Use of TRV2:prf recombinant vectors to induce silencing of traits in tomato plant disease resistance
Baylor S. Blewett1,*
1 College of Agricultural Science, University of Idaho, Moscow, ID 83843-0000, USA
Received: May 3rd, 2025 Revised: [in progress] Published: [in progress]
*For correspondence: (text [censored]; email blew0814@vandals.uidaho.edu)
Abstract
Silencing of the Prf gene expression in tomato plants via Virus Induced Gene Silencing (VIGS) notably decreases the plants’ ability to effectively resist the infection of Pseudomonas syringae pv. tomato. A Prf-silencing assembly recombinant with the TRV2 plasmid native to Escherichia coli transformed into an Agrobacterium strain was introduced into young tomato seedlings to observe how Prf knock-down affects the immunity of the plants. After infection with P. syringae, we observed a notable decline in overall health with plants integrated with the construct compared to their unmodified and PDS-combined counterparts. The unique reaction of Prf silencing with VIGS was also compared to similar gene expression factor EF1, which had no notable difference in gel results between the TRV2 blank and TRV2::Prf recombinant unlike the Prf targeted primers who near completely deactivated the expression of the Prf fragment within its construct. Without the ability to manufacture Prf proteins, the Pto effector cannot properly recognize or combat viral DNA integration into plant cells via inducing cell death.
Keywords: Prf gene, plant disease resistance, Pseudomonas syringae
Introduction
Plants, although typically inactive physically, host a myriad of intriguing defense mechanisms meant to protect the organism from the various harmful attacks its environment and ecosystem may inflict upon it. These active resistance traits are oftentimes chemical in nature involving complex pathways of input and exchange to drive biological responses such as generation of reactive oxygen and nitrogen species, changes in the concentration of ions across the plasma membrane and proteolysis which serves as the basis of gene expression (Ekengren et al., 2003). Of the many diseases and pests that affect plants, perhaps the most prolific are viruses, who are more enigmatic than most pathogens and have been a subject of interest for decades for botanists and agricultural scientists due to their potential for destruction and widespread diaspora. The earliest form of imbuing plant cross-breeding with virally resistant traits was by inserting viral genetic material into plants in a process known as pathogen-derived resistance (PDR) (Burch-Smith et al., 2004). Due to this method's pattern of providing resistance traits to newly grown tissues as opposed to the initially infected tissue, it was later determined that the molecular basis for PDR was post-transcriptional gene silencing (PTGS), a form of RNA interference that degrades targeted mRNA strands produced within an organism (Burch-Smith et al., 2004). Virus-induced gene silencing or VIGS was a development made on the discovery that viruses actively compete with plant pathogen defense mechanisms by creating suppressant proteins to combat PTGS. Researchers were then able to utilize VIGS via recombinant viruses armed with specified targeted sequences of RNA which could limit or “knock-down” the expression of targeted genes within plant hosts (Burch-Smith et al., 2004; Robertson 2004).
VIGS has a number of advantages over other rapid silencing procedures such as convenience of inoculation methods (agroinfiltration), its lack of requirement for transformation and its potential in operating with multi-copy genes (Purkayastha & Dasgupta, 2009). The primary application of VIGS is to observe gene influence from a “front-to-back” approach, silencing genes with the recombinant virus to record their phenotypic expression and cross-gene interactions without the need for transformation or selection, reducing the inoculation to expression time to 2-3 weeks (Robertson, 2004; Purkayastha & Dasgupta, 2009). An important instance of this process being demonstrated involved an enzyme referred to as Phytoene desaturase (PDS) which bio-synthesizes carotenoids, a compound that allows chlorophyll to resist photo-bleaching. A recombinant virus TMV transformed with a sequence encoding pds was inoculated into Nicotiana benthamiana plants which silenced PDS, triggering the leaves to grow pale white as a visual indicator of transformation (Purkayastha & Dasgupta, 2009). A particular grouping of organisms popular for studying plant-pathogen interaction and surveying their modification through VIGS are those of Pseudomonas syringae pv. tomato and the tomato. Within particular strains of tomato, a host resistance (R) gene Pto which encodes a protein kinase is present, this gene specifically provides defense against P. syringae pv. Tomato strains that express either AvrPto or AvrPtoB effector proteins (Ekengren et al., 2003). Prf is a protein that falls within interest along this resistance pathway, as it has similarity to the R proteins utilized in Pseudomonas resistance and is required for Pto-mediated resistance, however its role in the greater scheme is still relatively unknown (Ekengren et al., 2003; Burch-Smith et al., 2004). The primary objective of the experiment was to determine the function of the Prf gene in terms of resistance to Pseudomonas syringae pv. tomato (abbreviated Pst) in tomato plants. The observations from other studies indicates a likely hypothesis that resistance to the disease Pst in the resistant RG-PtoR (Rio Grande-PtoR) strain of tomato is regulated by the indirect recognition of the Pst effector by Prf. Thus VIGS could be utilized to repress expression of the Prf gene in RG-PtoR tomato plants and after transformation with a recombinant vector containing a gene-silencing construct, it can be observed if there was any alteration of the amount of resistance to Pst in the Prf-silenced plant.
Materials and Methods
Figure 1. A flowchart depicting the individual steps in the experimental design for surveying the effect of Prf silencing in tomato plants. Steps are organized as pre-ligation (green), pre-transformation of tomato plant (yellow), plant infection survey (orange) and Prf versus EF1 gene factor observations (red). Sequencing results were spaced between the plant infection survey and the completion of the Prf vs. EF1 observations are separately displayed as blue by order of completion.
The experiment was conducted over 3 months and was subdivided into several smaller tasks to better structure progress. The overall stages of experimentation will be described by categorization in the results section (Figure 1). The first section encompassed the isolation of the genome from the tomato plant and the plasmid vector of choice, the TRV2 from Escherichia coli; it also involved the creation of primers meant to separate out a specific fragment of DNA encoding the Prf gene. After purification on electrophoresis gel verification, the vector and gene of interest were ligated and then chemically transformed into E. coli to serve as a duplication host. After uptake and division, the recombinants were isolated from their bacterial hosts and sequenced in gel to ensure the ligation process was successful. The vectors were then re-transformed into Agrobacterium by electrophoresis before being injected into the leaves of tomato seedlings. After being allowed to grow, inserting the T-DNA region containing the Prf silencing construct from the bacteria into the genome of new plant tissue, a sample was collected and processed to be sent out for sequencing and plated in the lab to monitor bacterial growth. After 2 weeks, the plantlets were vacuum inoculated with infectious P. syringae to observe if the T-DNA was integrated into the plant’s genome. Phenotypic traits were recorded after 2 days and visibly infected tissue was harvested for both bacterial concentration calculations as well as for plant RNA isolation. After the amount of P. syringae present in the Prf construct and negative control tomato plants was estimated, the RNA was converted to cDNA and subdivided into 4 primer/vector combinations (Figure 1). These 4 subdivisions were further split into 5 cycle breaks allowing for the difference between the effects of Prf and EF1 gene expressions to be better visualized.
Results
The first observable results produced in the experiment was the electrophoresis gel generated from the whole isolated tomato plant genome. The bands appeared universally across all of my co-experimentalists' wells as 2 primary bands lower down the ladder than the large genomic artifacts present near the starts of the lanes (Figure 2). Smaller pieces appear infrequently down the lane, but are relatively insignificant due to the strong presence of the two center bands.
Figure 2. Electrophoresis gel results ran on the PCR product of the whole tomato genome. Ladder is to the far left of the image, the fifth well indicated by the arrow depicts the outcome of my sample.
Following the initial genome isolation, primers were combined with reaction buffer, Taq polymerase and dNTPs and ran through polymerase chain reaction (PCR) to amplify the targeted Prf fragment. These highly duplicated genes were digested with restriction enzymes alongside their intended vectors, TRV2 and ligated together into the TRV2::Prf construct. This ligation was transformed into E. coli with limited success. Only 3 bacterial colonies grew from the self-ligation plates over the two day 37℃ incubation period while none grew on the recombination with TRV2 plate. This likely could have been due to an error in the decontamination process conducted on the spreader, which was immersed into ethanol and quickly burned over a Bunsen flame in order to fully rid the tool of remaining E. coli bacteria. The spreader could have accidentally been used too early on the secondarily applied recombination plate, killing the bacteria and preventing their colonization process. The plasmids were then isolated from the E. coli and were checked in an electrophoresis gel to ensure that the restriction enzymes properly processed the fragment and vector (Figure 3). The gel observed three different varieties of sample made up to this point; one being the undigested TRV2 which did not appear in the gel possibly due to contamination or improper application of primers or RE, the second being the RE digested plasmid which had an expected appearance and size, and the third being the Prf PCR fragment on its own, which did not visually stand-out (as expected).
Figure 3. Electrophoresis gel results for the restriction enzyme verification step, run after PCR. Personal wells are indicated by the red arrows, the leftmost first row contains undigested TRV2 plasmid, the second row contains RE digested TRV2, the third has Prf PCR fragment.
After disposing of the faulty digests and retaining my plasmids (replacing my samples with Ryan’s) the DNA was transformed into Agrobacterium via electrophoresis, with my solution producing a 6.0 reading. Three different Agrobacterium solutions were created, the primary focus being on the suspension containing the Prf silencing construct, however the other two contained positive and negative expression controls, PDS and a blank TRV2 plasmid respectively. These solutions were run on a gel and all produced expected results, each vector appeared high on the lane due to their relatively similar masses. The bacterial suspensions were then injected into the leaflets of young tomato plants via a syringe to allow the Agrobacterium to integrate the T-DNA fragment of the plasmid containing our genes of interest into new cells created by the seedling as it grows. Over two weeks the PDS treated plant unfortunately died, and one of the Prf plants was seemingly contaminated showing whiting on its leaves.
Figures 4-7. Images of tomato plants after inoculation of transformed Agrobacterium and before infection with P. syringae. Figure 6 and 7 (leftmost) depicts the plants syringe-injected with the Prf constructs representing the experimental subjects. Figure 8 was treated with the blank TRV2 vector with any silencing genes present serving as the negative control. Figure 9 (rightmost) shows the PDS treated seedling which died during the Agrobacterium uptake phase, but would have represented the positive control due to PDS expression manifesting as a strong visual indicator whitening the leaves.
The two lab days following our observations on the plantlets involved harvesting and processing leaf samples for two different functions, one was for the sequencing of the TRV2::Prf construct, which would be received by us a few weeks after, as well as for RNA isolations (Figure 12). After the transformed plants had samples collected before testing out assumptions, we vacuum inoculated them with P. syringae to study how the disease interacted with them post-transformation. Only the Prf and negative control were infected as their symptoms were expected to be explicitly different. The Prf silenced plant demonstrated strong symptoms of infection, reddish and brown speckles coated new leaves while the overall appearance of the organism was wilty, some older leaves even showed signs of necrosis. The negative control plant, only treated with a placebo empty vector, had only displayed a very limited sign of sickness, very few spots showed and virtually no wilt occured.
Figures 8-11. Images of tomato plants after inoculation of P. syringae. The Prf gene-silenced plant is pictured to the far left (figure 9) followed to the right by a close-up of the leaf displaying the typical symptoms of Pst infection including red-brown speckles and withering leaves (figure 10). The two rightmost pictures (figures 11 & 12) display the post-inoculated blank TRV2 vector treated plant overall, and a closer view to its leaves depicting nearly no visible sign of sickness.
The RNA isolation product was converted into complementary DNA (cDNA) to provide stability for further observation of the level of mRNA produced from the introduced sequence in the Prf transformed plant. The RNA isolate solution was run in a 1% agarose gel to ensure its presence for conversion (Figure 12). The results came out as expected, nearly matching the band lengths and brightness of the original DNA genome signalling that the whole sequence has not been significantly changed. After the indication of clean, whole RNA, a reverse-transcriptase (RT) reaction was completed to convert the unstable, single-stranded RNA into cDNA for use in a downstream PCR reaction.
Figure 12. Recombinant DNA gel picture from the post-infection RNA isolation and conversion to cDNA. My sample well is marked by the initials “BB”, the fourth well right from the ladder.
The sequencing results arrived between the RNA isolation and cDNA conversion phases of the RT-PCR and were analyzed with the NCBI BLAST tool available through the National Institute of Health to determine accuracy in the construct. Likely due to a failure in ligation or to a mix-up involving the three Agrobacterium inoculants, the construct sent in to be sequenced was most accurately matched to a blank tobacco rattle virus plasmid, with no Prf fragment being detected in the solution. While these results lack insight to the composition of the gene fragment do not provide hard reassurance in the results found so far, learning the details of reading sequence data and understanding the various flaws and errors that can occur in the ligation, recombination or sequencing processes was still quite insightful.
Figure 13. Sequencing results from the Prf silencing solution. Results start dirtier and more clustered, but gradually clean-up by the thirtieth base pair, the whole sequence runs approximately 470 bp.
Leaf samples visible displaying signs of infection were also harvested and processed down to the residual P. syringae bacteria. These bacteria solutions were diluted into various concentrated mixtures and plated at original and 10 fold dilutions for bacteria gathered from the TRV2 blank plants, and 10^3 and 10^4 dilutions for the bacteria collected from the Prf construct transformed plants. The amount of bacterial colonies that formed represented the original number of bacteriums in the dilution sample and could be calculated to a current population concentration per square centimeter of leaf tissue using the OD600 reading. The concentrations are displayed in Figure 14 showing the exponential growth of bacteria within the plants, and notably, how much more the P. syringae colonized the Prf silenced plantlets.
Figure 14. Logarithmic chart depicting the calculated amount of bacterial growth in the TRV2 and TRV2::Prf transformed tomato plantlets. The left column depicts the bacteria count in the original inoculum introduced to the plant leaves, the right column shows the calculated concentration (cfu/cm^2) of bacteria after 5 days of exposure. Within each column is an orange bar (TRV2 counts) and a green bar (TRV2::Prf counts). The y-axis is logarithmic with a factor of 10, due to both overall bacterial growth being immense and the difference in scale between growth in the two tomato plants.
Following the sequencing of the TRV2::Prf construct, the cDNA solution was then divided into 2 PCR solutions. Another two solutions were produced under the guidance of the lab TA which only contained empty TRV2 plasmids to serve as negative controls for the secondary comparison factor. To test if the gene expression of P. syringae immunity is uniquely controlled through the use of our Prf targeted primers, Expression Factor 1 (EF1) primers was applied into a TRV2 and TRV2::Prf recombinant PCR solution to compare with our specialized Prf primers in the opposing set of PCR mixtures. The first and third solutions only containing blank plasmids show similar results, over time the plasmids are replicated in the thermocycler, unaffected by the primers included as they do not explicitly target any individual pieces of the sequence. The fourth PCR solution depicts a similar pattern to mixtures 1 and 3, in the EF1 primer immersion, the Prf is not extruded from its construct as the expression factor does not have a matching sequence within its recombinants. However, the second solution has far fainter bands due the silencing of the breakdown of the Prf gene triggered by the inclusion of the Prf specific primers demonstrating that the Prf expression is silenced by the matching primers as opposed to the EF1 groupings.
Figure 15. UV gel image of the Prf and EF1 comparison phase. The five rows on the far-right encompass the first PCR solution and contain blank TRV2 vectors with Prf forwards and reverse primers. The next five wells moving towards the right contain the TRV2::Prf assembly with Prf primers, and is much more muted in appearance. The next block has the blank TRV2 plasmid in EF1 primers, and the last five contain the Prf construct with the EF1 primers.
Discussion & Conclusion
Overall, the experiment was quite unfortunately plagued with issues that slowed the latter half of the experiment and forced compromises and stopgaps to be employed constantly, however, certain important patterns did emerge during particular sections of experimentation. The introduction of the TRV2::Prf gene silencing construct vastly reduced the tomato plant’s immunity to P. syringae and therefore shows the influence the gene expression has over the trait of disease resistance. The growth of the tomato plants after injection with transformed Agrobacterium provided the most insight on the effects of how gene expression can massively change the phenotypic traits of plants. The tomato seedling exposed to PDS-recombinant bacteria grew bleached leaves dramatically demonstrating a successful transformation effort. The desired target factor needed the second layer of selection, exposure to P. syringae, to observe how infection displays itself on a plant that has had its ability to produce the Prf protein drastically diminished. The Prf silenced plant had strong visual indication of successful infection by the virus, with leaf tissue being coated in rust colored speckles, decay on the edge of some leaflets and overall wilting across the whole of the plantlet. The comparison between the effects of the plasmids when exposed to EF1 primers or the Prf primers also provided strong evidence of how the silencing of Prf is not induced by other related expression factors. From the sections of the experiments that functioned correctly, the hypothesis of Prf resistance to Pst in the resistant strains of tomato (such as our RG-PtoR) being regulated by the indirect recognition of the Pst effector by Prf seems to be corroborated. Even from portions that did not produce expected results, most did not explicitly show contradicting evidence of Prf’s influence on plant immunity and Pst recognition, but rather resulted in dud visualizations that depicted a clear error in application or an issue with the sample provided. Were these protocols to be redone, results from co-experimentalists seems to support that more evidence would be given to the position that the silencing of Prf expression in tomato plants reduces the organisms’ ability to construct recognition complexes with Pto, slowing the immune response to Pst, increasing the rate and severity of infection.
Perhaps the most notable error of the whole of the experiment runs as far back as the potential tomato plant genomic DNA contamination that occurred in late January, when after the protocol, the DNA bands did not properly materialize on the gel. While the correction cycle focusing on a different set of both PCR primers and DNA seemed to indicate that the DNA sample itself was prone to errors and not the primers designed for the Prf fragmentation, both were replaced, however several following procedures failed to properly produce results and either resulted in replacement by adjacent and available substitutes or were simply kept in the experiment and brought to the next phase with only notations about prior issues being attached. While replacements of samples with co-experimentalist or TA samples managed to occasionally produce expected results in the post-Agrobacterium transformed plant section, the final check of the TRV2::Prf assembly sequence after the treated seedlings were inoculated with Pst was the largest communication error in the experiment and would be a definite priority for being re-submitting. Due to the result of the blank TRV2 vector, I strongly believe that the incorrect solution was sent to the analysis company as I was late to submission and may have selected the wrong tube when filing it to the TA. While the error is quite small and comes down to clerical responsibility on my part, the resulting large hole of near useless data is bothersome and would be a focus in any re-doing. The bacterial isolation of Pst was also problematic and likely produced unexpected results due to either contamination or improper use of tools, while not the central focus of the experiment, it again would have gathered more convincing data to back the hypothesis and would have been a repeat if it was able to happen. It would be ideal to begin the entire experiment over again from the initial point of contamination in the genomic DNA, while more properly sorting out problems by eliminating and restarting protocols that result in visible failures. However,time was the clear limiting constraint, and it had to be respected in regards to the linear structure of the whole experimental design and what could be allocated to completing the whole of the lab.
Acknowledgements
Thank you Dr. Xiao and Chloe Ramirez (whoever is reading this, make sure the other is thanked) for being so attentive, knowledgeable and helpful with this experiment; despite some of my results, I still feel good about what I have learned and done here and you both were absolutely critical to that feeling. I am certain that I will use what I have learned here going forwards with my interest in genetics and biotechnology.
References
Arunima Purkayastha and Indranil Dasgupta. (2009). Virus-induced gene silencing: A versatile tool for discovery of gene functions in plants. PLant Physiology and Biochemistry, 47: 967-976.
Dominique Robinson. (2004). VIGS VECTORS FOR GENE SILENCING: Many Targets, Many Tools. Annual Review of Plant Biology, 55: 495-519.
Sophia K. Ekengren, Yule Liu, Michael Schiff, S. P. Dinesh-Kumar, and Gregory B. Martin. (2003). Two MAPK cascades, NPR1, and TGA transcription factors play a role in Pto-mediated disease resistance in tomato. The Plant Journal, 36: 905-917.
Tessa M. Burch-Smith, Jeffrey C. Anderson, Gregory B. Martin and S. P. Dinesh-Kumar. (2004). Applications and advantages of virus-induced gene silencing for gene function studies in plants. The Plant Journal, 39: 734-746.