Patent Application
20100017911
Virus-based transient-expression vectors are routine tools used in plant molecular biology laboratories throughout the world for rapidly expressing or silencing genes in plants. They also can be important tools in plant genomics to screen unknown sequences for function. Yet, available vectors have been developed from a limited number of rather similar viruses of herbaceous plants. Notable examples are the vectors based on Tobacco mosaic virus (TMV) (Dawson et al., 1989; Donson et al., 1991; Shivprasad et al., 1999; Rabindran and Dawson, 2001). Tree crops offer special challenges. Even if existing vectors could infect trees, the time required for systemic infection and analysis of the expressed genes in trees generally exceeds the stability of known virus-based vectors. Yet, the challenges of breeding restraints and the decades required for improving trees greatly increase the need for useful virus-based vectors.
Citrus tristeza virus (CTV) is a member of the complex Closteroviridae family that contains viruses with mono- , bi-, and tripartite genomes transmitted by a range of insect vectors including aphids, whiteflies, and mealybugs (Bar-Joseph et al., 1979; Dolja et al., 1994; Agranovsky, 1996; Karasev, 2000). The long flexuous virions (2000 nm×10-12 nm) of CTV are encapsidated by two coat proteins: the major coat protein (CP) covering about 97% of the virion and the minor coat protein (CPm) completing encapsidation of the other terminus. The single-stranded RNA genome of CTV is approximately 19.3 kb, divided into twelve open reading frames (ORFs) (Pappu et al., 1994; Karasev et al., 1995) (
The major lesson that has been learned so far from virus-based vector design is that building an effective vector requires understanding of the regulation of viral gene expression (Shivprasad et al., 1999). It is fairly easy to insert a reporter gene into your favorite virus and monitor expression in protoplasts, or for a limited time in portions of an herbaceous plant. It is much more difficult to create a vector that both expresses the inserted gene at a sufficient level and is stable long enough to be useful.
The present invention is based on the inventors′ work regarding several strategies for utilizing CTV as a basis for constructing a transient-expression vector for citrus trees. According to one embodiment, the invention relates to a method of transfecting a woody tree with a gene of interest. In a specific embodiment, the method enables the transfection of woody trees under field conditions to express beneficial proteins. The method includes the step of innoculating a woody tree with a CTV vector engineered to comprise said gene of interest. The gene of interest has a 5′ end and a 3′ end. Positioned upstream of the 5′ end is an additional sgRNA control element (CE). Innoculation of the tree results in a systemic infection of the woody tree such that a majority of the virus population of the systemic infection comprises the CTV vector. Also, the CTV vector is stable enough to enable expression of the gene of interest for at least two months. Inoculation may occur by convention methods in the art. In one embodiment, inoculation occurs by grafting a portion of tissue infected with CTV vector to another tree.
The inventors have recognized that it is desired to sufficiently form virions such that the inoculum can be amplified by protoplast passage for inoculation of trees. Because of limitations of working with a viruses in woody plants, the requirements for a transient-expression vector are much more rigorous than the short term experiments possible with herbaceous plants. Not only should the levels of expression of the foreign gene be sufficiently high, but the resulting construct must be stable enough for systemic infection of trees and analysis of the foreign gene's effect on the tree. The time required is months to years. The inventors describe herein the results of different types of virus-based transient-expression vectors using CTV, and a demonstration of an ‘add-a-gene’ construct that is unusually stable. The vectors of the present invention are capable of producing copious quantities of a foreign protein for several years. The vectors are remarkably useful at equipping trees in the field with the ability to protect themselves or cure themselves of diseases.
The CTV produces ten 3′-coterminal sgRNAs that function as mRNAs for the ten 3′ genes (Hilf et al., 1995), with only the 5′-most ORF of each thought to be translated. There appear to be general rules that determine the levels of production of these sgRNAs. First, genes located nearer the 3′ terminus are usually expressed at higher levels. The two 3′ most genes, p23 and p20, have the highest levels of sgRNAs. Also, when genes are moved closer to the 3′ terminus, expression levels increase. Positioning the low expressed p33 gene near the 3′ terminus results in a level of expression comparable to the highest expressed genes (p20 and p23) (Satyanarayana et al., 1999). However, there are exceptions. The CP gene, located at position 5 from the 3′ end, is expressed higher than the p13 and p18 genes, located at positions 3 and 4. The second rule is that ORFs with an upstream nontranslated region are generally expressed higher than ORFs that overlap with the preceding ORF. The p23, p20, p13, CP, p6, and p33 ORFs have upstream nontranslated sequences (Pappu et al., 1994). With the exception of the 5′-most genes (p33 and p6), these are the more highly expressed genes.
The cis-acting sequences that regulate the expression of the 3′ genes generally are located immediately upstream of their ORFs. We refer to these sequences as ‘controller elements’ (CE) instead of ‘promoters’ because we have not been able to determine whether the mode of production of the 3′ sg mRNAs is by promotion from an internal sequence of the minus strand or termination during minus strand synthesis followed by amplification of positive-stranded sg mRNAs (Gowda et al., 2001). Each 3′ CE produces two sgRNAs in addition to the expected 3′-terminal positive-stranded MRNA: a negative-stranded sgRNA with sequence complementary to the sg mRNA, and a positive-stranded 5′-terminal sgRNA that apparently terminates prematurely near the CE during genomic RNA synthesis. The 3′ CEs generally consist of one or two stem-loop structures with a downstream (plus sense) +1 site corresponding to the 5′ terminal adenosine of the mRNA (Gowda et al., 2001; Ayllón et al., 2005). Mutational analysis and characterization of the context of the +1 site of the sg mRNAs showed opportunities for manipulation of levels of expression several fold up or down (Ayllón et al., 2003).
In a specific embodiment, the invention pertains to a CTV vector engineered to comprise and express gene of interest, said gene of interest having a 5′ end and a 3′ end; and an additional sgRNA CE positioned upstream of the 5′ end of said gene of interest. In a more specific embodiment, the heterologous sgRNA CE is BYV CP sgRNA CE. In an alternative embodiment, the sgRNA is an added homologous CTV CPsgRNA CE.
In an alternative method embodiment, the invention pertains to a method of producing a tree expressing a gene of interest. The method includes transfecting a first tree with a CTV vector engineered to comprise said gene of interest, said gene of interest having a 5′ end and a 3′ end; and a heterologous sgRNA CE positioned upstream of the 5′ end of said gene of interest. A portion of the first tree is then grafted to a portion of a second tree.
The earliest prototype plant virus-based vectors were gene-substitution vectors, in which a viral ORF not necessary for replication was substituted by a foreign gene ORF. However, since the coat protein gene was the expendable gene, this version of vector failed to move or only moved poorly in intact plants (French et al., 1986; Takamatsu et al., 1987; Dawson et al., 1988; Chapman et al., 1992; Scholthof, 1999). In contrast, CTV has three genes that can be deleted with continued efficient replication and movement of the virus in citrus trees—p33, p18, and p13 (T. Satyanarayana, unpublished data). Although the p33 and p18 genes are expressed at relatively low levels, the p13 gene is expressed at moderate levels. Thus, the inventors chose to replace most of p13 ORF with the ORF of GFP, resulting in vector construct CTV-dp13/GFP (
The in vitro generated RNA transcripts of pCTV-dp13/GFP were used to inoculate Nicotiana benthamiana protoplasts. CTV-dp13/GFP replicated well, producing normal amounts of genomic and sgRNAs as demonstrated by Northern blot hybridization analysis (
The ability of this vector prototype to form virions was assessed via passaging of progeny virions that were extracted from the transcript-inoculated protoplasts to a next set of protoplasts (Satyanarayana et al., 2001). CTV-dp13/GFP was passaged efficiently through a sequential series of protoplasts, with the number of protoplasts infected and the yield of virus amplified to high levels, demonstrating the facility to form good virions (
The inventors previously examined the production of some CTV proteins fused to GFP during replication in protoplasts. A viral construct with a fusion of GFP to the C-terminus of p20 produced copious amounts of the fusion protein that fluoresced brightly and accumulated in the amorphous inclusion bodies, which represent a characteristic feature of the CTV infection (Gowda et al., 2000). Similarly, a construct with GFP fused to the C-terminus of p23 produced large amounts of the fluorescing fusion protein (T. Satyanarayana, unpublished data). Both constructs replicated normally, with little effect on regulation of the other genes, and could be passaged efficiently from protoplast to protoplast. Yet, neither construct infected plants, apparently because the fused viral protein was not functional in plants. Here it was chosen to produce similar constructs, but to ‘cleave’ the fusion protein and provide a functional viral protein. Because CP is one of the highest expressed proteins, we chose to examine fusions to CP. One approach is to use the 2A peptide of the Foot-and-mouth disease virus (FMDV) that mediates processing of the FMDV polyprotein by disrupting translation, which results in production of two polypeptides (Ryan et al., 1991; Donnelly et al., 2001b). The ‘cleavage’ occurs between the carboxy-terminal glycine residue of the 2A peptide and the amino-terminal proline residue of the 2B protein of FMDV. Insertion of the FMDV 2A region followed by a proline residue in a synthetic polyprotein has been previously shown to mediate a ‘cleavage’ of the polyprotein with an efficiency estimated at ˜85% (Ryan and Drew, 1994). N-terminal extension of the 2A region by 14 amino acid residues from the C-terminus of the FMDV 1D protein located immediately upstream of 2A improves the ‘cleavage’ activity to ˜99% (Donnelly et al., 2001a). Thus, in CTV-fCP/GFP a sequence encoding the polypeptide comprising 14 amino acid residues of the ID region followed by 18 amino acid residues of the 2A protein plus an additional proline codon were fused between the last codon of the GFP ORF and the first codon of the CP ORF. The expression of the fused ORF was directed by the CP sgRNA CE (
Upon inoculation of protoplasts with in vitro synthesized transcripts of pCTV-fCP/GFP, a larger sgRNA corresponding to the expected size of the GFP-ID-2A-CP mRNA was produced, which did not affect the levels of accumulation of the other sgRNAs (
‘add-a-gene’ vectors based on TMV that produced an extra sgRNA and a foreign protein have been previously produced (Dawson et al., 1989; Donson et al., 1991; Shivprasad et al., 1999). With CTV, mini-replicon constructs were created with the GFP ORF expressed from a sg mRNA and examined its production in protoplasts, demonstrating the feasibility of expressing foreign ORFs with CTV CEs (Satyanarayana et al., 2003). However, these CTV constructs were missing most of the 3′ genes and could not infect trees. Since an objective of this work was to construct a vector that would express foreign genes in citrus trees, constructs containing all of the genes required for passage and infection of plants were needed. A first question was, where to position an extra gene for stable and high-level expression? The inventors chose to insert the foreign ORF between the CP and CPm genes because the CP sgRNA CE was defined better than other CEs (Gowda et al., 2001; Ayllón et al., 2004). A next question was, what to use as a CE for the foreign ORF? With TMV, insertion of homologous sequence repeats caused vector constructs to be unstable (Dawson et al., 1989), but heterologous repeats using a promoter sequence from a related virus were relatively stable (Donson et al., 1991). Based on the TMV results, Peremyslov et al. (1999) created a vector from the closterovirus Beet yellows virus (BYV) using an extra heterologous CE from Beet yellow stunt virus.
A set of three ‘add-a-gene’ constructs with an additional CE were examined (
Inoculation of protoplasts with transcripts derived from cDNAs of each of these constructs resulted in efficient replication of the vectors (
The ability of these vectors to be passaged in protoplasts was examined as described above.
Similar to the parental wild type virus (
The BYV sgRNA CE has not been characterized. Thus, an arbitrary length of sequence upstream of the BYV CP ORF for CTV-BC1/GFP was chosen. To examine the effect of different lengths of this upstream sequence, a series of six vector constructs were built based on CTV-BC1/GFP with progressively shorter fragments inserted in front of the GFP ORF (Table 1). These constructs were analyzed for levels of accumulation of the GFP sg mRNA (
Vectors with the highest levels of GFP expression, CTV-BC1/GFP, CTV-BC5/GFP, and CTV-CC/GFP, were examined during a series of eleven sequential passages in protoplasts (which represented 44 days of replication) for their stability. The patterns of sgRNAs remained unchanged during these passages for all three vector constructs (
After the above series of passages in protoplasts, progeny virions were concentrated via sucrose cushion centrifugation and used for inoculation of citrus trees (Robertson et al., 2005). At five weeks after inoculation, GFP-expressing vectors were found to have replicated and moved systemically in the young trees as demonstrated by ELISA using CTV-specific antiserum (data not presented) and by observation of GFP fluorescence (
GFP fluorescence was detected in phloem-associated cells of young, developing parts of the trees: in the bark of stems, in young leaves and petioles, and in young roots of infected seedlings (
The stability of the inserted GFP gene in CTV-BC1/GFP, CTV-BC5/GFP, and CTV-CC/GFP in citrus trees was examined at different times after inoculation by analyzing the pattern of sgRNAs and by monitoring GFP expression in the growing areas of the trees. Since the infected cells in the tree are at different stages of replication, it is difficult to get good sgRNA profiles from RNA extracted directly from the trees. To avoid this complication, virions were extracted from growing parts of the trees and used for inoculum for protoplasts. Total RNA isolated from protoplasts 4 days after inoculation was analyzed by Northern blot hybridizations and compared to the corresponding RNA samples isolated earlier during the protoplast passaging of the vectors. At 6 weeks after inoculation of the trees, no changes in sgRNAs patterns were detected and GFP fluorescence in the trees was uniform for all of the vectors (data not shown), demonstrating that the insertions remained in all vectors.
The GFP fluorescence was further monitored in all infected trees for several months. At one year after inoculation of trees, the GFP fluorescence in plants infected with CTV-CC/GFP and CTV-BC1/GFP was still visible, but the number of fluorescent cells was reduced in some trees. The progeny virions analyzed by amplification in protoplasts showed patterns of sgRNAs that demonstrated loss of the GFP insert for both vectors in each set of three trees that were examined in this experiment. The patterns represented mixed populations of intact vector and virus in which the extra GFP sgRNA was partially or completely lost, as indicated by the downward shifts of the p61 and CPm sgRNAs (
To examine the stability of CTV-BC5/GFP further, one of the plants in which the vector remained stable for two years was used as inoculum for ten new citrus trees. Analysis of the virus population six months after infection revealed that only one plant out of ten contained virus with the insertion deleted as a minor component of population, which was noted by an appearance of an additional band corresponding to the wild type p61 sgRNA on Northern hybridization blots (
To examine the competitiveness of the original vector, CTV-BC5/GFP, with the recombinants, the virus mixture isolated from the plant with the mixed population (
Thus, the vector appeared to be competitive with the recombinant wild-type-like virus. To examine whether these recombinant viruses were less competitive than the wild type virus for some unknown reason, we examined the competitiveness of the vector with the wild type virus in another experiment in which five citrus trees were simultaneously inoculated with two viruses—wild type CTV and CTV-BC5/GFP. After two months and four months of infection, the virus populations in the flush of new growth at the top of the trees were analyzed. The resulting virus in the new growth was a mixture of both viruses (
The ‘add-a-gene’ vectors worked best of the strategies that that were examined. Insertion of the foreign gene between the two coat proteins genes was examined. Of the constructs examined, the choice and place of CEs greatly affected their effectiveness. The heterologous CE from BYV when used to control CP expression in CTV-CB/GFP resulted in too little CP for efficient passage for amplification of inoculum to infect trees. The vector using the heterologous BYV sgRNA CE to control GFP expression produced slightly less GFP MRNA, but produced sufficient amounts of CP from the native CE to allow efficient passage and amplification. Remarkably, the size of the heterologous BYV sgRNA CE substantially affected the levels of the GFP mRNA. There was no clear relationship between size and strength of the CE. As the inserted sequences were shortened, the levels of mRNA decreased, increased, and decreased again, suggesting that structural interactions that affect the function of the CE were altered upon changing the lengths of sequences. The differences were great enough to justify empirically testing different lengths of sequence for optimal expression.
From observations of several different CTV constructs, it appears that the level of CP produced cannot be reduced much below the wild type level without reducing the yield of virions.
Previous work that examined expression of an extra gene in TMV clearly demonstrated that there was competition between the different sgRNAs—increases in one sgRNA resulted in decreases in others (Shivprasad et al., 1999). Additionally, manipulation of the sgRNA promoters within wild type TMV similarly changed the ratio of the 30K and CP sgRNAs (Grdzelishvili et al., 2000). In designing a TMV-based vector, production of the sgRNA for the extra (inserted) gene occurred at the expense of the sgRNAs of the native genes (30K and CP). Thus, with TMV, the optimal vector resulted from a compromise between reduced levels of CP and movement protein that were still sufficient for virion assembly and movement and a corresponding increase in production of the foreign protein (Shivprasad et al., 1999). Similar competition in production of viral sgRNAs has been shown with Barley stripe mosaic virus (Johnson et al., 2003) and Sindbis virus (Raju and Huang, 1991). In contrast, there appears to be no competition in production of the different sg mRNAs by CTV. Inserting a new gene or increasing or decreasing the levels of expression of different genes had little or no effect on the levels of sg mRNA production by the other genes (Ayllón et al., 2003).
CTV appears to differ from alpha-like RNA viruses in production of sg mRNAs. It is not yet certain whether closteroviruses produce sg mRNAs by initiation from a promoter on the negative strand like alpha-like RNA viruses, or by terminating negative strand synthesis at the CE and using that minus-stranded sgRNA as a template for the mRNAs (Gowda et al., 2001). However, the production of sg mRNAs by CTV clearly differed from that of alpha-like RNA viruses in that the ‘+1 nt’ tolerated considerable modification (Ayllón et al., 2003; 2004). Another unusual characteristic of CTV is that the p23 gene controls the ratio of plus to minus strands of the sgRNAs (Satyanarayana et al., 2002). The lack of competition of sgRNAs is another line of evidence suggesting that the mechanisms of production of sg mRNAs by closteroviruses differ from those of alpha-like RNA viruses.
A major advantage of virus-based vectors is a high level of expression, and, thus, high-titer viruses like TMV and Potato virus X (PVX) (Chapman et al., 1992) were utilized first. Although CTV is more limited in tissue tropism than these viruses, the levels of foreign gene production per infected protoplast by the CTV-based vectors were similar to that of TMV. A major limitation of virus-based vectors has been their lack of stability. For example, the vectors that are used most widely as laboratory tools are those based on TMV and PVX, but their progeny populations are overcome by recombinants that have lost most of the inserted sequences during the systemic infection of their hosts within a couple of weeks. In contrast, CTV-based vectors have remained the major component of the population for as long as five years. Without being bound to any theory, the inventors expect that eventually the vector will be overtaken by recombinants. Even though the vector can compete with the wild type virus for years, it is likely that the wild type virus will have a competitive advantage in the long term. It should be noted that the construct CTV-BC1/GFP did not appear as stable as CTV-BC5/GFP. Other CTV-BC1 constructs with different foreign genes that appear to be as stable as CTV-BC5/GFP. Thus, other factors such as errors incorporated during cloning and propagation of the cDNA or founder effects during the movement and systemic infection of trees might determine which vectors last longer.
Instability generally has been blamed on high error rates of RNA virus replication and high rates of recombination. In fact, the error rate of RNA viruses at one time was thought to be so high that it was argued that it would be impossible to utilize RNA viruses as transient-expression vectors (Van Vloten-Doting et al., 1985). Yet, the sequence of CTV appears to be unusually stable. The inventors have discovered that the sequences of different isolates of CTV maintained in different countries and in different varieties of citrus for more than a hundred years were essentially identical (Albiach-Marti et al., 2000). However, recombination of CTV appears not to be limited. Most wild populations contain multiple defective RNAs that apparently resulted from facile recombination (Mawassi et al., 1995). For TMV-based vectors, it was shown that repeated sequences decreased stability, apparently by increasing recombination rates (Dawson et al., 1989; Donson et al., 1991). This instability could be partially overcome by using promoters from different tobamoviruses instead of repeated promoters (Donson et al., 1991; Rabindran and Dawson, 2001). However, repeated sequences in the CTV-based vector resulting from duplicated CP sgRNA CEs controlling the GFP and CP genes did not appreciably decrease stability. The stability of the CTV-based vectors appears not to be due to reduced recombination, but to increased competitiveness with the potential wild-type-like recombinant. When inoculated simultaneously with the wild type virus, the CTV-based vectors with an extra gene were able to compete effectively with the wild type virus during replication and movement throughout citrus trees. In contrast, the TMV-based vectors compete poorly with the wild-type-like recombinants for both cell-to-cell and long-distance movement (Rabindran and Dawson, 2001). Thus, the TMV recombinants quickly overcome the vector during spread in the plant. Even though recombinants with the inserted sequences deleted are produced in the CTV populations, there appears to be little selection for these recombinants to be increased proportionally in the populations.
CTV is a large virus with ten 3′ genes expressed through sg mRNAs. The addition of an extra gene had no obvious effect on the virus. In the experiments described herein, the vector constructs appeared to be competitive with the wild type virus.
A limitation of CTV is that it is generally limited to phloem-associated cells. Thus, the vectors described here may not be the best for expression of genes in other tissues. However, the inventors have found that this potential limitation can be reduced somewhat by producing proteins with secretion signal peptides to export the protein out of the cell into the intercellular space where the protein or peptide is dispersed in the liquid films between cells (unpublished data).
Especially in woody plants, the transient-expression vector is a valuable tool to complement stable transformation. The potyvirus, Plum pox virus, recently was used to express GFP to examine movement of the virus in susceptible and resistant stone fruit trees (Lansac et al., 2005; Ion-Nagy et al., 2006). Transformation of citrus trees can take a year to produce a 2-inch tall plant and 5 to 20 years until the effect of the transgene on mature tree characteristics can be examined. The value of the virus-based vector is that once it infects a tree, it can be easily and quickly graft-inoculated to unlimited numbers of other susceptible trees of different varieties or species or different ages, including mature trees, and the virus-based vector can express new genes in trees or remove existing gene functions by RNA silencing relatively quickly. Although CTV has expressed GFP stably for more than five years, we do not advocate using the vector for expression in the field. Instead, the vector should be used to quickly identify genes that could improve trees followed by permanent expression through genetic transformation.
The full-length cDNA clone of CTV T36, pCTV9R (Satyanarayana et al., 1999; 2003), was the basis of all constructs in this study. pCTV9R was digested with SpeI and XmaI restriction endonucleases and a resulting 7.6 kb fragment (nts 11659-19293 plus the NotI site) encompassing ORFs p6 to p23 plus the 3′ nontranslated region was ligated into pGEM7Z+ (Promega) between XbaI and XmaI restrictions sites to generate pGEM-3′CTV. The wild type GFP ORF (Prasher et al., 1992) was inserted into pGEM-3′CTV by overlap-extension PCR (Higuchi et al., 1988) and subdloning (see below). The full-length CTV vector constructs were produced by ligation of PmeI-NotI fragments of GFP-harboring pGEM-3′CTV derivatives into similarly digested pCTV9R.
To engineer pGEM-dp13/GFP, the sequence encoding the first 112 amino acids of the p13 ORF was replaced by the GFP ORF. The resulting ORF began with the original p13 translation start codon and encoded the complete GFP fused to seven C-terminal amino acids (SLLPCDN) of the p13 ORF. This design preserved the p20 CE that overlaps the p13 ORF. To construct pGEM-fCP/GFP, the ORF of GFP and a sequence of FMDV 1D-2A plus an additional proline codon (encoding amino acids: GLEARHKQKIVAPVKQTLNFDLLKLAGDVESNPGP) were incorporated into pGEM-3′ CTV. pGEM-fCP/GFP possessed a modified ORF beginning with the original CP translation start codon and comprising the complete GFP ORF, the sequence encoding the 1D-2A peptide plus proline codon, followed by the CTV CP ORF.
The GFP ORF was inserted in the region between the CPm and CP genes to produce pGEM-CB/GFP, pGEM-BC1/GFP through pGEM-BC7/GFP, and pGEM-CC/GFP. Construct pGEM-CB/GFP contained an insertion of the GFP ORF followed by the sequence of CP sgRNA CE of BYV (nts 13499-13637) (Peremyslov et al., 1999) upstream of the translation start codon of the CTV CP ORF (between corresponding nts 16151 and 16152 in CTV genome). A set of constructs pGEM-BCl/GFP through pGEM-BC7/GFP contained insertions of the BYV CP sgRNA CE sequence of various lengths (Table 1), the restriction site for PacI, the GFP ORF, followed by the restriction site for XhoI downstream termination codon of the CPm ORF (between corresponding nts 16058 and 16059 in CTV sequence). The construct pGEM-BC1/GFP was used as a template for overlap-extension PCR to replace the sequence of BYV sgRNA CE with the CTV CP sgRNA CE (nts 16059-16151) to produce pGEM-CC/GFP.
PmeI—NotI fragments of pGEM-dp13/GFP, pGEM-fCP/GFP, pGEM-CB/GFP, pGEM-BC1/GFP through pGEM-BC7/GFP, and pGEM-CC/GFP were subdloned into pCTV9R digested with the same enzymes to produce full-length CTV constructs pCTV-dp13/GFP, pCTV-fCP/GFP, pCTV-CB/GFP, pCTV-BC1/GFP through pCTV-BC7/GFP, and pCTV-CC/GFP, respectively (
The procedures for the isolation of mesophyll protoplasts from N. benthamiana leaves and their transfection with SP6 RNA polymerase-derived transcripts of CTV cDNAs linearized with NotI were carried out as described by Satyanarayana et al. (1999). The expression of GFP fluorescence in infected protoplasts was observed at 4 dpi with a Zeiss Stemi SV 11 UV-fluorescence dissecting microscope (Carl Zeiss Jena, GmbH., Jena, Germany) and with a confocal scanning microscope Leica TCS SL (Leica Microsystems, Inc., Exton, Pa.). Protoplasts were harvested and divided into two portions: one used for total nucleic acids isolation, and the other stored at −70° C. for subsequent protoplast passage of virions. The total RNA isolated from protoplasts at 4 dpi were analyzed by Northern blot hybridization using a 3′ positive-stranded RNA-specific riboprobe (Satyanarayana et al., 1999). Passaging of virions in crude sap through up to eleven successive cycles in protoplasts for amplification of the virus and virion assembly assay were done as described previously by Satyanarayana et al. (2000).
Amplified progeny virions from the final passages in protoplasts were extracted and concentrated by sucrose cushion centrifugation, and the concentrated virions were used for mechanical inoculation of small C. macrophylla trees as described by Robertson et al. (2005). Double antibody sandwich indirect ELISA was performed as described previously (Garnsey and Cambra, 1991) to confirm infection in inoculated plants. Samples of leaves, stems, or roots were taken at different time points beginning at 2 weeks after inoculation of citrus trees. Transverse and longitudinal sections of plant material were prepared by hand with a razor blade. Unfixed specimens were mounted in water and GFP fluorescence was observed with the UV-fluorescence dissecting microscope with an attached camera Olympus Q-color 5 (Olympus America, Inc., Center Valley, Pa.).
Analysis of Virus Population Accumulated in Citrus Trees Infected with CTV-Based Vector Constructs
To examine virus populations in citrus trees, bark of young trees infected with the CTV-based vector constructs was peeled and ground with liquid nitrogen. CTV virions were extracted with 40 mM phosphate buffer, pH 7.4 added according to the ratio: 3 ml of buffer for 0.5 g of bark tissue. Extracts were clarified at 4000 g, and 100·1 of supernatant containing virions was used for inoculation of N. benthamiana mesophyll protoplasts as described by Satyanarayana et al. (1999). Total RNA isolated from protoplasts at 4 dpi was analyzed by Northern blot hybridization as described above. The genetic stability of GFP-expressing constructs was evaluated by the pattern of sgRNA accumulation.
The following list includes the full cites of references described above and additional related references. The disclosures of all references cited herein are incorporated in their entirety to the extent non inconsistent with the teachings herein.
Hammond-Kosack, K. E., Staskawicz, B. J., Jones, J. D. G., Baulcombe, D. C., 1995. Functional expression of a fungal avirulence gene from a modified Potato virus X genome. Mol. Plant Microbe Interact. 8, 181-185.
