Progress 10/01/19 to 09/30/20
Outputs
Target Audience:Researchers who are interested in utilizing genome editing technologies toward the improvement of crop species. Changes/Problems:As described under Accomplishments, we found that our original strategy for obtaining high efficiency HR-mediated genome editing proved to be unsuccessful. Therefore we are adopting the innovative technique known as prime editing as a better means for accomplishing the goals of the project. What opportunities for training and professional development has the project provided?This project provided excellent training of genome editing technologies for a postdoctoral research associate. It also provided a valuable research experience for an undergraduate worker in the lab. How have the results been disseminated to communities of interest?
Nothing Reported
What do you plan to do during the next reporting period to accomplish the goals?As described under Accomplishments, we are now focusing on the promising technique referred to as prime editing as a means of introducing precise changes to the tobacco genome.
Impacts
What was accomplished under these goals?
Based on the information obtained during that past reporting period, we generated a CRISPR-Cas9 construct and an associated correcting template designed to introduce a simple amino acid substitution into the nitrate reductase genes of tobacco (designated Nia1 and Nia2). The constructs were introduced into tobacco protoplasts via PEG/Ca++ precipitation. At total of 855 hygromycin resistant T0 plants were regenerated from protoplasts. All 855 plants were screened for mutation at the Nia1 and Nia2 loci. Twenty-three percent of the plants were wild type at both Nia loci; seventy-six percent of the plants had an NHEJ-mediate insertion/deletion (indel) event at Nia1, Nia2, or both. A total of 3 plants were identified that possessed what appeared to be "clean" HR events where the Ser codon at position 523 had been replaced with a codon specifying Asp. The three plants possessing the putative HR events were grown to maturity and self-fertilized. In addition, a few plants containing only NHEJ-mediated mutation events were also carried on to the next generation as controls. Once the T1 generation plants had achieved sufficient size, they were genotyped using primers directed against the hptII selectable marker gene to identify individuals lacking the CRISPR-Cas9/correcting template construct. Plant lacking hptII were subsequently genotyped by DNA sequence analysis of PCR products spanning the region of interest within the Nia1 genes. The putative HR events were present in the progeny of all three parental T0 plants that possessed the mutation. Furthermore, we could readily identify every Nia1 allele combination among the progeny of each parental plant (HR/HR; HR/3bp del; 3bp del/3bp del). Because our PCR-based genotyping across the target region of Nia1 was limited to a small fragment of the gene, we subsequently expanded our characterization of Nia1 in the HR/HR progeny. Unfortunately, the extended sequence verification of Nia1 revealed an unexpected anomaly located 141 bp downstream of the CRISPR-Cas9 cut site. At this site, a 450 bp sequence corresponding to a region of the CRISPR-Cas9 vector involved in its replication in E. coli was observed. The 450 bp of vector DNA was not contiguous, as the first 126 bp of this insert was in the antisense orientation with respect to the CRISPR-Cas9 cassette, while the remaining 324 bp was in the sense orientation Thus, the three individuals that we initially believed to have cleanly incorporated the change via HR-mediated editing proved to have vector sequence inserted 141 bp downstream of the Cas9 cut site. Because of the our lack of success in developing an efficient HR-mediated genome editing protocol in tobacco, we decided to pursue a different approach. In 2019, a publication appeared in the journal Nature describing an novel genome editing technology referred to as "prime editing" (Anzalone et al., 2019, Search-and-replace genome editing without double-strand breaks or donor DNA, Nature doi:10.1038/s41586-019-1711-4). We believe that prime editing represents the most innovative breakthrough in genome editing for the past several years. By creating a Cas9 nickase fused to a reverse transcriptase, and extending the guide RNA to prime the synthesis of the sequences needed for correction at the site of chromosomal nicking, these authors were able to create a system where by precision nucleotide substitutions could be introduced without having to rely on the very inefficient process of HR-mediated DNA repair. We obtained the core prime editing construct described in the Nature paper from Addgene, and are in the process to modifying it with the appropriate promoter and termination sequences for expression in plants. Once these adaptations are complete, we will use this vector to target Nia1/2. By using the prime editing approach, we can go back to using Agrobacterium transformation for introducing editing constructs, and avoid the exceptionally labor and time intensive process of protoplast transformation and regeneration. Furthermore, once the plant-directed prime editing vector has been validated for one target, it will subsequently be very easy to apply this same vector toward any new target in the tobacco genome for introducing the types of genome editing changes that could previously only have been accomplished via HR-directed repair.
Publications
- Type:
Journal Articles
Status:
Published
Year Published:
2020
Citation:
Wu. Q., La Hovary, C., Chen, H-Y., Li, X., Eng, H., Vallejo, V., Qu, R. and Dewey, R.E. 2020. An efficient Stevia rebaudiana transformation system and in vitro enzyme assays reveal novel insights into UGT76G1 function. Sci. Reports 10:3773.
- Type:
Journal Articles
Status:
Published
Year Published:
2020
Citation:
Lewis, R.S., Drake-Stowe, K.E., Heim, C., Steede, T., Smith, W. and Dewey, R.E. 2020. Genetic and agronomic analysis of tobacco genotypes exhibiting reduced nicotine accumulation due to induced mutations in Berberine Bridge Like (BBL) genes. Front. Plant Sci. 11:368.
|
Progress 10/01/18 to 09/30/19
Outputs
Target Audience:Researchers who are interested in utilizing genome editing technologies toward the improvement of crop species. Changes/Problems:
Nothing Reported
What opportunities for training and professional development has the project provided?This project provided excellent training the genome editing technologies for a postdoctoral research associate. As a result of the experience the postdoc obtained while working on this project, she was able to acquire a good job at a major agricultural biotechnology company. How have the results been disseminated to communities of interest?A poster was presented at a genome editing conference held in Raleigh, NC in November, 2019. What do you plan to do during the next reporting period to accomplish the goals?We will continue following the research plan outlined in the proposal.
Impacts
What was accomplished under these goals?
Progress Toward Objectives 1 and 2 As explained in my Project Proposal, we have come to the conclusion that the system that would be most effective for inducing and recovering HR-mediated genome editing events in tobacco would be through the introduction of the editing vectors into protoplasts, followed by the regeneration of the protoplasts back to whole plants. The use of protoplasts toward genome editing proceeded in two stages: (1) working out the conditions whereby we could efficiently regenerate whole plants from tobacco protoplasts; and (2) the subsequent introduction of genome editing constructs into the protoplasts by PEG/Ca++ precipitation, followed by regeneration of whole plants in the presence of antibiotics to kill all nontransformed individuals. We have been successful in accomplished Objective 1, the development of an efficient protoplast regeneration protocol in the absence of DNA transformation. Using the protocol that we optimized, we were able to take numerous protoplast from both K326 and TN90 tobacco lines all the way through the rooting process, and whole plant growth in soil. Once we were confident that we were capable of regenerating whole plants from tobacco protoplasts, we initiated experiments that included the CRISPR-Cas9 construct targeting the NtEPSPS-2a and -2b genes, together with the correcting template needed to introduce the sequence changes at codons 102 and 106 that can confer glyphosate tolerance. This protocol differs a little from the one in which we regenerated nontransgenic plants from protoplasts, as an additional PEG/Ca++ precipitation step is included, followed by washes to remove the PEG and Ca++ prior to transfer to a medium that promotes cell wall regeneration (then subsequently cell division). In addition, once callus clusters reach sufficient size, hygromycin is added to the culture media to select for those individuals carrying an integrated transgene. Given that the protoplasts are very sensitive to any type of environmental perturbation, the transfection procedure undoubtedly will result in a reduction in regeneration efficiency. Furthermore, the antibiotic selection will also reduce the final numbers of events the will survive and can be propagated back to whole plants. Nevertheless, for the experiments where we've introduced the NtEPSPS genome editing nuclease + correcting template, we generated dozens of individuals at various stages throughout the process that have survived thus far, the most advanced of these were transferred to soil. Progress Toward Objectives 3 and 4 We have conducted protoplast transfections and deep sequencing using numerous constructs, as described in the original proposal. Once we began the in silico analyses of the very large sequence datasets produced by the Illumina sequencing, it became apparent that the Cpf1 nuclease we were working with was minimally functional (and possibly completely nonfunctional) in tobacco protoplasts. The original manuscript describing the new Cpf1 class of editing enzymes used two different sources for the Cpf1 gene; one gene was isolated from a bacterial species of the genus Acidaminococcus (designated AsCpf1) and the other gene was isolated from the genus Lachnospiraceae (designated LbCpf1). The Cpf1 genes from both species worked well in the original study conducted using human culture cells. When we initiated this project, we arbitrarily chose to have the AsCpf1 gene custom synthesized for expression in tobacco (by the company GenScript) for our experiments in tobacco. By the time we obtained the AsCpf1 construct and cloned it into our plant expression vectors, a two papers came out demonstrating that in two different plant species the LbCpf1 gene works very well, but AsCpf1 is only minimally active. Since we had already invested a lot of time and money (the custom synthesis of a gene the size of AsCpf1 is quite expensive) on our AsCpf1-based constructs, we decided to go ahead and use it anyway, but similar to the recent two plant papers, we observed little genome editing activity from AsCpf1 expressed in tobacco protoplasts. We have since ordered a custom synthesized LbCpf1 gene from GenScript, and intend to repeat this portion of the experiment once we've finished generating the constructs with this new Cpf1 gene. Seven constructs targeting the BBLd gene and seven constructs targeting the SuRA gene were tested for their transient expression in tobacco protoplasts. For the CRISPR-Cas9 and CRISPR-Cas9 nickase systems, constructs were generated using three different sgRNAs. This was done because the "rules" of what constitutes an optimal target site for CRISPR-Cas9 are still unclear, and occasionally a given sgRNA will not work well, or at all. Our inability to access multiple ARCUS enzymes, however, is limited (due to the fact that they can only be made by Precision Biosciences). Therefore, only one ARCUS construct was available for testing at each target gene. All target sites for BBLd and SuRA were selected to be within 50 bp of each other, thus enabling us to use an identical correcting template of all nucleases tested for a given gene target. To determine whether the ratio of HR-mediated versus NHEJ-mediated editing events could be skewed in favor of HR according to the type of nuclease employed, 48 hrs after PEG/Ca++ transfection with the various genome editing constructs, the protoplasts were harvested and PCR was conducted on genomic DNA preparations using primers flanking the region containing the targeted sites. The complete interpretation of the dataset is beyond what can be easily explained in this report, but the results of the protoplast experiments clearly support the hypothesis that genome editing events can be skewed in favor of those mediated by HR (which are the most desirable for any type of complex genome editing) if one were to use either an ARCUS enzyme or a CRISPR-Cas9 nickase, rather than the conventional CRISPR-Cas9 system. For the CRISPR-Cas9 nickase, for two out of the three constructs tested for BBLd, and all three constructs targeting SuRA yield only HR events. This is not unexpected, given that the activity of the enzyme merely nicks the strand, and doesn't make the complete double-stranded break that is necessary to activate the NHEJ repair machinery. For the ARCUS enzymes, 27.2% of the genome editing events in BBLd and 21.2% of the editing events in SuRA were HR-mediated. This is a very high percentage in comparison to regular CRISPR-Cas9 constructs, where HR events only ranged from 0.5 - 1.9% of the total, depending on the specific construct. Although CRISPR-Cas9 nickase and ARCUS enzymes greatly favor HR-mediated genome editing over NHEJ compared to the traditional CRISPR-Cas9 system, the latter clearly has much greater overall mutagenic activity within the transient protoplast system. This type of differential in overall activity between CRISPR-Cas9 and ARCUS, however, is unlikely to be recapitulated within the context of generating stable transformed plants. Despite the fact that the ARCUS enzymes yielded overall mutagenesis efficiencies of 0.5% and 0.1% for BBLd and SuRA, respectively, in the transient protoplast system, we observe frequencies of ~15 - 20% when these same nuclease constructs were used to generate knockout mutations in their respective gene targets when used in stable transformation experiments. Thus, the overall conclusions from the experiments conducted thus far would greatly favor the use of ARCUS enzymes for conducting HR-mediated genome editing, while CRISPR-Cas9 is clearly superior when the goal is merely to obtain NHEJ-mediated knockout mutations. If one is constrained to using the CRISPR-Cas9 system, however, our results greatly favor the use of the nickase version of the enzyme (generated by replacing the Asp at position 10 of the enzyme for an Ala).
Publications
- Type:
Book Chapters
Status:
Accepted
Year Published:
2020
Citation:
Chandrakanth, N.N. and Dewey, R.E. 2020. Tobacco: Past, present and future. In U. Dreher, ed. New Research on Tobacco, Nova Sci. Publishers, Hauppauge, NY. pp. 1-22.
|