The fight against crop damage from biotic and abiotic stresses has led and continues to lead research in the fields of soil science, stress signaling and hormone responses, plant immunology and immunology more generally.
One breakthrough application from the discovery of RNA interference (RNAi) is the ability to use this method of host protection to specifically target particular pathogens that a plant may be susceptible to. The most prominent example of the application of this technology was its use to save the papaya industry from the Papaya Ringspot Virus (a great explanation of the virus and the use of transgenic technology to control it can be found here).
The discovery and development of CRISPR technology has also been used to tackle plant stress issues in multiple ways, whether that to be to direct endonucleases to invading genetic material or as a tool to insert sequences that allow the plant to generate its own resistance abilities.
However, as the researchers behind a recent paper in the Plant Biotechnology Journal state, whilst the ability to protect plants against DNA-based viruses using CRISPR is increasing, the tool hasn’t found implementation in the effort to control RNA viruses primarily due to it being limited to the restriction of target DNA.
Therefore, the researchers took some recent findings into the ability of some variants of the Cas9 endonuclease (the common nuclease used with CRISPR) which had the ability target RNA sequences within prokaryotes to test whether they could be used to develop resistance to RNA viruses in plants.
Using cucumber mosaic virus (CMV) and tobacco mosaic virus (TMV) as the two RNA viruses to be targeted, the researchers took the Cas9 protein from Francisella novicida (FnCas9) and used it in combination with guide RNAs directed toward a variety of targets within the CMV and TMV genomes, then assessed whether the construct had any effect of infection rates and severity compared to control plants.
The first part of the experiment focused on CMV, a virus with a genome of three parts, each with similar sequences. The FnCas9 was to be directed to 23 sites within the genome selected by the researchers and complementary sequences to those sites were individually synthesised and inserted into separate vectors also containing the sequence for the FnCas9 nuclease.
Constructed vectors were inserted into Agrobacterium which was then injected into a bottom quarter of 25 day old N. benthamiana leaves. A control vector was injected into the opposite bottom quarter of the same leaf and the CMV infection was inoculated in the top half of the leaf.
Figure 1(c) from article, demonstrating the inoculation sites of the experimental vector, control vector and CMV on subject leaves.
Five days later, discs were cut from the bottom quarters of the leaf for analysis of infection rates. The analysis was performed by quantifying virus RNA presence using Reverse Transcription Quantitative Polymerase Chain Reaction (RT-qPCR) and found that the quantity of CMV RNA in the discs containing the FnCas9 + guide RNA was between 40% and 60% less than that contained in the control discs.
Within the same experiment the researchers toyed with different PAM spacers but found that there didn’t seem to be any variability between the different spacers used and the performance of the FnCas9 construct.
The three RNA target sequences which resulted in the best virus resistance were selected for further testing. FnCas9 constructs containing the individual RNA targets and controls were again transiently expressed in N. benthamiana plants and the following day the plants were challenged with the CMV virus. Two weeks later, control plants showed symptoms of CMV infection while severe infection symptoms were avoided in the plants inoculated with one of the three FnCas9 vectors. ELISA-based quantification and RT-qPCR quantification of viral RNA again showed reductions of between 40% and 50% compared to control plants.
Figure 2 from article showing the difference in viral effects between control and transgenic plants (a) and comparison in viral load using ELISA (b) and RT-qPCR (c).
The researchers took a different tact to researching the ability of the FnCas9 constructs to control TMV. They attached a Green Fluorescence Protein to the virus and then targeted the virus with FnCas9 constructs directed to three selected sequences within the TMV genome. Accordingly, expression of TMV would result in detectable fluorescence, whilst the suppression of the virus would result in a reduced amount of fluorescence.
One week after inoculation with the virus, bright green fluorescence was observed throughout the control plants but treated plants showed significantly reduced fluorescence comparatively. Again, RT-qPCR confirmed the reduced virus load in the treated leaves.
Researching the Mechanism
The results appear to show that the FnCas9 plus RNA guides are working well in repressing the RNA viruses. To check whether the synthetic constructs are the true cause of the effect seen and, if so, to delve into how the construct is working, the researchers performed a number of tests.
The first test used immunoprecipitation targeted at a flag placed at the N-terminus of the FnCas9 protein to capture and purify the FnCas9 + RNA guide. When the purified proteins were sequenced and amplified using RT-PCR the CMV virus was also detected, indicating the guide RNA was associating the viral RNA and therefore the FnCas9 protein was being successfully directed to the viral RNA genome.
To check whether the FnCas9 was necessary to the inhibition of the viruse, the FnCas9 was replaced with a GUS reporter gene. When expressed with the guide RNA the CMV virus progression was no longer repressed by the synthetic construct. Therefore, the FnCas9 protein was required for virus repression.
The next method used to check the mechanism was to add a nuclear localisation signal to the FnCas9. Restricting the protein to activity only within the nucleus of the plant cells again resulted in reduced virus repression compared to that seen previously.
Point mutations were then made to the active regions of the FnCas9 to disrupt its endonucleolytic activity. Interestingly, the disruption to the ability of the protein to cut the RNA sequences made no difference to the ability of the construct to reduce viral load and disease severity.
The last test the constructs were put to was test whether the ability to bind the viral RNA effected the productivity of the construct. Mutations were again made to the construct, this time at the RNA-binding arginine-rich motif (ARM) of FnCas9. The inability of the FnCas9 protein to bind to the RNA resulted in reduced ability of the construct to repress the RNA virus.
Accordingly, the ability of the FnCas9 synthetic constructed directed to the two RNA viruses to curtail infection relied not the endonuclease cutting the viral RNA but actually relied on its ability to bind to the viral genome, an unexpected result which the researchers later gave the name CRISPR Interference (CRISPRi) in the discussion section of the paper.
Stability of Resistance
Transient expression of the FnCas9 constructs resulted in repression of virus load, but to be of assistance to crop protection, stable repression of virus is required.
Therefore, the three best performing constructs targeting CMV were used to transform Arabidopsis thaliana plants. Homozygous T2 plants for each construct were created and selected for challenge with CMV infection. While control plants showed severe effects of infection two weeks after being infected, transgenic plants mild symptoms in most lines with two lines having no obvious symptoms. ELISA-based quantification confirmed the virus inhibition.
The same lines were continuously harvested until the sixth generation was produced and then challenged with infection. The randomly selected lines showed a stable resistance to the virus .
The study demonstrates a novel method of reducing the effects of a couple of important RNA viruses that can have dramatic impact on plant viability and productivity. Whilst the logic behind the study is sound and seemed destined to work given FnCas9 had previously been demonstrated to target RNA, the mechanism by which the endonuclease repressed was not anticipated.
Acting effectively like RNAi, the FnCas9 seems to block the translation of the virus’ RNA genome, halting its ability to infect the plant cells. Further, constitutive expression was shown to be stably inherited in subsequent generations.
The researchers point out in the discussion section of the paper that this strange method of reducing viral loads could have some advantages, such as a reduction in the production of mutant forms of the virus from cleaved RNA strands escaping the cells. However, it does have a similar limitation to RNAi in the creation of evolutionary pressure on the virus to mutate to avoid the repression of the site of the genome targeted by the FnCas9.
Although the same problem of the “arms race” between virus and plant arises in the context of this research, the findings certainly give some hope that CRISPR can be used to reduce the effects of viral damage to crops.