following protocol employed was based on Prod'homme et al (2001), and involved fixing the protoplasts on poly-L-lysine coated coverslips. To prepare coverslips for adherence of protoplast cells, a stock solution of poly-L-lysine was prepared by dissolving 100 mg of poly-L-lysine hydrobromide (Sigma-Aldrich) in 100 mL of dFL:0. The solution was filter sterilized through a 0.22 urn filter (Fisher Scientific) and stored in 5 mL aliquots at -20°C. A working solution of poly-L-lysine (50 ug/mL) was prepared by making a 1/20 dilution of the original stock with sterile dt^O. Coverslips, previously sterilized in 95% ethanol and air dried, were submerged in a sterile Petri dish containing the working solution and incubated at 37°C for 1 hour. The coverslips were then 145 removed using sterile forceps and allowed to air dry on an angle. The protoplasts were adhered to the poly-L-lysine coated coverslips by placing several drops of cells on the surface of the coverslip and allowing them to settle for 20 minutes at room temperature. To fix the protoplasts, the coverslips were submerged in cold (4°C) 95% ethanol for 20 minutes. The cells were then washed and immunostained. The main issue with this method of fixation was that the protoplasts shrunk into a compact mass due to the ethanol treatment, rendering them useless for microscopy. Another fixation solution was tested which consisted of 1% glutaraldehyde (Sigma- Aldrich) in BY-2 media. While this solution did not cause the protoplasts to shrink, it evidently took too long to kill the cells as they had very high levels of autofluorescence on all channels during microscopy, which masked any potential expression patterns. If the protoplasts were not killed quickly, the cells may have had time to mount a stress response, resulting in the production of autofluorescent compounds such as phytoalexins (Pierce and Essenberg, 1987). Acceptable fixation, resulting in normal cell morphologies with minimal background autofluorescence, was eventually achieved using two separate solutions. A fixative consisting of 6% paraformaldehyde (Sigma-Aldrich) and 10 mM ethtlene glycol tetra-acetic acid (EGTA) (Sigma-Aldrich) in IX PBS was found to work sufficiently if the protoplasts were fixed prior to adherence to a coverslip. In this protocol, 450 uL of protoplasts was mixed with 450 uL of fixative and incubated at room temperature for 30 minutes. The cells were then centrifuged at 800 RPM in an Eppendorf J415D centrifuge at room temperature for 3 minutes. Four hundred and fifty microlitres of the supernatant was then removed and the pellet was re-suspended via gentle inversion. The re-suspended protoplasts were then adhered to the coverslip as mentioned 146 previously. Another fixative solution that produced cells sufficient for microscopy was based on Banjoko and Trelease (1995) and consisted of 4% formaldehyde (Sigma- Aldrich) in 0.5X protoplast culture media. This fixative solution also produced protoplasts which were acceptable for fluorescent microscopy. However, none of the above fixative solutions resulted in protoplasts that were similar in appearance to living cells. The sensitive nature of protoplasts may preclude this possibility. 3.3 Permeabilization of Tobacco BY-2 Protoplasts In order to allow for the efficient entry of antibodies into the protoplast cytoplasm, the cells had to first be permeabilized to create small pores through which the antibodies could pass. Since immunostaining was never successful, it was not possible to assess the degree to which permeabilization affected the desired final outcome. However, experiments were performed in which the entire immunostaining protocol was applied to both permeabilized and non-permeabilized protoplasts. In some cases, secondary antibody fluorescence was observed, however fluorescence was present throughout the cytoplasm in all cells, including negative controls, suggesting non-specific binding of an antibody was occurring. These cells were useful however, in demonstrating that protoplasts that were permeabilized showed a much higher level of immunofluorescence than their non-permeabilized counterparts, suggesting permeabilization did have a significant positive effect on the entry of the antibodies into the cells. Permeabilization was performed using two different methods. The first method simply involved submerging a protoplast coated coverslip in cold (-20°C) methanol for 10 minutes. The second method was based on Banjoko and Trelease (1995) and involved 147 permeabilizing the cells in 10% Triton-X 100 (Sigma-Aldrich) in IX PBS by mixing via gentle inversion at room temperature for 15 minutes in a 1.5 mL microfuge tube. Since both methods appeared to yield similar results, the latter Triton-X 100 method was preferentially used as it allowed for "tube based" immunostaining (see following section), negating the use of poly-L-lysine coated coverslips. 3.4 Immunostaining of Tobacco BY-2 Protoplasts As mentioned above, protoplasts were initially fixed, permeabilized, and immunostained on poly-L-lysine coated coverslips. The advantage of using coverslips as opposed to 1.5 mL microfuge tubes was that very small volumes of primary and secondary antibodies could be used during immunostaining. By placing a single drop (-20 uL) of an antibody dilution onto a piece of parafilm, the coverslip needed only to be lowered carefully, cell side down, onto the drop. For RdRp detection, the primary antibody used was a polyclonal rabbit anti-RdRp antibody (see section 2.10.2). The secondary antibody was Alexafluor 488 or 555 (Invitrogen), which were conjugated to green and red fluorescent fiuorophores respectively. Despite the advantage of being able to use small volumes of antibody, the coverslip method proved unreliable as the multiple washes applied throughout the entire immunostaining procedure often washed the majority the protoplasts off the coverslip. Since protoplast transfection produces relatively few transformants to begin with, the loss of 95% of the cells using the coverslip method was not acceptable. A "tube based" method, which involved fixing, permeabilizing, and immunostaining the protoplasts in 15 mL conical tubes and 1.5 mL microfuge tubes, was tested. 148 Centrifugation followed by the careful decanting of the supernatant was used between treatments and washes. Centrifugation speeds and times were optimized to pellet, but not destroy, the protoplasts. A speed of 1000 X g for 8 minutes was found to provide the most optimal conditions. The "tube based" method had the advantage of retaining almost all of the cells throughout the procedure, however larger volumes of each antibody dilution was required (approximately lmL compared to 20 uL with the coverslip method). Unfortunately, even with high cell retention, the exact conditions for successful immunostaining were never elucidated. Non-specific binding of the primary and/or secondary antibody continued to be a problem throughout the course of the aforementioned experiments. 3.5 Conclusions and Preliminary Protocol The successful immunostaining of pPOL was never accomplished, and the use of fluorescent protein tags was eventually employed so as to hasten the pace of research. However, most of a fluorescent immunostaining protocol was experimentally optimized for use on tobacco BY-2 protoplasts. Since this protocol may be useful in future experiments, it is outlined in detail as follows: 1. Isolate and transfect tobacco BY-2 protoplasts as usual. Allow to incubate for the appropriate time post-transfection. 2. Prepare fresh fixative solution (2.5 mL/sample) (4% formaldehyde in 0.5X protoplast culture media). 3. Pour the protoplasts into a 15 mL conical tube and mix with 2.5 mL fixative solution. Incubate at room temperature for 15 minutes on a rocker or nutating 149 mixer. 4. Centrifuge at 1000 X g for 8 minutes to pellet the fixed protoplasts. NOTE: Fixed protoplasts are significantly stronger than live protoplasts, however they can be warped with prolonged or too fast centrifugation. This speed and time have worked in the past but may need to be further optimized. 5. Wash the protoplasts four times with 1.5 mL of IX PBS, mixing for 1.5 minutes between washes after the initial wash. Remove the supernatant with a pulled Pasteur pipette. 6. After the final wash, re-suspend the protoplasts in 1 mL of IX PBS and pour into a 1.5 mL microfuge tube. NOTE: Protoplasts can be stored at 4°C at this point. 7. Fix by adding 33 uL of fixative solution and mix via rocking at room temperature for 15 minutes. 8. Wash four times as above, end with pellet. 9. Dilute primary antibody to desired concentration in IX PBS and use to re-suspend above protoplast pellet. Incubate at room temperature for 1 hour on nutating mixer. NOTE: An anti-RdRp diltution between 1:10,000 and 1:50,000 was found to minimize non-specific binding, however the optimal dilution for RdRp detection was not discovered. 10. Wash the protoplasts as above for a total of three washes. End with pellet. 11. Dilute secondary antibody to the desired concentration in IX PBS and use to re- 150 suspend above protoplast pellet. Incubate on nutating mixer in the dark at room temperature for 1 hour. NOTE: Using Alexafluor 488 or 555 (Invitrogen), a 1:1000 dilution was used for all experiments while the concentration of the primary antibody was adjusted. 12. Wash three times as above, and re-suspend protoplasts to the desired concentration in 1 X PBS. Mount protoplasts on slide for microscopy as usual. 151