CRISPR/CAS9 MUTAGENESIS FOR GENETIC CONTAINMENT OF FOREST TREES

• Sponsoring Institution: National Institute of Food and Agriculture
• Project Status: TERMINATED
• Funding Source: OTHER GRANTS
• Reporting Frequency: Annual
• Accession No.: 1013743
• Grant No.: 2017-33522-27098
• Project No.: OREZ-FES-884
• Proposal No.: 2017-03820
• Program Code: HX
• Project Start Date: Sep 1, 2017
• Project End Date: Aug 31, 2021
• Grant Year: 2017

Project Director
Strauss, S. H.

Recipient Organization
OREGON STATE UNIVERSITY
(N/A)
CORVALLIS,OR 97331

Performing Department
Forest Ecosyst & Society

Non Technical Summary
The undesired release of genetic material to the environment from transgenic plants presents a major ecological concern and regulatory barrier to their use. Transgenic trees may provide a valuable resource for wood production, biofuels, bioremediation, and many other applications, however commercial uses are severely limited by regulations and associated ecological and legal risks that are largely the result of transgene dispersal. Efficient, reliable, and bisexually effective methods for genetic containment may be essential to the broad use of vegetatively propagated transgenic trees and other perennial biofuels crops.Stability in expression of a mitigation transgene is critical to its regulatory approval and commercial use. Technologies that do not depend on gene expression--which can be modified by environment, development, and epigenetics--would be preferable for regulatory assurance of stability. Clustered regularly interspaced short palindromic repeats utilizing the Cas9 enzyme (CRISPR-Cas) is a new technology in genetic modification that provides users with the ability to precisely and directly alter the genome. A major advantage of CRISPR-Cas over alternative technologies is the capability to produce permanent structural changes in gene sequences that disable their function. These changes can include the deletion of essential domains or entire parts of genes, and can be identified in juvenile tissues years before floral onset. The loss of both copies of essential parts of a unique gene is unlikely to undergo a repair or reversion. Thus, genetic containment by gene editing should produce stable, predictable, and reliable forms of containment, facilitating regulatory decisions.Our project will focus on testing CRISPR-Cas as a method of genetic engineering in three different areas of interest. First, we will conduct greenhouse and field tests of existing CRISPR-Cas edited poplars and eucalypts to determine their floral and vegetative phenotypes, including their genetic stability and non-target vegetative effects over years in the field. We have chosen the genes leafy (LFY), agamous (AG), tapetal development and function 1 (TDF1), meiotic recombination 8 (REC8), and embryo sac development arrest 33 (EDA33), as well as an overexpressed early flowering gene (FT). We chose LFY for poplar and eucalyptus and AG for poplar because they have shown promising results of floral modification in previous studies using other genetic engineering technologies, and we use the FT gene to induce early flowering so that trees will flower within months instead of having to wait years for trees to flower in the field, providing us with the ability to study floral morphology under greenhouse conditions. We have chosen to use TDF1, REC8, and EDA33 in eucalyptus based off extensive literature searches for promising gene homologs that have affected sterility in Arabidopsis. Our second area of interest will be to examine off-target mutagenesis and chimerism caused by Cas9 nuclease and Cre recombinase at a genomic scale. Finally, third, we will develop and test the efficiency of a system for asexual excision of CRISPR-Cas loci during in vitro plant regeneration.

Animal Health Component

0%

Research Effort Categories

Basic

25%

Applied

75%

Developmental

0%

Classification

Knowledge Area (KA)	Subject of Investigation (SOI)	Field of Science (FOS)	Percent
201	0670	1080	100%

Knowledge Area
201 - Plant Genome, Genetics, and Genetic Mechanisms;

Subject Of Investigation
0670 - Short rotation woody crops, including holiday trees;

Field Of Science
1080 - Genetics;

Keywords

agamous

eucalyptus

gene editing

leafy

populus

recombinase

wuschel

biotechnology

Goals / Objectives
1. Objective 1: Greenhouse study of FT-accelerated flowering background in poplar and Eucalyptus (Year 1)This objective deals with poplar and eucalyptus that have been engineered for early flowering. We will retransform the plants with CRISPR-Cas, taking out various genes of interest that affect sterility or floral morphology in order to study the efficacy, efficiency and other effects the gene knockouts may have on the plants, as well as any effects of the early flowering gene (FT) itself in greenhouse conditions.a. Assess vegetative and floral morphology in Eucalyptus LFY and novel gene constructsWe will study floral morphology, sterility, and vegetative morphology of retransformed FT eucalyptus with CRISPR-induced knockout (KO) events in each of the four target genes: LFY, TDF1, REC8, and EDA33. Studying these genes with an FT background will allow us to study floral morphology quickly under greenhouse conditions and determine if retransformation phenotypes differ from wild type (WT) transformed eucalypts of the same constructs.b. Assess vegetative and floral morphology in poplar LFY and AG constructsWe will start with transformed poplar that have KO mutations in LFY and AG. Using the LFY and AG KO lines, we will retransform the poplars with the FT gene. The FT transgene will allow us to quickly assess floral morphology and sterility, which will be compared to natural flowering in field studies (objective 2). We will study vegetative morphology as well to determine if the presence of FT affects growth significantly by comparing to WT poplars and FT-only events.2. Objective 2: Greenhouse and field study of wild type background poplar and Eucalyptus (Years 1-3)This objective deals with using CRISPR-Cas to transform wild type (not early-flowering) poplar and eucalyptus with the same gene constructs as in objective 1 for greenhouse (GH) as well as field trials. This will allow us to study the effects of CRISPR-Cas without any influence from the FT gene. It will also allow us to assess vegetative and floral phenotypes under natural field conditions on a normal flowering and developmental timeline.a. Assessment of naturally flowering poplarWe will complete greenhouse assessments of vegetative morphology for CRISPR-induced LFY and AG KO events. Trees will then be planted in Oregon for a field trial where they will be allowed to flower naturally. Trees will be assessed for floral and vegetative morphology in the field, sterility, stability, and any off-target effects.b. Assessment of naturally flowering EucalyptusWe will transform wild type Eucalyptus using CRISPR-Cas to target LFY, TDF1, REC8, and EDA33. Once transformed, trees will undergo a GH trial for vegetative phenotype assessment before they are planted in a field trial in Israel where they will be allowed to flower naturally. Trees will be assessed for floral and vegetative morphology in the field, sterility, stability, and any off-target effects.3. Objective 3: Evaluation of off-target effects and chimerism (Years 1-3)We will assess and record any unintended or unusual phenotypic variation in GH and field trials throughout the duration of the project. An example of what we might look for, based on a previous study, is FT eucalypts retransformed using ribonucleic acid interference targeting LFY produced one event with wavy leaves around flowers, when we didn't expect difference in vegetative traits, a rare but unexpected variation. Other variations to look for may include differences in growth, structure, or color. We will also use bait-capture sequencing to look for off-target mutations within the genome. This is done by looking at regions in the genome that are similar to our target region that may have been inadvertently affected by the CRISPR-Cas nuclease.4. Objective 4: Design and test cre-lox CRISPR excision system (Years 2-3)We will assess the tissue-specific removal efficiency of the cre-lox system in the shoot apical meristem. The purpose of doing so is to develop a system that can remove the CRISPR-Cas nucleases from the genome so that they can no longer affect the genome in unintended ways. We will determine the efficiency of this system in two steps. First, we will determine the recombinase activity of the Cre enzyme and the transcriptional activity of the wuschel (WUS) gene promoter by using a green fluorescent protein (GFP) reporter surrounded by the removal sites (lox sites). This will allow us to see where and when the WUS gene is active. We chose to use WUS because it is highly expressed in newly generated shoots, but not in calli or meristematic tissue. Second, we will determine the actual expression level of WUS by retransforming constitutively GFP-expressing poplar tissue with a WUS-Cre fusion protein. If the rates of GFP excision in shoots are satisfactory, we will replace the GFP reporter with a highly active CRISPR Cas9 cassette and we will define the rates of mutagenesis and excision versus just mutagenesis (no Cre present) for the particular Cre-CRISPR combination selected.

Project Methods
1. Objective 1: Greenhouse study of FT-accelerated flowering background in poplar and Eucalyptus (Year 1)a. Assess vegetative and floral morphology in Eucalyptus LFY and novel gene constructsTransformation is completed for LFY constructs and has begun for the three new CRISPR constructs. The greenhouse (GH) study will contain five loss-of-function events each for LFY and the other target floral genes (TDF1, REC8, EDA33), and at least five control events of Cas9 and/or WT. The population will be analyzed for stem growth, height, plant morphology, floral morphology, and fertility.b. Assess vegetative and floral morphology in poplar LFY and AG constructsWe used the construct 35S:FT to retransform five KO events of LFY and AG in poplar in order to induce early flowering, as well as five Cas9-only events and controls. Twenty events will be grown and at least two flowering retransformed events per original KO event will be studied in depth. It will be analyzed similarly to the FT-eucalypt population above except that fertility won't be studied as FT poplar generally does not produce viable pollen/seed.2. Objective 2: Greenhouse and field study of wild type background poplar and Eucalyptus (Years 1-3)a. Naturally flowering poplarWT clones have been transformed with CRISPR-Cas. For each target (two duplicate AG loci and one LFY locus) in each of the male/female clones, we will propagate and plant 5 KO events, 3 heterozygous events, 2 transgenic but not mutated events, 4 Cas9 controls, and 5 sources each of the two types of WT transgenic events. This is in order to determine the roles that LFY and AG play in the floral and vegetative development process of poplar and how stable these gene mutations are throughout the lifetime of the plant. Trees will be grown in the GH and measured for growth and vegetative features. Traits measured could include leaf area, total chlorophyll content, petiole length, floral or vegetative phenology, leaf shape, and leaf specific mass. They will then be assessed in the field for growth, floral and vegetative morphology, within- and among-tree chimerism in vegetative and floral characteristics. We will inspect all plants for floral buds and flowers, and quantify variation in morphology. Floral morphology will be analyzed by noting the presence/absence and number of expected floral whorls. For dioecious poplar clones, trees will be checked for either male or female fertility. Floral phenotypes and fertility will be assayed yearly once flowering begins to assess stability across growing seasons. Floral assessments in the field grown poplars are likely to begin after the three-year grant period.b. Naturally flowering EucalyptusWe used CRISPR-Cas to produce KO mutations of LFY in WT variety hybrid (SP-7) eucalypts and are currently doing the same with TDF1, REC8, and EDA33. We will propagate 5 events that are KOs for each gene, plus 2 Cas9-only events and 5 WT controls each. These will be studied in the GH at OSU for vegetative effects, then field tested in Israel. Once trees begin flowering in field, we will inspect all plants for buds and flowers, and quantify variation in floral morphology. Trees will be assayed for both male and female fertility yearly. Studying trees in the field will allow us to assess any variation in phenotypic expression between naturally flowering trees and GH early-flowering plants.3. Objective 3: Evaluation of off-target effects and chimerism (Years 1-3)We will examine the extent of off-target mutagenesis from CRISPR and Cre-lox induced mutations from GH and field environments. We will collect tissue samples and extract DNA for long-term storage as the experiments are conducted, but genomic analysis will be conducted during year 3. We will use bait-capture sequencing to focus attention on the parts of the genome most likely to suffer unintended mutations. These are areas of the genome that are very similar sequentially to our target site. We plan to sequence a variety of poplar and eucalypt materials representing a wide time frame.4. Objective 4: Design and test Cre-lox CRISPR excision system (Years 2-3)We plan to design and test an excision system using the Cre recombinase enzyme directed by lox removal sites (Cre-lox) that is intended to remove the CRISPR-Cas nucleases from the genome so that they can no longer have any effect. We will first use a GFP marker to study promoter and recombinase activity in poplar, and then test its mutagenesis and excision rate with a CRISPR gene in place of the marker. We will direct expression of the Cre gene by utilizing a WUS-Cre fusion protein. We will then characterize cellular patterns of GFP expression using confocal, fluorescence microscopy. The extent of excision will be assessed using the GFP marker as a visual indicator, as well as by using PCR that will be more sensitive and can also detect chimerism in shoots that do not express GFP.If we find satisfactory rates of excision, we will then test whether the system can provide both high rates of mutagenesis and excision. We will transform an active CRISPR cassette in the absence of the Cre gene, but with the same plasmid otherwise, so that the change in rate of mutagenesis compared to stable CRISPR-Cas overexpression can be determined. For these experiments we will use plant specific Cre gene and lox sites that were provided to us by Pioneer/DuPont, and that has been shown to be highly effective in plants.

Progress 09/01/17 to 08/31/21

Outputs
Target Audience:Our target audience for this reporting period was professors, researchers, postdocs, industry members, students, government regulatory bodies, and the general public. Changes/Problems: Nothing Reported What opportunities for training and professional development has the project provided?Students and postdocs working on this project have attended conferences to give poster presentations, disseminating their research and providing them with valuable networking opportunities.This project has also resulted in three students completing their graduate degrees and advancing their education. How have the results been disseminated to communities of interest?During this reporting period, we have made efforts to reach our targetaudiences through two thesis publications, posters and oral presentations, discussions with industry collaborators, and education activities via OSU classes. What do you plan to do during the next reporting period to accomplish the goals? Nothing Reported

Impacts
What was accomplished under these goals? 1a. In fall 2015, we retransformed two early flowering (FLOWERING LOCUS T (FT)) overexpression events inEucalyptushybridE. grandisxE. urophylla(clone SP7) with three different CRISPR Cas9 constructs targeting LEAFY (LFY). This work is summarized in Elorriaga et al. 2021 (PMID: 33774917). Overall, we found lfy loss-of function mutants are completely sterile in fast-flowering eucalypt transgenics, with no significant effects on growth. Using the same two FT-only events inEucalyptushybridE. grandisxE. urophylla(clone SP7), we retransformed eucalypt tissue with three CRISPR Cas9 constructs targeting EMBRYO SAC DEVELOPMENT ARREST 33 (EDA33), MEIOTIC RECOMBINATION 8 (REC8),andTAPETAL DEVELOPMENT FUNCTION 1 (TDF1). We generated and studied multiple verified loss of function events per construct, and assessed their impact on floral development, pollen viability, and pollen germination. tdf1 loss of function mutants were, as expected from Arabidopsis literature, completely male sterile with no viable pollen produced. eda33 knockouts were predicted to produce viable fruits with defective valve margin senescence resulting in female sterility. The eda33 knockouts were male sterile, with no pollen produced, and had shorter styles at late stages of floral development compared to controls. We were unable to obtain fruits with seed set in the greenhouse with multiple controlled crosses using E. grandis pollen obtained from a seed orchard in South Africa to assess the fruit indehiscence phenotype in eda33 knockouts. rec8 loss of function mutant flowers produced WT levels of discrete pollen grains which were severely misshapen and had 0% viability. The flowers had normal morphology and produced nectar. Summaries of the eucalyptus mutant analysis can be found in the following presentations: https://www.dropbox.com/s/736y0qxov6qw5iz/Sonali%203-min%20undergrad%20symposium%20final%20presentation%202021.pdf?dl=0 https://www.dropbox.com/s/k8sgtengc6owhco/Daniel%20URSA%202021%203%20min%20revised.pdf?dl=0 1b. We retransformed the heat inducible construct (HSP::AtFT) into selected 717 and 353 KO mutants as this construct gave us normal and frequent flowering after three weeks of heat induction in previous experiments. We selected 36 FT subevents events withLFYmodified, 33 FT subevents withAGmodified, 13 FT subevents with CAS9 only (i.e., empty vector control) events, and one FT-only events, with clonal replication (ramets). We supplied heat shock conditions in two block-replicated greenhouses, and 34% of 717 and 19% of 353 flowered. All ramets with KO mutation inLFYand inAGhad sterile flower-like organs. No evidence of ovules or stigmas was seen. lfy loss-of-function mutants still formed inflorescences where catkins would have formed, but indeterminate vegetative structures took their place. Ag mutants in male and female poplar clonal backgrounds were distinctive, both produced petaloid/sepaloid indeterminate flowers with some sterile carpeloid structures. 2a. We completed the GH assessments of vegetative morphology for CRISPR-inducedLFYandAGKO poplar events prior to the start of this proposal. In November 2017 we planted a field trial consisting of 136 trees from clone 353 and 180 trees from clone 717. After initial replacement, only 2 trees have died. Growth and survival data was collected during winter 2019, 2020, and 2021. Trees were singled in 2019. SPAD data, leaf collections, and dbh/height collections were conducted on this trial in summer 2021 and have yet to be analyzed for volume index, leaf weight, or leaf area. 2b. The GH trial of wild-type (i.e., not early flowering) eucalypt SP7 targetingLFYis now has concluded. These results were published in Elorriaga et al., 2021. Additionally, we completed our greenhouse vegetative analysis of nine predicted knock-out events in EDA33, seven in REC8, and five in TDF1. We also selected ten Cas9 only control events. We found no significant effect ongrowth, leaf traits, or oil duct morphology in edited knockout mutants compared to controls (i.e., Cas9 events). Please refer to the below products: https://ir.library.oregonstate.edu/downloads/8k71nq237 https://isbr.info/symposium-welcome 3. We received sequencing information from 96 individual hybrid poplar (field grown) and eucalyptus (GH) ramets corresponding toLFYandAGKOs, heterozygous mutants, transgenic non-mutants, and WT controls. We used a targeted sequencing approach using 20,000 designed RNA probe baits. These baits were designed to be centered on an off-target sgRNA binding site with up to 5bp of mismatch to the target sequence in the respective reference genome. Each sample was sequenced to 2GB on a nova-seq Illumina platform. Coverage of ~100-1200x was achieved around probe sites. We began analysis of off-target CRISPR/Cas9 mutations and novel somatic mutations accumulated during the transformation/regeneration process using the program Mutect2 https://gatk.broadinstitute.org/hc/en-us/articles/360037593851-Mutect2. Analysis of CRISPR/Cas9 on-target, off-target, and novel somatic mutations is still ongoing. For clones 717 and 353, we found off-target mutations likely induced by Cas9 at 3-4 bp of mismatch to the perfect target site, at lower frequency. Off-target mutations within potential off-target windows were 62% in overall abundance compared to novel somatic mutations within the same windows in the 717 clone. This evidence suggests that off-target mutagenesis israre compared with novel somatic variation in vegetatively propagated plants. All off-target mutations were found within the AG targeting constructs, suggesting gRNA design contributes most to off-target rates. Mutect2 identification of true novel mutations ranged from 5-80% depending on model parameter inputs and construct type, showing an unsurprising balance between sensitivity and accuracy. We will receive a completed reference genome for the SP7 eucalyptus clone for better assessment of off-target mutagenesis in our Eucalyptus leafy CRISPR/Cas9 population. 4a. We constructed and transformed 717 poplars with an editing an excision construct which included a shoot meristem specific promoter (AtCSP3) driven CRE recombinase gene fused to the rat glucocorticoid receptor (GR) (to enable cell specific and chemical inducible gene expression), and sgRNAs targetingLEAFY (LFY) and GIBERELLIC ACID INSENSITIVE (GAI)gene to produce sterile, semi-dwarf poplars. We produced 5 edited events with deletion events in GAI. 3/5 events showed strong dwarf phenotypes, one showed a semi-dwarf habit, and one was a loss-of-function event for GAI with putatively normal growth habit. All events were biallelic loss-of-function mutants for LFY. One of these events, with normal growth habit and lfy mutant was completely excised, with a 252bp footprint remaining in the genome consisting of T-DNA borders, a LoxP site and non-coding vector DNA. Other dwarf and semi-dwarf events were unable to excise after repeated dexamethasone treatments. The results of these experiments are summarized in the following presentations: https://people.forestry.oregonstate.edu/steve-strauss/sites/people.forestry.oregonstate.edu.steve-strauss/files/StraussASPB2021.pdf, https://people.forestry.oregonstate.edu/steve-strauss/sites/people.forestry.oregonstate.edu.steve-strauss/files/GoralogiaASPB_POSTER2021pdf.pdf. 4b. After screening 13 Arabidopsis or Populus trichocarpa promoters for regenerating shoot meristem activity using promoter:GFP transgenic constructs, we assayed SHOOT MERISTEMLESS, WUSCHEL, and COLD SHOCK PROTEIN 3 in detailed experiments using stable transgenic events as source material in regeneration assays. All developmental promoters investigated had undesirable expression in callus tissue during regeneration. CSP3 had the most consistent shoot meristem expression among transgenic events, and was selected for further construct development. WUSCHEL had the least consistent expression.

Publications

• Type: Journal Articles Status: Published Year Published: 2021 Citation: Elorriaga, E., Klocko, A. L., Ma, C., Marc du Plessis, An, X., Myburg, A. A. and Strauss, S. H. (2021) Genetic containment in vegetatively propagated forest trees: CRISPR disruption of LEAFY function in Eucalyptus gives sterile indeterminate inflorescences and normal juvenile development. Plant Biotechnol J., https://doi.org/10.1111/pbi.13588
• Type: Journal Articles Status: Published Year Published: 2021 Citation: Klocko, A.L.; Goddard, A.L.; Jacobson, J.R.; Magnuson, A.C.; Strauss, S.H. RNAi Suppression of LEAFY Gives Stable Floral Sterility, and Reduced Growth Rate and Leaf Size, in Field-Grown Poplars. Plants 2021, 10, 1594. https://doi.org/10.3390/plants10081594
• Type: Journal Articles Status: Published Year Published: 2021 Citation: Goralogia, G.S., Redick, T.P. & Strauss, S.H. Gene editing in tree and clonal crops: progress and challenges. In Vitro Cell.Dev.Biol.-Plant 57, 683699 (2021). https://doi.org/10.1007/s11627-021-10197-x
• Type: Journal Articles Status: Other Year Published: 2021 Citation: CRISPR knock-outs of the poplar orthologs of AGAMOUS and LEAFY in rapid-flowering trees give sterile indeterminate floral organs. AL Klocko, E. Elorriaga, C. Ma, and SH Strauss. Manuscript being prepared.

Progress 09/01/19 to 08/31/20

Outputs
Target Audience:Our target audience for this reporting period was professors, researchers, postdocs, industry members, students, government regulatory bodies, and the general public. Changes/Problems: 4. Objective 4: Design and test cre-lox CRISPR excision system (Years 2-3) (60% completion) a. Construction of a CRISPR excision vector system Under objective 4a, we sought to create a WUS-cre protein fusion to increase abundance throughout SAM cells. To control the activity of cre protein in the plant we instead opted to add the glucocorticoid receptor peptide fused C-terminally to cre, which controls entry of the protein into the nucleus upon detection of the hormone dexamethasone. To avoid complications with recombinase function we will not be testing WUS peptide fusions at present, as we believe our promoter fragments should provide an adequate percentage of SAM cells to reach fixation for excision of the transgene. During the process of developing the gene editing and transgene excision vector system, we made several modifications due to unforeseen issues. Upon vector assembly we observed complete T-DNA excision inE. coli. To fix this issue, we designed, had synthesized and constructed a cre recombinase with an intron from the ArabidopsisCOLD REGULATED 15A (COR15A)gene. Some introns may unintentionally increase or decrease gene expression; to control for this we choseCOR15Aas it is known to have no impact on expression. To add additional control in the system, we also fused cre c-terminally with the rat glucocorticoid receptor peptide (GR).This limits cre activity to when the dexamethasone hormone is applied. Upon initial testing, we determined thepUBQ10:GFP-Cas9construct was not easily visible in transformed explants under an epifluorescent dissecting microscope. To solve this problem, we modified thepUB10:GFP-Cas9gene to instead produce two separate peptides via a peptide cleavage domain fromThosea asignavirus 2A (T2A). This allows GFP to be more easily visible in multiple cell compartments. Objective 2. Greenhouse and field study of wild type background poplar andEucalyptus b. Assessment of naturally floweringEucalyptus Under objective 2b, the originally planned field trial of these natural (not early flowering) CRISPR eucalypts in Israel will not go forward as our collaborators there have decided they are unable to support it. We no longer plan to test CRISPR mutated but naturally flowering Eucalyptus due to the lack of a suitable partner willing to conduct field trials with flowering CRISPR-transgenic trees. What opportunities for training and professional development has the project provided?Students and postdocs working on this project have attended conferences to give poster presentations, disseminating their research and providing them with valuable networking opportunities. This project has also resulted in three students completing their graduate degrees and advancing their education. How have the results been disseminated to communities of interest?During this reporting period, we have made efforts to reach our targetaudiences through two thesis publications, posters and oral presentations, discussions with industry collaborators, and education activities via OSU classes. What do you plan to do during the next reporting period to accomplish the goals? Objective 1. Greenhouse study of FT-accelerated flowering background in poplar andEucalyptus(Year 1) a. Assess vegetative and floral morphology inEucalyptusLFYand novel gene constructs We are currently taking what we hope are near final data on floral morphology and fertility, as well as hand pollinating flowers in hope of obtaining fruits for assessment of seed formation and capsule dehiscence. These results should be complete and a manuscript prepared by summer 2021. b. Assess vegetative and floral morphology in poplarLFYandAGconstructs Data on floral and vegetative morphology are nearly complete and a manuscript is nearly ready for preparation. Objective 2. Greenhouse and field study of wild type background poplar andEucalyptus(Years 1-3) a. Assessment of naturally flowering poplar Statistical analysis of growth based on size through the growing season in 2020 are nearly complete and will be part of the manuscript on floral sterility in poplar cited above. Monitor, control weeds, and care for CRISPR-inducedLFYandAGKO plants growing in the field. Take/analyze leaf vegetative data after summer 2021. Monitor flowering in the field and assess floral morphology and fertility once it begins. b. Assessment of naturally floweringEucalyptus This objective was abandoned as our collaborator has declined to do the field trial for regulatory and political reasons. Objective 3. Evaluation of off-target effects and chimerism (Years 1-3) We are currently analyzing the sequencing data to identify off-target mutations, identify novel somaclonal mutations, and determine T-DNA integration locations and copy number. Analyze the returned results of the bait-capture sequencing experiment by aligning the reads and identifying variants. Identify likely variants present in individual trees producedde novofrom their parental clone, PCR amplify those regions and perform Sanger sequencing to confirm variants. Estimate the mutation rate and spectra produced by different CRISPR/Cas9 constructs and rate of new somaclonal mutations induced by propagation and/or transformation to compare. From sequence data, characterize the T-DNA insertion location and copy number in independent transgenic events. Objective 4. Design and test cre-lox CRISPR excision system (Years 2-3) We are in the process of re-regenerating transgenic explants on chemical induction media to obtain transgene free, edited explants for analysis. Because of current excision system shortcomings, we built another vector platform using a chemically induced transactivation system based on the human estrogen receptor, and will be comparing the excision rates of this system. Test the suitability of SAM promoters during a second round of regeneration on confirmed transgenic lines coupled with our scoring system to assess expression domain and GFP signal intensity, and choose the three best for further analysis. Determine the editing rate from excision constructs by using sgRNAs which target thePHYTOENE DESATURASE (PDS)gene for mutation, resulting in a chlorotic shoot phenotype. Assess the rate of editing and transgene excision by comparing the GFP signal lost in transgenic shoots against the proportion of chlorotic shoots induced during transformation, and observe any chimerism through remaining GFP signal. Characterize 5 successfully excised events to validate complete fixation for excised transgene, insertion site(s) using TAIL-PCR, and the remaining inert pieces of the T-DNA still present in the plant genome.

Impacts
What was accomplished under these goals? 1a. In fall 2015, we retransformed two early flowering (FLOWERING LOCUS T (FT)) overexpression events inEucalyptushybridE. grandisxE. urophylla(clone SP7) with three different CRISPR Cas9 constructs targeting LEAFY (LFY). We generated 59 FT CRISPR Cas9 insertion events, 10 FT Cas9 events, one FT escape event, and we had the original two FT-only events. In January 2018 we selected several of the events and began a greenhouse trial to evaluate flower morphology. In the GH trial we had 42FT insertion events, 6 Cas9FTcontrol events, 2FTonly (early flowering background but not retransformed) flowering control events, 2 escapes (i.e. alsoFTflowering controls but underwent transformation process), and wild type (WT) controls. Of the 42 FT insertion events, 38 had predicted KO mutations on both alleles. However, only 32 had at least one flowering ramet (i.e., four events had no flowering ramets and their form was similar to that of WT ramets). The four insertion events that did not have KO mutations, had instead in frame mutations (i.e., deletion or insertion) in at least one allele. These four insertion events had flowers that resembled those from the FT controls (i.e., FT Cas9, FT escape, and Ft-only events). We observed altered floral phenotypes in the32 FT insertion events with KO mutations. The altered flowers appeared to have lost determinacy. The phenotype of the altered flowers ranged from underdeveloped bisexual floral-like structures with two to three repeated layers of bract-like and pedicel-like organs with underdeveloped (i.e., sterile) stamens and underdeveloped gynoecia to bud-like structures with many repeated layers of bracts and pedicels with no reproductive organs at all. We performed gene expression analysis on select genes directly downstream and upstream (or at the same level) ofLFYwith qPCR. We found that genes upstream or at the same developmental stage asLFY(i.e.,FT, SPL3, SPL9, CAL, FUL1,andFUL2) had dramatically higher expression in the KOs than in the flowering controls.For genes downstream ofLFY(i.e.,AP3, PI, AG, SHP,andSTK),we saw the complete opposite. A manuscript was submitted to New Phytologist and is currently under review. Using the same two FT-only events inEucalyptushybridE. grandisxE. urophylla(clone SP7), we retransformed eucalypt tissue with three CRISPR Cas9 constructs targeting EMBRYO SAC DEVELOPMENT ARREST 33 (EDA33), MEIOTIC RECOMBINATION 8 (REC8),andTAPETAL DEVELOPMENT FUNCTION 1 (TDF1). We obtained 43 PCR positive retransformed events: 14EDA33,10REC8, 11 TDF1,and 8CAS9. In May 2020, we selected insertion events with mutations in both alleles, specifically six targeting EDA33, ten targeting REC8, and six targeting TDF1. We also generated the following flowering controls: eight FT Cas9 events, two FT escapes, and the two original early flowering lines. Evaluation of effects on floral and vegetative morphology, fertility, and pollen viability are underway. 1b. We retransformed the heat inducible construct (HSP::AtFT) into selected 717 and 353 KO mutants as this construct gave us normal and frequent flowering after three weeks of heat induction in previous experiments. For the heat induction experiment, we have 2,295 explants in 717 from threeLFYKOs, twoAGKOs, and two CAS9 control events transformed with HSP::AtFT; 212 FT subevents were obtained and 182 events have been propagated for the study. We selected 170 ramets corresponding to 36 FT subevents events withLFYmodified, 33 FT subevents withAGmodified, 13 FT subevents with just CAS9 (i.e., empty vector control) events, and one FT-only events. We heated the glasshouses to 42°C for two months for a total of eight hours per day from 9:30am to 5:30pm. 34% of ramets flowered and 3% died. Flowers were imaged with an SLR camera and also with a Keyence Digital microscope. All of the ramets with KO mutation inLFYand inAGhad sterile flower-like organs. No evidence of ovules or stigmas were seen. In 353 we have 2,543 explants from twoLFYKO events, twoAGKO events, and two CAS9 control events that have been co-cultivated with HSP::AtFT; 256 FT subevents were obtained and 120 subevents were propagated for the heat shock experiment. We selected 2 ramets for each subevent and 10 ramets corresponding to an HSP:AtFT event. We heated the glasshouses to 42°C between 8AM and 12PM and between 4PM and 9PM for two months. 19% of ramets flowered and only one ramet (0.4%) died. Flowers were imaged with an SLR camera and a Keyence Digital microscope. All of the ramets with KO mutation in LFY and in AG had sterile altered reproductive structures. No evidence of stamens was seen. 2a. We completed the GH assessments of vegetative morphology for CRISPR-inducedLFYandAGKO poplar events prior to the start of this proposal. In November 2017 we planted a field trial consisting of 136 trees from clone 353 and 180 trees from clone 717. After initial replacement, survival is at 100%. Growth and survival data (height, DBH) was collected in 2019 and 2020. Trees were singled to promote growth of the leader stem in summer 2019. 2b. The GH trial of wild-type (i.e., not early flowering) eucalypt SP7 targetingLFYis now concluding after six months of analysis. No significant differences in vegetative growth were found. We transformed 1,515 leaf explants of wild type SP7 withEDA33, REC8, TDF1, and an empty control vector containing CAS9. We have obtained 58 PCR positive events: 18EDA33,11REC8, 18 TDF1,and 11CAS9. We selected nine predicted knock-out events in EDA33, seven in REC8, and five in TDF1.We also selected ten Cas9 events. We found no evidence that eliminating gene function of the three target genes (i.e., EDA33, REC8, and TDF1) would affect growth or leaf traits when comparing to the empty vector controls (i.e., Cas9 events). 3. We received sequencing information from 96 individual hybrid poplar (field grown) and eucalyptus (GH) ramets corresponding toLFYandAGKOs, heterozygous mutants, transgenic non-mutants, and WT controls. We used a targeted sequencing approach using 20,000 designed RNA probe baits. They were designed to be centered on an off-target sgRNA binding site with up to 5bp of mismatch to the target sequence in the respective reference genome. Each sample was sequenced to 2GB on a nova-seq Illumina platform. Coverage of ~100-1200x was achieved around probe sites. We are currently analyzing the sequencing data to identify off-target mutations, identify novel somaclonal mutations, and determine T-DNA integration locations and copy number. 4a. We constructed and transformed 717 poplars with an editing an excision construct which included a shoot meristem specific promoter driven CRE recombinase gene fused to the rat glucocorticoid receptor (GR) (to enable cell specific and chemical inducible gene expression), and sgRNAs targetingPHYTOENE DESATURASE (PDS)gene. This gene target was chosen to generate chlorotic phenotypes and visually score editing efficiency. We obtained a 26% editing rate on a per explant basis. 8% of edited shoots showed a dramatic loss of GFP signal, however subsequent PCR analysis showed substantial chimerism in shoots which scored GFP positive and negative. 4b. We transformed Arabidopsis andPopulus trichocarpapromoter regions of theSHOOT MERISTEMLESS (STM), WUSCHEL (WUS), ERECTA (ER), COLD SHOCK PROTEIN 3 (CSP3),andENHANCER OF SHOOT REGENERATION 1 (ESR1)genes fused to GFP into thePopulus tremula x alba717-1B4 clone. We PCR confirmed at least 10 events for all constructs. We developed a scoring system for assaying GFP expression during regeneration, and are assessing promoter utility by regenerating confirmed transgenics. Arabidopsis CSP3 (1.3kb fragment) and STM promoters (3.6kb fragment) show the highest and most shoot meristem specific expression in poplar, so we included these promoters in our gene editing and excision constructs. Expression analyses will be complete within the next two months and a manuscript prepared for publication.

Publications

• Type: Theses/Dissertations Status: Published Year Published: 2020 Citation: Nahata S. 2020. Sequence Analysis and Vegetative Growth in Transgenic Eucalyptus with CRISPR-Cas9-induced Mutations in the Eucalyptus Homologs of the Floral Genes EMBRYO DEVELOPMENT ARREST 33 (EDA33) and TAPETAL DEVELOPMENT AND FUNCTION1 (TDF1). Environmental Sciences, Oregon State University. M.S.
• Type: Theses/Dissertations Status: Published Year Published: 2020 Citation: Elorriaga E. 2020. Functional Characterization and Classification of Genes Essential to Flower Induction, Flower Development, and Seed Development in Populus and Eucalyptus. Molecular and Cellular Biology, Oregon State University. Ph.D.
• Type: Journal Articles Status: Submitted Year Published: 2020 Citation: Elorriaga E, Klocko A, Ma C, du Plessis M, An X, Myburg A, and Strauss SH. 2020. CRISPR disruption of LEAFY function in Eucalyptus gives sterile indeterminate inflorescences and normal juvenile development. New Phytologist, submitted.

Progress 09/01/18 to 08/31/19

Outputs
Target Audience:Our target audience for this reporting period was professors, researchers, postdocs, industry members, students, government regulatory bodies, and the general public. Changes/Problems: 4. Objective 4: Design and test cre-lox CRISPR excision system (Years 2-3) a. Construction of a CRISPR excision vector system Under objective 4a, we sought to create a WUS-cre protein fusion to increase abundance throughout SAM cells. To control the activity of cre protein in the plant we instead opted to add the glucocorticoid receptor peptide fused C-terminally to cre, which controls entry of the protein into the nucleus upon detection of the hormone dexamethasone. To avoid complications with recombinase function we will not be testing WUS peptide fusions at present, as we believe our promoter fragments should provide an adequate percentage of SAM cells to reach fixation for excision of the transgene. During the process of developing the gene editing and transgene excision vector system, we made several modifications due to unforeseen issues. Upon vector assembly we observed complete T-DNA excision in E. coli. To fix this issue, we designed, had synthesized and constructed a cre recombinase with an intron from the Arabidopsis COLD REGULATED 15A (COR15A) gene. Some introns may unintentionally increase or decrease gene expression; to control for this we chose COR15A as it is known to have no impact on expression. To add additional control in the system, we also fused cre c-terminally with the rat glucocorticoid receptor peptide (GR).This limits cre activity to when the dexamethasone hormone is applied. Upon initial testing, we determined the pUBQ10:GFP-Cas9 construct was not easily visible in transformed explants under an epifluorescent dissecting microscope. To solve this problem, we modified the pUB10:GFP-Cas9 gene to instead produce two separate peptides via a peptide cleavage domain from Thosea asigna virus 2A (T2A). This allows GFP to be more easily visible in multiple cell compartments. Objective 2. Greenhouse and field study of wild type background poplar and Eucalyptus b. Assessment of naturally flowering Eucalyptus Under objective 2b, the originally planned field trial of these natural (not early flowering) CRISPR eucalypts in Israel will not go forward as our collaborators there have decided they are unable to support it. We no longer plan to test CRISPR mutated but naturally flowering Eucalyptus due to the lack of a suitable partner willing to conduct field trials with flowering CRISPR-transgenic trees. What opportunities for training and professional development has the project provided?Students and postdocs working on this project have attended conferences to give poster presentations, disseminating their research and providing them with valuable networking opportunities.? How have the results been disseminated to communities of interest?During this reporting period, we have made efforts to reach our targetaudiences through one thesis publication, posters and oral presentations, discussions on industry collaborations, and other outreach activities. What do you plan to do during the next reporting period to accomplish the goals?Objective 1. Greenhouse study of FT-accelerated flowering background in poplar and Eucalyptus (Year 1) a. Assess vegetative and floral morphology in Eucalyptus LFY and novel gene constructs REC8, TDF1, EDA33 mutant and CAS9 control events have been selected and are now being readied for a greenhouse trial to start in winter. Screen mutants and propagate KO events for GH studies Once events are selected, begin GH trial of novel gene targets in FT SP7 and take data on vegetative and floral morphology to assess the effects of the genes Continue to analyze gene expression of genes upstream of LFY and we will finalize analysis on genes downstream of LFY b. Assess vegetative and floral morphology in poplar LFY and AG constructs Continue to conduct the greenhouse study of 353 KOs targeting LFY and AG containing the FT transgene background Begin heat induction experiment with 717 KOs containing FT gene and monitor floral development/phenotype We will continue with PCR confirmation of the 353 KOs with HSP::AtFT, propagate positive FT subevents, and prepare for heat induction experiments in the greenhouse Analyze results of heat induction experiment for both 717 and 353 once concluded Objective 2. Greenhouse and field study of wild type background poplar and Eucalyptus (Years 1-3) a. Assessment of naturally flowering poplar Monitor and care for CRISPR-induced LFY and AG KO plants growing in the field Collect growth data (height, diameter at 6" above soil level) during winter 2019, and take/analyze leaf vegetative data in summer 2020 once leaves emerge and reach maturity Trees were singled to promote growth of the leader stem in summer 2019, and will be singled in summer 2020 Monitor for flowering in the field, although we don't expect flowering to begin for at least another year b. Assessment of naturally flowering Eucalyptus Complete mutants screening of SP7 with EDA33 and TDF1 Propagate SP7 KOs and conduct greenhouse assessment Take GH data on vegetative phenotypes and analyze to determine if novel gene targets affect vegetative traits compared to Cas9 and WT controls Objective 3. Evaluation of off-target effects and chimerism (Years 1-3) Analyze the returned results of the bait-capture sequencing experiment by aligning the reads and identifying variants Identify likely variants present in individual trees produced de novo from their parental clone, PCR amplify those regions and perform sanger sequencing to confirm variants Estimate the mutation rate and spectra produced by different CRISPR/Cas9 constructs and rate of new somaclonal mutations induced by propagation and/or transformation to compare From sequence data, characterize the T-DNA insertion location and copy number in independent transgenic events Objective 4. Design and test cre-lox CRISPR excision system (Years 2-3) Test the suitability of SAM promoters during a second round of regeneration on confirmed transgenic lines coupled with our scoring system to assess expression domain and GFP signal intensity, and choose the three best for further analysis Determine the editing rate from excision constructs by using sgRNAs which target the PHYTOENE DESATURASE (PDS) gene for mutation, resulting in a chlorotic shoot phenotype Assess the rate of editing and transgene excision by comparing the GFP signal lost in transgenic shoots against the proportion of chlorotic shoots induced during transformation, and observe any chimerism through remaining GFP signal Characterize 5 successfully excised events to validate complete fixation for excised transgene, insertion site(s) using TAIL-PCR, and the remaining inert pieces of the T-DNA still present in the plant genome

Impacts
What was accomplished under these goals? Objective 1. Greenhouse study of FT-accelerated flowering background in poplar and Eucalyptus (Year 1) - 90% completion a. Assess vegetative and floral morphology in Eucalyptus LFY and novel gene constructs Using two early flowering (FLOWERING LOCUS T (FT)) overexpression events in Eucalyptus hybrid E. grandis x E. urophylla (clone SP7), we co-cultivated 1,346 leaf explants with novel CRISPR constructs targeting EMBRYO SAC DEVELOPMENT ARREST 33 (EDA33), MEIOTIC RECOMBINATION 8 (REC8), and TAPETAL DEVELOPMENT FUNCTION 1 (TDF1). So far we have obtained 21 PCR positive events (6 EDA33, 6 REC8, 3 TDF1, and 6 CAS9), which are being readied for a greenhouse trial to begin in winter. In January 2018 we began the GH trial for FT SP7 targeting LEAFY (LFY) with 38 LFY KO events, 6 Cas9 FT control events, 2 FT only (early flowering background but not retransformed) flowering control events, 2 escapes (i.e. also FT flowering controls but underwent transformation process), and wild type (WT) controls. We observed altered floral phenotypes in the LFY KO mutants (27 of 36 predicted KO events had the altered phenotype, 3 had WT flowers, and 6 produced no flower buds). The altered flowers appeared as if they had diminished floral determinacy. no floral organs or unorganized floral tissue. We performed gene expression analysis on select genes directly downstream and upstream of LFY with qPCR. We found that genes upstream or at the same developmental stage as LFY (i.e., FT, SPL3, SPL9, CAL, FUL1, and FUL2) had dramatically higher expression in the KOs than in the flowering controls. For genes downstream of LFY (i.e., AP3, PI, AG, SHP, and STK), we saw the complete oppositeA manuscript has beeing readied for submission. b. Assess vegetative and floral morphology in poplar LFY and AG constructs We retransformed the heat inducible construct (HSP::AtFT) into selected 717 and 353 KO mutants as this construct gave us normal and frequent flowering after three weeks of heat induction in previous experiments. For the heat induction experiment, we have 2,295 explants in 717 from three LFY KOs, two AG KOs, and two CAS9 control events transformed with HSP::AtFT; 212 FT subevents were obtained and 182 events have been propagated for the study. We selected 170 ramets corresponding to 36 FT subevents events with LFY modified, 33 FT subevents with AG modified, 13 FT subevents with just CAS9 (i.e., empty vector control) events, and one FT-only events. We heated the glasshouses to 42°C for two months for a total of eight hours per day from 9:30am to 5:30pm. 34% of ramets flowered and 3% died. Flowers were imaged with an SLR camera and also with a Keyence Digital microscope. All of the ramets with KO mutation in LFY and in AG had sterile flower-like organs. No evidence of ovules or stigmas were seen. In 353 we have 2,543 explants from two LFY KO events, two AG KO events, and two CAS9 control events that have been co-cultivated with HSP::AtFT; selection and PCR confirmation to obtain events for the study are underway. The heat shock study will begin in January 2020. Objective 2. Greenhouse and field study of wild type background poplar and Eucalyptus (Years 1-3) - 75% completion a. Assessment of naturally flowering poplar We completed the GH assessments of vegetative morphology for CRISPR-induced LFY and AG KO poplar events prior to the start of this proposal. In November 2017 we planted a field trial consisting of 136 trees from clone 353 and 180 trees from clone 717. Trees were scored for survival and we found that the 353s had a 93.4% survival rate and the 717s had a 93.3% survival rate. Dead trees were replaced with trees of the same construct and event (when possible), and those trees will be monitored closely to ensure they survive through the winter. b. Assessment of naturally flowering Eucalyptus We transformed 1,515 leaf explants of wild type SP7 with EDA33, REC8, TDF1, and an empty control vector containing CAS9. We have obtained 58 PCR positive events (18 EDA33, 11 REC8, 18 TDF1, and 11 CAS9). Among 11 REC8 events, 9 of them are mutants confirmed by mutant screening and sequencing. There are 7 selected are biallelic REC8 KO events, and 10 Cas9 only events plus WT controls. The GH studies for REC, EDA33 and TDF1 are now done and data are being analyzed. The GH trial of eucalypt SP7 targeting LFY is now concluding after six months of analysis. No significant differences in vegetative growth were found. Objective 3. Evaluation of off-target effects and chimerism (Years 1-3) - 20% completion We harvested DNA from 96 individual hybrid poplar (field grown) and eucalyptus (GH) ramets corresponding to LFY and AG KOs, heterozygous mutants, transgenic non-mutants, and WT controls. We have contracted with Arbor Biosciences LLC for a bait-capture and sequencing approach. For bait target sequences, we have provided 20,000 genomic loci that correspond to potential off-target binding sites. After bait capture, samples will be sequenced on an Illumina NovaSeq 6000 platform and sequenced to 1Gbp depth per sample. Our samples have been submitted and our bait-capture probes are currently being synthesized. Objective 4: Design and test cre-lox CRISPR excision system (Years 2-3) - 60% completion a. Construction of a CRISPR excision vector system We constructed an easily adaptable vector platform for gene editing and transgene excision by modifying existing gene editing vectors. Our new system uses golden gate assembly, with independent module vectors containing Cas9, sgRNAs, and site-specific recombinases.To make our plant transformation backbones work in this system, we added recognition sequences for the cre and FLP recombinases on separate plasmids. We eliminated 35S promoters from any individual expression units to ensure recombinase expression would not be influenced by neighboring genes. These were replaced by Arabidopsis UBIQUITIN 10 (UBQ10) or NOS promoters. We made 2 versions of each backbone, one with the recombinase recognition sites flanking all internal genes, and one that retained the plant resistance marker after excision.To provide a visual marker for transgene incorporation, we added an N-terminal green fluorescent protein (GFP) fusion to Cas9. We also modified the recombinase cassette to make developmental promoter incorporation simple and cloned in cre and FLP recombinases. In summary, we have developed an easily adaptable system that allows transgenic plants to be identified visually, and allows us to witness transgene excision via loss of GFP. We are coupling this with sgRNAs targeting PHYTOENE DESATURASE (PDS) to generate chlorotic phenotypes and visually score editing efficiency. We are currently transforming Populus tremula x alba 717-1B4 to assess GFP signal and editing rates. b. Determining suitable developmental promoters to drive the expression of recombinase during regeneration To determine the best promoters to drive recombinase expression in the shoot apical meristem (SAM) during regeneration, we amplified Arabidopsis and Populus trichocarpa promoter regions of the SHOOT MERISTEMLESS (STM), WUSCHEL (WUS), ERECTA (ER), YAOZHE (YAO), COLD SHOCK PROTEIN 3 (CSP3), and ENHANCER OF SHOOT REGENERATION 1 (ESR1) genes, which have been reported to have strong expression at the SAM, and made transcriptional fusions of these promoters to GFP. We have successfully cloned all promoter fragments and have transformed 10/12 constructs into the Populus tremula x alba 717-1B4 clone. The YAO and ESR1 constructs did not show visible GFP signal and were excluded from future analysis. We have PCR confirmed at least 10 events for 3/10 constructs, and are processing the remainder. We developed a scoring system for assaying GFP expression during regeneration, and will be assessing promoter utility by regenerating confirmed transgenics. Initial studies during transformation implicate STM, WUS, and CSP3 as viable promoters for use in the excision system.

Publications

• Type: Theses/Dissertations Status: Published Year Published: 2019 Citation: Lu H. 2019. Identification and Functional Characterization of Genes that Regulate Biomass Production, Floral Development, and Leaf Senescence in Populus Species using Genetic Engineering and Bioinformatics Tools. Oregon State University. Ph.D. https://ir.library.oregonstate.edu/concern/graduate_thesis_or_dissertations/nk322k610

Progress 09/01/17 to 08/31/18

Outputs
Target Audience:Our target audience for this reporting period was professors, researchers, postdocs, industry members, students, government regulatory bodies, and the general public. We have made efforts to reach these audiences through one peer-reviewed publication, posters and oral presentations, discussions on industry collaborations, and other outreach activities. Changes/Problems:Under objective 2b, we were originally planning on initiating a field trial with our collaborators in Israel of WT-background Eucalyptus transformed with LFY, REC8, EDA33, and TDF1 in order to assess natural flowering of the trees containing these constructs. However, our collaborators have since decided that time and regulatory barriers are going to be too much to contend with and they are no longer interested in planting a field trial for this project. We will continue this work in the greenhouse to assess the vegetative phenotypes of these trees but we will no longer be able to assess their floral phenotypes under natural flowering. Multiple-year field trials of Eucalyptus are very difficult in Oregon due to the unsuitable growing environment (frost damage). What opportunities for training and professional development has the project provided?This project provided training on literature searches, cloning CRISPR constructs, molecular techinques and statistical analysis for three graduate students (two PhDs and one MS student). The project also provided one postdoc with the opportunity to strengthen skills on qPCR (and associated analyses) and bioinformatics. Two faculty research assistants were able to further their education in transformation techniques, greenhouse management, and program management skills. Furthermore, one undergraduate thesis student is gaining invaluable opportunity to particiapte in the research on this project, including to help with experimental design and the underlying decision making processes, fundamental and advanced molecular biology techniques, and preliminary statistical work. Three other undergraduate students who are non-research focused have had the opportunity to learn about data collection, spreadsheet use, greenhouse management, pest identification, and field work. How have the results been disseminated to communities of interest?Results to date have been published in one peer-reviewed journal and disseminated through posters and oral presentations. What do you plan to do during the next reporting period to accomplish the goals? Objective 1. Greenhouse study of FT-accelerated flowering background in poplar and Eucalyptus (Year 1) a. Assess vegetative and floral morphology in Eucalyptus LFY and novel gene constructs Continue to select and perform PCR verification of SP7/FT transformed with EDA33, REC8, TDF1, and CAS9 to get more events for GH studies Screen mutants and propagate KO events for GH studies Once events are selected, begin GH trial of novel gene targets in FT SP7 and take data on vegetative and floral morphology to assess the effects of the genes Continue to analyze gene expression of genes upstream of LFY and we will finalize analysis on genes downstream of LFY b. Assess vegetative and floral morphology in poplar LFY and AG constructs Continue to conduct the greenhouse study of 353 KOs targeting LFY and AG containing the FT transgene background Begin heat induction experiment with 717 KOs containing FT gene and monitor floral development/phenotype We will continue with PCR confirmation of the 353 KOs with HSP::AtFT, propagate positive FT subevents, and prepare for heat induction experiments in the greenhouse Analyze results of heat induction experiment for both 717 and 353 once concluded Objective 2. Greenhouse and field study of wild type background poplar and Eucalyptus (Years 1-3) a. Assessment of naturally flowering poplar Monitor and care for CRISPR-induced LFY and AG KO plants growing in the field Collect growth data (height, diameter at 6" above soil level) during winter 2018, single trees to promote growth of the leader stem in spring 2019, and take/analyze leaf vegetative data in summer 2019 once leaves emerge and reach maturity Monitor for flowering in the field, although we don't expect flowering to begin for several more years b. Assessment of naturally flowering Eucalyptus Complete mutants screening of SP7 with EDA33 and TDF1 Propagate SP7 KOs and conduct greenhouse assessment Take GH data on vegetative phenotypes and analyze to determine if novel gene targets affect vegetative traits compared to Cas9 and WT controls Objective 3. Evaluation of off-target effects and chimerism (Years 1-3) Use the computer script to identify the flanking regions around our off-target sites and design primers for bait-capture studies Order baits or oligos for targeted capture followed by sequencing for off-target sites Analyze sequencing data to determine the rate of off-target mutation from CRISPR constructs targeting LFY or AG in poplar (LFY and AG) and Eucalyptus (LFY) compared to non-CRISPR controls Objective 4. Design and test cre-lox CRISPR excision system (Years 2-3) We will create constructs with several different lengths of promoter with and without the WUSCHEL (WUS) gene (both transcriptional and translational fusions) with a fluorescence reporter gene like GREEN FLUORESCENT PROTEIN (GFP) to determine which promoters work best Transform those constructs and analyze reporter expression during regeneration Design and clone the constructs that we will use for testing an effective excision system that will remove itself after gene editing has occurred Transform and identify lines with the cre-lox construct and assess the rate and cell-specificity of excision

Impacts
What was accomplished under these goals? Objective 1. Greenhouse study of FT-accelerated flowering background in poplar and Eucalyptus (Year 1) - 55% completion a. Assess vegetative and floral morphology in Eucalyptus LFY and novel gene constructs Transformation is underway for the greenhouse (GH) study that will look at the effects of three novel gene constructs on vegetative and floral morphology. Using two early flowering (FLOWERING LOCUS T (FT)) overexpression events in Eucalyptus hybrid E. grandis x E. urophylla (hereafter referred to as SP7), we co-cultivated 1,346 leaf explants with novel CRISPR constructs targeting EMBRYO SAC DEVELOPMENT ARREST 33 (EDA33), MEIOTIC RECOMBINATION 8 (REC8), and TAPETAL DEVELOPMENT FUNCTION 1 (TDF1). We also co-cultivated explants containing an empty vector with the Cas9 enzyme as a control, and subculture and selection are still ongoing for all. Once sufficient transgenic knockout (KO) events are selected they will be propagated for GH studies. So far we have obtained 21 PCR positive events (6 EDA33, 6 REC8, 3 TDF1, and 6 CAS9). In January we began the GH trial for FT SP7 targeting LEAFY (LFY) with 38 LFY KO events, 6 Cas9 FT control events, 2 FT only (early flowering background but not retransformed) flowering control events, 2 escapes (i.e. also FT flowering controls but underwent transformation process), and wild type (WT) controls. We have measured height, diameter, leaf shape, leaf area, tree crown/shape, SPAD, scored for flowering, and assessed floral morphology based off comparisons to control flowers. All trees have been photographed individually as well as in comparison groups, and select flowers were sampled for microscopy. There was a high rate of flowering overall with only a few events that did not produce flower buds. We observed altered floral phenotypes in the LFY KO mutants (27 of 36 predicted KO events had the altered phenotype, 3 had WT flowers, and 6 produced no flower buds). The altered flowers developed a bract-in-bract look and appeared as if they had diminished floral determinacy. Upon bud dissection and microscopy, we determined that these flowers appeared to be sterile, either with no floral organs (just vegetative layers inside buds), or unorganized floral tissue. To further evaluate the effectiveness of LEAFY as a gene that induces sterility when knocked out, we performed gene expression analysis on select genes directly downstream and upstream of LFY. We sampled late buds for qPCR analysis from two events of each KO construct as well as empty vector Cas9 controls and WT controls. Of the genes we looked at, preliminary results indicate that APETALA3 (AP3-1) and SEEDSTICK (STK) expression may be significantly reduced in KO mutants compared to the flowering controls, but that AP3-2 expression did not follow suit. We are continuing this work into next year. b. Assess vegetative and floral morphology in poplar LFY and AG constructs We transformed early flowering (FLOWERING LOCUS T (FT) gene) constructs into ten 717 KO mutants and conducted a GH study with FT subevents of two KO constructs as well as controls in 2017 and 2018; however, none of the plants flowered after several months. We tried trimming the plants to stimulate floral organ initiation and development, but did not see any flowering after that either. We also transformed the same early flowering (FT) construct into selected events (3 LFY KOs, 1 LFY transgenic escape, 2 AGAMOUS (AG) KOs, and 2 Cas9 control events) in male clone 353. FT subevents from one LFY KO gene have been planted in soil for the GH study. So far, 46 trees have flowered (28 trees of the LFY KOs and 18 control trees) from a total of 227 trees studied. For all FT poplar we measured leaf shape, leaf area, diameter, and SPAD to assess vegetative morphology regardless of whether or not they flowered. Since we could not induce flowering in 717 in normal GH studies, we retransformed the heat inducible construct (HSP::AtFT) into selected 717 and 353 KO mutants as this construct gave us normal and frequent flowering after three weeks of heat induction in previous experiments. For the heat induction experiment, we have 2,295 explants in 717 from three LFY KOs, two AG KOs, and two CAS9 control events transformed with HSP::AtFT; 212 FT subevents were obtained and 182 events have been propagated for the study. In 353 we have 2,543 explants from two LFY KO events, two AG KO events, and two CAS9 control events that have been co-cultivated with HSP::AtFT; selection and PCR confirmation to obtain events for the study are underway. Objective 2. Greenhouse and field study of wild type background poplar and Eucalyptus (Years 1-3) - 25% completion a. Assessment of naturally flowering poplar We completed the GH assessments of vegetative morphology for CRISPR-induced LFY and AG KO poplar events prior to the start of this proposal. In November 2017 we planted a field trial consisting of 136 trees from clone 353 and 180 trees from clone 717. The trial was irrigated over the summer and data collection will begin in winter 2018 after they have concluded their first growing season in the field. Trees were scored for survival and we found that the 353s had a 93.4% survival rate and the 717s had a 93.3% survival rate. Dead trees were replaced with trees of the same construct and event (when possible), and those trees will be monitored closely to ensure they survive through the winter. b. Assessment of naturally flowering Eucalyptus We have transformed 1,515 leaf explants of wild type SP7 with EDA33, REC8, TDF1, and an empty control vector containing CAS9. We have obtained 58 PCR positive events (18 EDA33, 11 REC8, 18 TDF1, and 11 CAS9). Among 11 REC8 events, 9 of them are mutants confirmed by mutant screening and sequencing. 7 selected biallelic REC8 KO events, 10 Cas9 only events, and WT trees are rooted and planted in soil for GH study to analyze the effects on vegetative growth. Mutants screening for EDA33 and TDF1 are in progress. The GH trial of eucalypt SP7 targeting LFY is now concluding after six months of analysis. During those months we have measured height, diameter, leaf shape, leaf area, tree crown/shape. All trees were photographed. No significant differences were found between KOs and controls. No flowers were observed because they did not contain the early flowering (FT) background (thus they had WT background and we would not expect them to flower until their natural maturity). Objective 3. Evaluation of off-target effects and chimerism (Years 1-3) - 20% completion We identified and collected leaf tissue from the field or greenhouse for 96 individual ramets corresponding to LFY and AG KOs, heterozygous mutants, transgenic non-mutants, and WT controls for DNA isolation and off target analysis from the GH and field in poplar (field) and eucalyptus (GH). Off-target sites were then predicted using Cas-OFFinder with mismatches up to and less than 5 base pairs, and a computer script was written to automatically find the sequences around selected off-target sites for primer development. We plan to isolate and resequenced genomic targets corresponding to the off-target sites in the coming year. Objective 4. Design and test cre-lox CRISPR excision system (Years 2-3) - 0% completion Work under objective four is set to begin during the next reporting period.

Publications

• Type: Journal Articles Status: Published Year Published: 2018 Citation: Elorriaga, E., Klocko, A.L., Ma, C., and Strauss, S.H. (2018). Variation in Mutation Spectra Among CRISPR/Cas9 Mutagenized Poplars. Front. Plant Sci. 9. https://www.frontiersin.org/articles/10.3389/fpls.2018.00594/full