Source: OREGON STATE UNIVERSITY submitted to
DNA METHYLATION CONTROL FOR SITE-SPECIFIC RECOMBINASES: BIOSAFETY AND EFFICACY
Sponsoring Institution
National Institute of Food and Agriculture
Project Status
ACTIVE
Funding Source
OTHER GRANTS
Reporting Frequency
Annual
Accession No.
1030929
Grant No.
2023-33522-40414
Cumulative Award Amt.
$649,916.00
Proposal No.
2023-02481
Multistate No.
(N/A)
Project Start Date
Jul 1, 2023
Project End Date
Jun 30, 2026
Grant Year
2023
Program Code
[HX]- Biotechnology Risk Assessment
Project Director
Goralogia, G.
Recipient Organization
OREGON STATE UNIVERSITY
(N/A)
CORVALLIS,OR 97331
Performing Department
(N/A)
Non Technical Summary
Advances in biology are enabling more efficient genetic transformation and editing, including for biosafety-promoting traits such as for genetic containment. Many asexually propagated species, such as the forest trees we propose for study, are notably recalcitrant to transformation yet in need of biosafety-promoting transgenes or edits due to their potential large ecological impacts and propensity for long distance gene flow. However, the genes employed (such as CRISPR/Cas9) often need to be excised (removed) to facilitate regulatory approval or ensure plant health--particularly in these species for which sexual segregation is difficult or impossible. The most common tools for excision are the Cre and Flp recombinases, however, their activity is constrained by target-site DNA methylation in plants, making their use inefficient. Various means for demethylation are known to enhance efficiency, but have not been compared or optimized, nor have their off-target impacts examined at the locus, genome, or phenotypic levels. Using tobacco, Arabidopsis, and poplar as study systems, we will: 1) Develop novel recombinase target sites that are devoid of potentially methylated cytosines but are active in supporting excision; 2) Test a variety of constitutive, inducible, and site-specific methylation inhibitors for enhancement of transgene excision; and 3) Examine local, genome-scale and phenotypic impacts of the most effective methylation control systems. This work will help enhance genome transformation and engineering tools relevant to biosafety technology applications and inform regulators of the degree to which they should be concerned with off-target impacts when recombinases are used for genome modification.
Animal Health Component
(N/A)
Research Effort Categories
Basic
50%
Applied
(N/A)
Developmental
50%
Classification

Knowledge Area (KA)Subject of Investigation (SOI)Field of Science (FOS)Percent
20106991080100%
Knowledge Area
201 - Plant Genome, Genetics, and Genetic Mechanisms;

Subject Of Investigation
0699 - Trees, forests, and forest products, general;

Field Of Science
1080 - Genetics;
Goals / Objectives
Project Alignment with Program GoalsThe scope of the proposed work most closely aligns closely with strategic goal 5a: "Research addressing phenotypic effects associated with on- or off-target errors in GE organisms developed using genome editing technology or other genetic engineering techniques and potential hazards or adverse effects to the environment associated with these phenotypic effects." For all agricultural crops, but especially for forest trees that provide many ecosystem services, adverse effects to genomes, phenotypes, and thus plant health are off-target effects of concern. This proposal aims to develop an improved system for transgene excision by control of DNA methylation, but also to assess the risks of unintended off-target epigenetic effects of attenuating DNA methylation at local and genome scales. This proposal also indirectly addresses goal 1c, "Development or evaluation of effective bio-confinement strategies, including molecular and/or genetic techniques, to limit gene transfer (gene flow) or outcrossing to sexually compatible organisms..." As discussed above, if our system for excision elevates the capacity for use of DEV genes to promote transformation, it may enable the much wider use of risk mitigation tactics such as the use of bioconfinement motivated edits and transgenes. Finally, by more effective removal of DEV genes, we hope to expand their use and thus improve the efficiency of transformation systems. Thus, it also addresses the "urgent need for increased plant transformation capacity in the USA" identified in the recent "Dear Colleague Letter: Advancing Plant Transformation" from NSF, that also includes USDA-NIFA interests, including those relevant to the BRAG program (John Erickson, BRAG Program Director, personal communication, Dec. 16, 2022).
Project Methods
Aim I:To select for highly efficient sites for follow up study in planta, we will clone direct repeats of the novel cytosine-free lox and FRT sites in the same conformation within our fluorescent-switch-reporter vectors. These sites in linearized plasmids will be tested in vitro using purified Cre or Flp recombinase to look for recombination products and compared against loxP and FRT sites for an assessment of efficiency. A total of 15 such lox and FRT sites will be cloned and tested in vitro (a total of 30 sites). The best 6 lox and FRT sites (a total of 12 constructs, alongside loxP/FRT and without recombinase controls) will be cloned within our fluorescent-switch-reporter-constructs with an antibiotic resistance gene and a heat-shock inducible Cre or Flp recombinase. These will be transformed into Arabidopsis by the floral dip method. T1 plants able to germinate on selective media will be grown in vitro and subjected to heat-shock treatments to induce Cre or Flp mediated excision (approximately 4hr pulses at 42C for up to two weeks); pilot studies with already developed vectors will first be undertaken to determine optimal heat treatments.We will assess recombination rates by analysis of leaf tdTomato and GFP fluorescence by microscopy and quantification by qPCR on isolated genomic DNA using primers within and spanning the recombination junction. Segregation ratios of resistant seeds will be determined by germination of T2 on selective media seed to look for any silencing effects which spread outside of the recombination-site flanked domains. The best two performing cytosine-free lox and FRT sites will be investigated in the readily transformable hybrid poplar (Populus tremula x alba) genotype 717-1B4 via standard organogenic Agrobacterium-mediated transformation (as has been employed in our laboratory for more than two decades), alongside loxP/FRT and no-recombinase controls. These will use the same constructs developed for Arabidopsis transformation, with a heat-shock inducible Cre or Flp containing module. Heat-shock treatments will be made at the CIM (callus-induction medium) to SIM (shoot-) transfer stage at 4 weeks after transformation, when large transgenic calli have developed (e.g., Figure 7). We will analyze excision outcomes using similar approaches, namely: 1) Fluorescent sector analysis by microscopy, 2) fluorescent quantification using a previously developed machine-vision and hyperspectral imaging platform (Yuan et al., 2022), and 3) DNA quantification of excision via qPCR analysis using primers outside or spanning the excision product. All treatments will employ at least three biological replications.Aim II:We aim to test three different classes of tools for the reduction of Cre-induced transgene silencing during transformation, which include: 1) Knockdown of the RdDM machinery (DRM2, first performed in the Liu et al (2021) study), 2) CRISPR-Cas9 mediated demethylation of the transgene through protein fusion to the human TET1 demethylase catalytic domain and gRNAs designed to target the T-DNA, and 3) Viral silencing suppressors to reduce siRNA-mediated silencing (P19) or reduce DNA methylation through interacting with plant demethylases (βC1).To study these effects, we will perform transformations in poplar using inducible (heat-shock) or constitutive versions of each approach. . To study the effect of expression of these components on regeneration and transgenic plant recovery, we will use our machine vision system to determine callus size and shoot regeneration rates in each treatment, and quantify the resultant number of fluorescence-positive shoots that we recover from each experiment. A minimum of 20 events, and five replicate plants per event, will be obtained for each transformed construct. To analyze the DNA methylation outcomes on the inserted T-DNA (excised or unexcised), we will isolate DNA from leaves of recovered transgenic events, perform bisulfite treatment, and amplify and clone five regions within the T-DNA, including the promoter and gene body of Cre, the antibiotic resistance gene, and the recombinase target sites and nearby DNA. Approximately 10 E. coli clones of these regions will be Sanger-sequenced for each event to quantify the numbers of methylated cytosines; at least five events of each construct type will be studied. After completion of this analysis, we will build vectors which place demethylation components and Cre recombinase inside of lox-flanked zones within the T-DNA, using the best two performing systems. A control construct without demethylation components but with Cre in the same conformation will be produced. Final comparisons of excision rates will be analyzed and the resulting plants will be used for genome-wide analysis of DNA methylation in Aim III.Aim III:Genome-wide methylation analysis. To assess the genome-wide impacts of DNA demethylation we will isolate DNA from two of the best performing construct types, with at least two independent transgenic events and three independent clonal ramets (trees) from each. We will also include two control events which went through the same transformation process but without excision/demethylation components. Using an Illumina NextSeq instrument at the OSU Center for Quantitative Life Sciences (CQLS) core facility, we will perform short read sequencing of bisulfite-converted DNA samples to 20x coverage. Bisulfite-converted sequence reads will be mapped to the phased, highly quality genomes for Populus tremula x alba 717-1B4 available at the University of Georgia. Cytosine methylation status for genome windows, as well as averages for various annotated segments (e.g., coding regions vs. introns and upstream regions) will be analyzed and statistically compared among constructs and gene insertion events using BatMeth2 . Finally, to assess if genome-scale changes are affecting the methylation of important genes, we will interrogate our bisulfite-seq data to analyze changes in methylation patterns within and near to about 10 genes for whom expression is known to be at least partly controlled by methylation, and for whom perturbations may affect development. These are likely to include poplar homologs of genes such as WUSCHEL, LEAFY COTYEDON 1 (LEC1), LEC2, BABY BOOM, RESPONSE REGULATOR 3, SOMATIC EMBRYOGENESIS RECEPTOR-LIKE KINASE 1 and YUCCA 2.Greenhouse trial. To statistically assess if there are vegetative consequences of induced Cre and demethylation component expression, we will perform a randomized greenhouse trial using transgenic and control 717-1B4 poplars. The controls will be the typical "escapes" that elude antibiotic selection, regenerate shoots and roots, and are found to be transgene-free based on PCR). The transgenics will include approximately five gene insertion events, and five replicate trees from each event, from all constructs tested under Aim II. They will be assigned randomly to blocks and assayed for height, diameter, leaf mass and area, relative chlorophyll content, and general crown morphology (e.g., extent of apical dominance through branch:height length-ratios) over approximately three months of growth.Quantitative AnalysisAll analyses will employ biological replicates. There are generally 12 explants per dish and 5-10 plates per treatment. Data is visually checked for normality and also subjected to Q-Q graphs and associated correlation statistics. When data are not normal in distribution, we transform to normality where possible, or employ attribute methods such as Chi-square and Fisher's Exact Tests where there are useful categories of phenotypes that can be recognized or constructed, or use non-parametric methods. For normal or close to normal data, we will inspect for outliers and heteroscedasticity, further transform as needed, and employ ANOVA, Tukey, and Students-T tests for assessment of whether observed differences among means are likely to be the result of chance.

Progress 07/01/23 to 06/30/24

Outputs
Target Audience:Our target audience for this reporting period was professors, researchers, postdocs, industry members, students, government regulatory bodies, and the general public. These were reached through internal scientific meetings at OSU and with collaborating research and industry groups, as well as lectures and invited talks. Changes/Problems: Aim I originally included heat-shock driven Cre and FLP reporter constructs for stable poplar transformation. Our new mScarlet3-intron reporter has excellent brightness and no background in the absence of recombinase, but the excision deletion distance is small (~300bp). We hypothesize this decrease in distance may increase recombination activity relative to prior reporters in planta, leading to premature excision in many transgenic events. We have switched to XVE/β-estradiol induction for Cre and FLP to improve recombination control and better assess difference in activity between sites. What opportunities for training and professional development has the project provided?During this reporting period, the project has enabled PD Goralogia to attend the Society for in vitro Biology annual meeting, and the IUFRO Tree Biotechnology meeting. It has also provided the opportunity for PI Goralogia to collaboratively interact and share tools with USDA-ARS and other academic groups. It has also provided the opportunity for Co-PI Strauss to discuss "clean" gene editing technology for forest trees within the context of biosafety and new transformation technologies (PlantGENE annual meeting). One student from the OSU URSA/Engage program worked on the project in 2024 and will continue to be involved 2025 until graduation (https://academicaffairs.oregonstate.edu/research/ursa-engage). One graduate student worked on the project in summer 2024 to assist with FLP/FRT site screening and develop molecular biology skills. How have the results been disseminated to communities of interest?During this reporting period, we have made efforts to reach our target audiences through four oral presentations, discussions with industry and USDA collaborators, and education activities via OSU classes. A description of progress was provided to industrial collaborators of our GREAT TREES research consortium on and September 4th, 2024. What do you plan to do during the next reporting period to accomplish the goals?Aim I: In the next reporting period, we will complete the analysis of stable transgenic recovery and excision rates using cytosine-free lox and FRT sites in Arabidopsis and poplar. We will also report the results of alternative in vitro screening approaches for cytosine-free FRT sites. Aim II: In the next reporting period, we will produce XVE/β-estradiol inducible versions of our viral silencing suppressor constructs and the result of their expression during excision in stable poplar transgenics. We will also produce CRISPR/Cas9-DNA demethylation constructs targeted to the lox sites and their effect on excision rates in stable poplar transgenics. Aim III: In the next reporting period, we will investigate whether enzymatic approaches to methylation sequencing are feasible for targeted analysis over the inserted transgenes, and comparisons of methylation status in viral silencing-suppressor transgenics vs. controls. We will also report any preliminary findings from the greenhouse trials on any phenotypes observed from silencing-suppressor transgenics.

Impacts
What was accomplished under these goals? Aim I (60% complete):We developed a screening system to determine if lox sites withcytosine-free spacer sequences could recombine in vitro. We used recombinant Cre protein to act upon PCR-derived plasmid backbones whose flanking ends contained lox sites with randomized T/A spacers produced by degenerate primers. We recovered hundreds of colonies, with nearly no background in negative controls. We sequenced a population of 100 lox sites produced in vitro for analysis. Of the 100 sequences, most were unique, with only eight spacers repeating twice, and none higher than twice. 25 sites were palindromic and thus of limited utility for use in planta. Three non-palindromic sites which occurred twice were selected for further study alongside three selected at random. Overall, our data suggests that Cre can act upon many cytosine-free spacers without a strong preference in vitro. We attempted to use the same system with FRT sites and FLP recombinase and encountered some challenges. This included a propensity for self-circularization in the absence of recombinase, even in recA- mutant E.coli strains, and some difficulty recovering recombinant FLP protein. We have thus changed our screening approach to avoid degenerate produced sites at the end of PCR fragments and are using negative sacB selection to select against non-FLP-recombinants. Given the permissiveness of Cre in our first screen, we selected FLP sites that were cytosine-free close mutational derivatives of the wtFRT sequence to move forward with testing in planta while the in vitro screening continues to take place. To detect excision of DNA fragments in planta we use an RFP gene (mScarlet3) whose coding sequence is separated by two introns. Within the introns are embedded lox or FRT sites, which concatenate together upon excision. Between the internal introns is a synthetic exon which encodes an inflexible linker, separating the two halves of the fluorescent protein. Thus prior to excision, an improperly folded mScarlet3 protein is produced, and after excision, the protein is correctly folded and red fluorescence is observed. Using this reporter, we have done pilot testing on loxP loxW7 (a preliminary top cytosine-free performer), and FRT through transient agroinfiltration of tobacco leaves. We also performed stable transformation of poplar using a heat-shock inducible Cre acting upon loxP and loxW7. In transient assays, loxP recombines the reporter at approximately double the rate of loxW7, which is equivalent to the rate difference detected from in vitro studies. Nearly all Cre and FRT sites selected for further study are cloned and currently being studied for transient analysis in tobacco and preparing for study in Arabidopsis. Pilot testing with stable transformation in poplar using loxP, loxW7 and heat-shock inducible Cre recombinase revealed premature excision during early transformation and strong toxic effects on event recovery in loxP-containing constructs with this reporter. Because of this, future work on identified cytosine-free spacers will be paired with a more tightly controlled Cre recombinase based on XVE/β-estradiol induction. Recovered pilot transgenic lines from these studies will also be used to troubleshoot our DNA-methylation downstream analysis. Aim II (40% complete): To test anti-silencing tools and their utility for assisting transgene excision we developed constructs overexpressing the viral silencing suppressors P19 and βC1, and artificial microRNAs which target the poplar DRM2 transcript. In poplar stable transformation, both P19 and βC1 overexpression resulted in inhibited shoot regeneration and aberrant effects in transgenic event leaf morphology. DRM2 knock-down events did not have any transformation phenotypes or apparent toxicity but also did not have beneficial excision phenotypes. Events are currently being analyzed for gene expression of DRM2 to determine the level of transcript reduction achieved with this construct. Overall, these initial studies showed that inducible expression of the anti-silencing factors (tied with recombinase expression) is likely ideal. Given βC1 is both a non-specific DNA-binding protein and recruits the DNA-demethylase ROS1 to reduce DNA methylation, we wanted to investigate if a recombinase could be fused to βC1 to inhibit methylation directed at or near the lox sites. We generated C-terminal fusions of Cre to βC1 with a flexible linker and tested their performance in tobacco transient infiltration. Cre- βC1 fusion proteins were fully capable of recombination compared to unfused controls, and had increased reporter signal later after injection. These fusion proteins will be further investigated in stable poplar transformation. In addition to these viral silencing factors, we will generate CRISPR-DNA demethylation constructs to try to inhibit silencing during the excision process. This will include Cas9-TET1 fusion constructs, as previously published, as well as newer versions developed by a collaborating group. Cloning of these constructs is still ongoing, but we expect to test these by early 2025. Aim III (10% complete): In this aim we will focus on analyzing the DNA-methylation outcomes after the use of these tools and any lingering epigenetic effects after transgene excision on the plant genome. We will also be analyzing the phenotypic effects of produced transgenics in greenhouse trials. To manage the DNA-methylation analysis over integrated transgenes, we are in the process of testing kit-based enzymatic methylation sequencing approaches as an alternative to conventional bisulfite sequencing. These include genomic DNA treatment with cytosine deaminases (TET2/APOBEC). The benefit of this approach is the ability to preserve gDNA integrity to enable large amplicon long-read sequencing to cover entire transgenes rather than a smaller panel of short amplicons. Bisulfite treatment damages gDNA into small sizes requiring discrete amplicon panels and thus would give us a limited view into transgene methylation status. We are currently using transgenic events from pilot tests to determine feasibility of this approach. If successful we expect to analyze transgenic products of Aim II by mid-2025. The same initial products are currently being propagated for hardening and planting for greenhouse trial.

Publications

  • Type: Conference Papers and Presentations Status: Accepted Year Published: 2024 Citation: Presentation at DOE workshop: Barriers in plant transformation: A focus on bioenergy crops September 2024. Leveraging oncogenes from a shooty Agrobacterium strain for altruistic transformation
  • Type: Conference Papers and Presentations Status: Accepted Year Published: 2024 Citation: Presentation at Society for in vitro Biology Annual Meeting, June 2024 "Agrobacterium T-DNA genes as tools to promote regeneration of transgenic woody plants https://biotechlab.forestry.oregonstate.edu/sites/default/files/Goralogia_SIVB_2024_Final.pdf
  • Type: Conference Papers and Presentations Status: Accepted Year Published: 2024 Citation: Presentation at IUFRO Tree Biotechnology Meeting, August 2024 Back to the Future: A co-transformation approach with Agrobacterium T-DNA genes promotes regeneration of transgenic woody plants https://biotechlab.forestry.oregonstate.edu/sites/default/files/Goralogia_IUFRO_2024_final.pdf