DEVELOPMENT OF NON-TRANSGENIC HLB RESISTANT CITRUS VARIETIES USING CRISPR-CAS9

<br>What was accomplished under these goals? Objective 1. We have completed yeast two hybridization (Y2H) for the 30 effectors including five defense-suppressing effectors and identified their targets in Valencia sweet orange. Overall, more than 100 putative targets have been identified using Y2H. We investigated the spatiotemporal CLas colonization in different tissues post ACP transmission. The spatiotemporal detection of CLas in different tissues after ACP transmission helps visualize the infection process of CLas in planta and subsequent HLB symptom development, and provides the knowledge supporting that young leaves should be the focus of HLB management. Here we show that one of Las-secreted proteins, SDE15, suppresses plant immunity and promotes Las multiplication. Transgenic expression of SDE15 in Duncan grapefruit suppresses the hypersensitive response (HR) induced by Xanthomonas citri subsp. citri (Xcc) and reduces the expression of immunity related genes. SDE15 also suppresses the HR triggered by the Xanthomonas vesicatoria effector protein AvrBsT in Nicotiana benthamiana, suggesting that it may be a broad-spectrum suppressor of plant immunity. SDE15 interacts with the citrus protein CsACD2, a homologue of Arabidopsis ACCELERATED CELL DEATH 2 (ACD2) based on Y2H, co-IP and pulldown assays. SDE15 suppression of plant immunity is dependent on CsACD2, and overexpression of CsACD2 in citrus suppresses plant immunity and promotes Las multiplication, phenocopying overexpression of SDE15. Identification of CsACD2 as a susceptibility target has implications in genome editing for novel plant resistance against devastating HLB. In addition, we have confirmed at least five more putative HLB susceptibility genes. We investigated flagella, one of the important virulence factors of Las. The expression of flagellar genes was higher in psyllids than in planta. The flagellar features of Las in planta suggest that Las movement in the phloem is not mediated by flagella. We also characterized the movement of Las after psyllid transmission into young flush. Our data support a model that Las remains inside young flush after psyllid transmission and before the flush matures. The delayed movement of Las out of young flush after psyllid transmission provides opportunities for targeted treatment of young flush for HLB control. 1.5. We found that SDE1 expression in the model plant Arabidopsis thaliana caused severe yellowing in mature leaves, reminiscent of both 'Ca. L. asiaticus' infection symptoms and accelerated leaf senescence. Induction of senescence signatures was also observed in the SDE1-expressing A. thaliana lines. These signatures were apparent in older leaves but not in seedlings, suggesting an age-associated effect. Furthermore, independent lines of transgenic Citrus paradisi (L.) Macfadyen (Duncan grapefruit) that express SDE1 exhibited hypersusceptibility to 'Ca. L. asiaticus'. Similar to A. thaliana, transgenic citrus expressing SDE1 showed altered expression of senescence-associated genes, but only after infection with 'Ca. L. asiaticus'. These findings suggest that SDE1 is a virulence factor that contributes to HLB progression, likely by inducing premature or accelerated senescence in citrus. 1.6. We found that HLB is a pathogen-triggered immune disease. CLas infection of Citrus sinensis stimulated systemic and chronic immune responses in phloem tissue including reactive oxygen species (ROS) production as indicated by H2O2, callose deposition, and induction of immune related genes. Systemic cell death of companion and sieve element cells, but not surrounding parenchyma cells, was observed following ROS production triggered by CLas. Death of companion cells was often observed before that of sieve elements. ROS production triggered by CLas localized in phloem-enriched bark tissue. The H2O2 concentration in exudates extracted from phloem enriched bark tissue of CLas infected plants kills citrus protoplast cells in vitro, which was suppressed by uric acid (a ROS scavenger) and gibberellin. RNA-seq analysis showed that gibberellin induces the expression of genes encoding H2O2 scavenging enzymes, but inhibits RBOHD that encodes a NADPH oxidase for ROS production. Foliar spray of HLB positive citrus with antioxidants and gibberellin significantly reduced both H2O2 concentrations and cell death in phloem tissues induced by CLas and reduced HLB symptoms. RNA-seq analyses of CLas infected and healthy C. sinensis demonstrates that CLas infections induce the expression of genes encoding NADPH oxidases and triggers downregulation of antioxidant enzyme genes, supporting that CLas causes oxidative stress. CLas also stimulate the expression of immune related genes. In summary, HLB is an immune-mediated disease while mitigating ROS via antioxidants and promoting plant growth can reduce cell death of phloem tissue caused by CLas, thus controlling HLB. 1.7. We have conducted RNAi analyses of 7 putative susceptibility genes. Objective 2. 2.1. We have developed multiplex genome editing toolkits for citrus that significantly improve citrus genome editing efficacy. CRISPR/Cas systems have been engineered for genome editing in many organisms, including plants. However, the gene editing efficiency in citrus via CRISPR technology remains too low to be implemented for genetic improvement in practice. Moreover, it is very difficult to obtain homozygous or biallelic knockout mutants in citrus. Here, we have developed multiplex genome editing toolkits for citrus including PEG-mediated protoplast transformation, a GFP reporter system that allows the rapid assessment of CRISPR constructs, citrus U6 promoters with improved efficacy, and tRNA-mediated or Csy4-mediated multiplex genome editing. Using the toolkits, we successfully conducted genome modification of embryogenic protoplast cells and epicotyl tissues. We have achieved a biallelic mutation rate of 89%, representing a significant improvement in citrus genome editing efficacy. 2.2 CRISPR-SpCas9 was successfully used to generate homozygous and biallelic citrus mutants in the T0 generation. Currently, the improved CRISPR technology is being used to generate homozygous mutants for putative HLB susceptibility genes including CsACD2. Cas12a ribonucleoprotein (RNP) was successfully used to edit the genome of protoplasts. We have generated embryogenic tissue cultures for Valencia sweet orange in Florida, Navel sweet orange in California, and Rio Red grapefruit in Texas. Genome editing of the putative HLB susceptibility genes including CsACD2 identified in objective 1 for the three different varieties are ongoing. Objective 3. For the genome modified citrus generated using Cas12a and SpCas9 in objective 2, we have conducted the off-target analyses. The plants are being kept in the greenhouse for analyses of horticultural traits and resistance against HLB. Some genome modified plants have been planted in an APHIS approved field site for evaluation. Objective 4. We have conducted surveys as planed regarding the consumer acceptance of genetically modified citrus. Objective 5. We have been giving presentations at the extension events on the collaborating states of FL, CA and TX including the Florida Citrus Expo, Grower Day, Florida Citrus Growers' Institute, California Citrus Growers Education Seminar Series, the UC Riverside Citrus Day and the Texas Citrus Showcase. We also published multiple extension materials regarding genome editing in industry magazines and the Cooperative Extension newsletters. <br><br><b>Publications</b><br>

<br>What was accomplished under these goals? Objective 1. 1.1 We have completed yeast two hybridization (Y2H) for the 30 effectors including five defense-suppressing effectors and identified their targets in Valencia sweet orange. Overall, more than 100 putative targets have been identified using Y2H. Confirmation of the targets is ongoing using co-immunoprecipitation and BiFC assays. We are conducting genome editing of the identified HLB S gene of Valencia sweet orange and Duncan grapefruit. 1.2 We investigated the spatiotemporal CLas colonization in different tissues post ACP transmission. At 75 day-post-ACP-removal (DPR), CLas was detected in roots of all trees, but in the mature leaf of only one tree, of the nine plants that were successfully infected via ACP transmission, consistent with the model that CLas moves passively from the source to sink. CLas was detected in 11.1%, and 43.1% mature leaves, which were unfed by ACPs during transmission, at 75, and 365 DPR, respectively, unveiling active movement to the source tissue. The difference in colonization timing of sink and source tissues indicates CLas is capable of both passive and active movement with passive movement being dominant. At 225 DPR, leaves fed by ACPs during the young stage showed the highest ratio of HLB symptomatic leaves and highest CLas titer, followed by that of leaves emerged post ACP removal, and mature leaves not fed by ACPs. Importantly, our data showed that ACPs were unable to transmit CLas via feeding on mature leaves. It is estimated that it takes at most three years for CLas to infect the whole tree. Overall, the spatiotemporal detection of CLas in different tissues after ACP transmission helps visualize the infection process of CLas in planta and subsequent HLB symptom development, and provides the knowledge supporting that young leaves should be the focus of HLB management. 1.3 Here we show that one of Las-secreted proteins, SDE15, suppresses plant immunity and promotes Las multiplication. Transgenic expression of SDE15 in Duncan grapefruit suppresses the hypersensitive response (HR) induced by Xanthomonas citri subsp. citri (Xcc) and reduces the expression of immunity related genes. SDE15 also suppresses the HR triggered by the Xanthomonas vesicatoria effector protein AvrBsT in Nicotiana benthamiana, suggesting that it may be a broad-spectrum suppressor of plant immunity. SDE15 interacts with the citrus protein CsACD2, a homologue of Arabidopsis ACCELERATED CELL DEATH 2 (ACD2) based on Y2H, co-IP and pulldown assays. SDE15 suppression of plant immunity is dependent on CsACD2, and overexpression of CsACD2 in citrus suppresses plant immunity and promotes Las multiplication, phenocopying overexpression of SDE15. Identification of CsACD2 as a susceptibility target has implications in genome editing for novel plant resistance against devastating HLB. In addition, we have confirmed at least five more putative HLB susceptibility genes. 1.4 We investigated flagella, one of the important virulence factors of Las. We analyzed the flagellar genes of Las. No flagellum was observed for Las in citrus and dodder. The expression of flagellar genes was higher in psyllids than in planta. In addition, western blotting using flagellin-specific antibody indicates that Las expresses flagellin protein in psyllids, but not in planta. The flagellar features of Las in planta suggest that Las movement in the phloem is not mediated by flagella. We also characterized the movement of Las after psyllid transmission into young flush. Our data support a model that Las remains inside young flush after psyllid transmission and before the flush matures. The delayed movement of Las out of young flush after psyllid transmission provides opportunities for targeted treatment of young flush for HLB control. 1.5. We found that SDE1 expression in the model plant Arabidopsis thaliana caused severe yellowing in mature leaves, reminiscent of both 'Ca. L. asiaticus' infection symptoms and accelerated leaf senescence. Induction of senescence signatures was also observed in the SDE1-expressing A. thaliana lines. These signatures were apparent in older leaves but not in seedlings, suggesting an age-associated effect. Furthermore, independent lines of transgenic Citrus paradisi (L.) Macfadyen (Duncan grapefruit) that express SDE1 exhibited hypersusceptibility to 'Ca. L. asiaticus'. Similar to A. thaliana, transgenic citrus expressing SDE1 showed altered expression of senescence-associated genes, but only after infection with 'Ca. L. asiaticus'. These findings suggest that SDE1 is a virulence factor that contributes to HLB progression, likely by inducing premature or accelerated senescence in citrus. Objective 2. 2.1. We have developed multiplex genome editing toolkits for citrus that significantly improve citrus genome editing efficacy. CRISPR/Cas systems have been engineered for genome editing in many organisms, including plants. However, the gene editing efficiency in citrus via CRISPR technology remains too low to be implemented for genetic improvement in practice. Moreover, it is very difficult to obtain homozygous or biallelic knockout mutants in citrus. Here, we have developed multiplex genome editing toolkits for citrus including PEG-mediated protoplast transformation, a GFP reporter system that allows the rapid assessment of CRISPR constructs, citrus U6 promoters with improved efficacy, and tRNA-mediated or Csy4-mediated multiplex genome editing. Using the toolkits, we successfully conducted genome modification of embryogenic protoplast cells and epicotyl tissues. We have achieved a biallelic mutation rate of 44.4% and a homozygous mutation rate of 11.1%, representing a significant improvement in citrus genome editing efficacy. 2.2 CRISPR-SpCas9 was successfully used to generate homozygous and biallelic citrus mutants in the T0 generation for two selected genes. Currently, the improved CRISPR technology is being used to generate homozygous mutants for putative HLB susceptibility genes including CsACD2. Here we further developed the non-transgenic genome editing of citrus technology via transient expression of Cas9- sgRNA DNA, RNA or ribonucleoprotein (RNP) into protoplasts by PEG-mediated transfection or into embryogenic cells of suspension culture by particle bombardment. Citrus protoplast or embryogenic cells appear to be recalcitrant to transformation using different approaches. Only transient expression of CRISPR/Cas9 DNA in protoplast cells led to efficient transformation and genome editing mutation rate of 78% was observed. We have finally generated non-transgenic genome modified mutations. The plants are under regeneration. We have generated embryogenic tissue cultures for Valencia sweet orange in Florida, Navel sweet orange in California, and Rio Red grapefruit in Texas. Genome editing of the putative HLB susceptibility genes including CsACD2 identified in objective 1 for the three different varieties are ongoing. Objective 3. For the genome modified citrus generated using Cas12a and SpCas9 in objective 2, we have conducted the off-target analyses. The plants are being kept in the greenhouse for analyses of horticultural traits. Some genome modified plants have been planted in an APHIS approved field site for evaluation. Objective 4. We are conducting surveys as planed regarding the consumer acceptance of genetically modified citrus. Objective 5. We have been giving presentations at the extension events on the collaborating states of FL, CA and TX including the Florida Citrus Expo, Grower Day, Florida Citrus Growers' Institute, California Citrus Growers Education Seminar Series, the UC Riverside Citrus Day and the Texas Citrus Showcase. We also published multiple extension materials regarding genome editing in industry magazines and the Cooperative Extension newsletters. <br><br><b>Publications</b><br>

<br>What was accomplished under these goals? Objective 1. We have completed screening of 20 putative Las effectors and 2 of them repressed plant defense. 1.1. We have completed yeast two hybridization (Y2H) for the 20 effectors including two defense-suppressing effectors and identified their targets in Valencia sweet orange. Overall, more than 50 putative targets have been identified using Y2H. Confirmation of the targets is ongoing using co-immunoprecipitation and BiFC assays. Meanwhile, we have conducted CTV-mediated gene silencing of 15 putative HLB susceptibility genes in collaboration with Dawson lab. Sweet orange plants carrying the CTV constructs were inoculated with Las via grafting. Interestingly, gene silencing of one of the putative HLB susceptible genes led to significant HLB tolerance. The plants showed mild HLB symptoms, similar growth as non-inoculated plants whereas the growth of control plants was significantly reduced and showed severe HLB symptoms. We are characterizing the putative mechanism of the HLB S gene. We are conducting genome editing of the identified HLB S gene of Valencia sweet orange and Duncan grapefruit. In addition, we also overexpressed the HLB S gene in Valencia sweet orange and will inoculate them with Las once they are one year old. 1.2. Here we show that one of Las-secreted proteins, SDE15, suppresses plant immunity and promotes Las multiplication. Transgenic expression of SDE15 in Duncan grapefruit suppresses the hypersensitive response (HR) induced by Xanthomonas citri subsp. citri (Xcc) and reduces the expression of immunity related genes. SDE15 also suppresses the HR triggered by the Xanthomonas vesicatoria effector protein AvrBsT in Nicotiana benthamiana, suggesting that it may be a broad-spectrum suppressor of plant immunity. SDE15 interacts with the citrus protein CsACD2, a homologue of Arabidopsis ACCELERATED CELL DEATH 2 (ACD2) based on Y2H, co-IP and pulldown assays. SDE15 suppression of plant immunity is dependent on CsACD2, and overexpression of CsACD2 in citrus suppresses plant immunity and promotes Las multiplication, phenocopying overexpression of SDE15. Identification of CsACD2 as a susceptibility target has implications in genome editing for novel plant resistance against devastating HLB. 1.3. We investigated flagella, one of the important virulence factors of Las. We analyzed the flagellar genes of Las. No flagellum was observed for Las incitrusand dodder. The expression of flagellar genes was higher in psyllids than in planta. In addition, western blotting using flagellin-specific antibody indicates that Las expresses flagellin protein in psyllids, but not in planta. The flagellar features of Las in planta suggest that Las movement in the phloem is not mediated by flagella. We also characterized the movement of Las after psyllid transmission into young flush. Our data support a model that Las remains inside young flush after psyllid transmission and before the flush matures. The delayed movement of Las out of young flush after psyllid transmission provides opportunities for targeted treatment of young flush for HLB control. This study suggested that we can utilize some flush specific promoters against Las infection. We are investigating whether the flagellum is involved in Las movement in planta and whether it induces plant immune responses by interacting with specific receptors. 1.4. Here, we investigated another important virulence factor of Las, the type IVc tight adherence (Tad) pilus. The Tad loci are conserved among members of Rhizobiaceae, including 'Ca.L. asiaticus' andAgrobacteriumspp. Ectopic expression of the 'Ca.L. asiaticus'cpaFgene, an ATPase essential for the biogenesis and secretion of the Tad pilus, restored the adherence phenotype incpaFmutant ofA. tumefaciens, indicating CpaF of 'Ca.L. asiaticus' was functional and critical for bacterial adherence mediated by Tad pilus. We are investigating whether the Tad pilus is involved in Las movement in planta and whether it induces plant immune responses by interacting with specific receptors. Objective 2. In this objective, we have been optimizing the citrus genome editing via the CRISPR technology and meanwhile, we are editing several putative HLB susceptibility genes including CsACD2 identified in Objective 1. 2.1. Recently, CRISPR-Cas12a (Cpf1) from Prevotella and Francisella was engineered to modify plant genomes. To further expand the tool box for genome editing of citrus, we employed CRISPR-LbCas12a (LbCpf1), which is derived from Lachnospiraceae bacterium ND2006, to edit acitrusgenome for the first time. LbCas12a was successfully used to modify two genes in Duncan grapefruit via Xcc-facilitated agroinfiltration. Finally, no potential off-targets were observed. Therefore, CRISPR-LbCas12a can readily be used as a powerful tool forcitrusgenome editing. 2.2 CRISPR-SpCas9 was successfully used to generate homozygous and biallelic citrus mutants in the T0 generation for two selected genes. Currently, the improved CRISPR technology is being used to generate homozygous mutants for putative HLB susceptibility genes including CsACD2. 2.3. Here we further developed the non-transgenic genome editing of citrus technology via transient expression of Cas9-sgRNA DNA, RNA or ribonucleoprotein (RNP) into protoplasts by PEG-mediated transfection or into embryogenic cells of suspension culture by particle bombardment. Citrus protoplast or embryogenic cells appear to be recalcitrant to transformation using different approaches. Only transient expression of CRISPR/Cas9 DNA in protoplast cells led to efficient transformation and genome editing mutation rate of 5.7% was observed. We further used hiTAIL-PCR analysis to show that the genome modified citrus cells are free of foreign DNA. We also detected low off-target mutations in genome modified citrus cells. Our findings represent a breakthrough in citrus breeding using the CRISPR technology. The non-transgenic genome editing method is being used to mutation the putative HLB susceptibility genes including CsACD2 identified in objective 1. 2.4. We have generated embryogenic tissue cultures for Valencia sweet orange in Florida, Navel sweet orange in California, and Rio Red grapefruit in Texas. Genome editing of the putative HLB susceptibility genes including CsACD2 identified in objective 1 for the three different varieties are ongoing. Objective 3. For the genome modified citrus generated using Cas12a and SpCas9 in objective 2, we have conducted the off-target analyses. The plants are being kept in the greenhouse for analyses of horticultural traits. Objective 4. We are conducting surveys as planed regarding the consumer acceptance of genetically modified citrus. Objective 5. We have been giving presentations at the extension events on the collaborating states of FL, CA and TX including the Florida Citrus Expo, Grower Day, Florida Citrus Growers' Institute, California Citrus Growers Education Seminar Series, the UC Riverside Citrus Day and the Texas Citrus Showcase. We also published multple extension materials regarding genome editing in industry magazines and the Cooperative Extension newsletters. <br><br><b>Publications</b><br>

<br>What was accomplished under these goals? Objective 1. Our hypothesis is that Las requires specific secreted effectors that directly or indirectly activate key host S genes or their products, which subsequently facilitate pathogen multiplication and trigger disease symptoms of HLB. To test this hypothesis, we will conduct functional characterizations of critical virulence effectors of Las and the associated key HLB S genes using mutagenesis, yeast two hybrid assays, ectopic expression, and RNA interference techniques. For the predicated SDEs, we have tested whether they suppress programmed cell death (PCD). For the SDEs suppressing PCD, we identified their targets (HLB susceptibility genes) using yeast two hybrid assay. Yeast two-hybrid, in vitro protein pull-down and in vivo bimolecular fluorescence complementation assays showed that SDE15 interacts with ACD2 (ACCELERATED CELL DEATH 2), a repressor of plant PCD and that it enhances the red chlorophyll catabolite reductase (RCCR) activity of ACD2 to remove porphyrin-related molecules, accumulation of which causes PCD. SDE15 promotes the chlorophyll break-down in planta and contributes to the development of yellowing symptom associated with HLB. We have conducted RNAi for 12 putative target genes. The silenced plants have been inoculated with Candidatus Liberibacter asiaticus (Las) to further confirm their involvement of HLB disease development. Objective 2. We hypothesize that HLB resistance can be obtained by modifying key citrus HLB S genes to be unresponsive to Las. We will alter key S genes using CRISPR-Cas9/sgRNA genome editing technology to develop HLB resistant citrus varieties. In this objective, we have been optimizing the non-transgenic genome editing of citrus using the CRISPR technology and generating biallelic mutants for the putative HLB S genes. We have generate tissue cultures formajor citrus scion varieties in Florida, California, and Texas and two rootstock varieties. In our study, we employed CRISPR-LbCpf1, derived from Lachnospiraceae bacterium ND2006, to edit the citrus genome. First, LbCpf1 was successfully used to modify the CsPDS gene in grapefruit Duncan via the Xcc-facilitated agroinfiltration. Next, LbCpf1 driven by either 35S or Yao promoter was used to edit the PthA4 effector binding elements in the promoter (EBEPthA4-CsLOBP) of CsLOB1. CsLOB1 is the canker susceptibility gene induced by the corresponding pathogenicity factor PthA4 of Xanthomonas citri, via binding to EBEPthA4-CsLOBP. Totally, seven 35S-LbCpf1-transformed Duncan plants were generated, designated as #D35s1 to #D35s7, and ten Yao-LbCpf1-transformed Duncan plants were created, designated as #DYao1 to #DYao10. LbCpf1-directed EBEPthA4-CsLOBP modifications were observed in three 35S-LbCpf1-transformed Duncan (#D35s1, #D35s4 and #D35s7). However, no LbCpf1-mediated indels were observed in the Yao-LbCpf1-transformed plants. Importantly, transgenic line #D35s4, containing the highest mutation rate, alleviates XccΔpthA4:dCsLOB1.4 infection. Therefore, CRISPR-LbCpf1 can be readily used as a powerful tool for citrus genome editing. Here, we conducted nontransgenic genome editing of citrus via the transient introduction of Cas9-sgRNA DNA, RNA or ribonucleoprotein (RNP) into protoplast by PEG-mediated transfection or into embryogenic cells in a suspension culture by particle bombardment. Citrus protoplast or embryogenic cells appear to be recalcitrant to transformation using different approaches. Only the transient expression of CRISPR/Cas9 DNA in protoplast cells led to efficient transformation, and a genome editing mutation rate of 5.7% was observed. We further used hiTAIL-PCR analysis to show that the genome-modified citrus cells are free of foreign DNA. We also detected low off-target mutations in genome-modified citrus cells. Our findings represent a breakthrough in citrus breeding using CRISPR technology. We are currently in the process of regenerating citrus from the genome modifed cells. Objective 3. We hypothesize that select modified citrus varieties will have HLB resistance and acceptable horticultural traits. We will evaluate the genome-modified citrus varieties for HLB resistance, tree growth and development, canopy development, and other related horticultural traits. This objective will be conducted once the genome modified plants are planted in the field. Objective 4. Acceptance of genetically modified citrus will be determined by the availability of educational material about the genome modification process. We propose to provide educational material and assess public sentiment to genome edited products through customer responses to educational material. We are conducting surveys as planed regarding the consumer acceptance of genetically modified citrus. Objective 5. Extension and outreach with stakeholders are hypothesized to facilitate the development and delivery of products to end-users. We propose to provide extension and educational outreach to stakeholders and consumers. We have been giving presentations at the extension events on the collaborating states of FL, CA and TX including the Florida Citrus Expo, Grower Day, Florida Citrus Growers' Institute, California Citrus Growers Education Seminar Series, the UC Riverside Citrus Day and the Texas Citrus Showcase. We also published multple extension materials regarding genome editing in industry magazines and the Cooperative Extension newsletters. <br><br><b>Publications</b><br>