Thank you for submitting your abstract for the symposium poster session! 
Poster presenters will be notified of their acceptance and if they have been selected for a lightning talk by Friday, May 8th.

• All posters should be less than or equal to 48x48". Push pins will be provided for hanging. 
• Plan to bring your poster to the Cargill Building on May 14th at 8:30 AM. You will hang your poster that morning in a fashion that allows 2 posters to be hung per side. Each poster board should display a total of 4 posters.  
• Poster presenters are expected to stand (or sit) with their poster and interact with attendees during the poster session (12:00 - 1:00 PM). 
• If you do not wish to eat during the poster session and you would like a box lunch put aside for you, please request this in advance by emailing Kit ([email protected]).

Our poster session is juried! Prizes will be awarded to the top three posters. Posters will be evaluated day-of by our visiting scholars; prizes will be announced at the reception following the symposium. A special thanks to the following organizations for their generous donations! 

• Minnesota Orchestra 
  1st prize: 2 tickets to a summer 2026 performance
• Embrace North Sauna
  2nd & 3rd prizes: 3-month sauna & cold-plunge membership

Tuning transcriptional output
Presented by Lucia Strader, Professor, Howard H. and Maryam R. Newman Chair, Salk Institute for Biological Studies

Arabidopsis gene expression is regulated by more than 1,900 transcription factors (TFs), which have been identified genome-wide by the presence of well-conserved DNA binding domains. Activator TFs contain activation domains (ADs) that recruit coactivator complexes; however, for nearly all Arabidopsis TFs, we lack knowledge about the presence, location, and transcriptional strength of their ADs ¹. To address this gap, we experimentally identified Arabidopsis ADs on a proteome-wide scale using a yeast library approach, finding that over half of the Arabidopsis TFs carry an AD. We annotated 1,553 ADs, the vast majority of which were previously unknown. We used the dataset generated to develop a neural network to accurately predict ADs and to identify sequence features necessary to recruit coactivator complexes. We uncovered six distinct sequence feature combinations that resulted in activation activity, providing a framework to interrogate activation domain sub-functionalization. Furthermore, we identified ADs within the ancient AUXIN RESPONSE FACTOR (ARF) family of TFs, uncovering conservation of AD positioning in distinct clades. Our findings provide a deep resource for understanding transcriptional activation, a framework for examination of function within intrinsically disordered regions, and a predictive model of ADs.

Viral delivery of gene editing reagents in crop plants
Presented by Steve Whitham, Professor, Dept. of Plant Pathology, Entomology, Microbiology, Iowa State University

Plant viruses are emerging as powerful and versatile platforms for delivering genome editing reagents, enabling rapid, systemic, and transformation-free modification of plant genomes. These approaches are particularly promising for crop species where stable transformation remains a major bottleneck. Viral vectors have been developed and optimized for gene silencing and protein expression, and more recently adapted for delivery of gene editing reagents in both model and crop species. Comparative analyses reveal that viral platforms differ in their capacity to induce somatic and heritable genome edits, emphasizing the importance of vector-host biology in shaping editing outcomes. Engineering strategies—such as promoter duplication, RNA processing systems, and multiplex guide design—further affect editing efficiency. We have also examined the role of RNA mobility signals and found that their presence is not sufficient to promote germline editing for a virus that lacks the intrinsic ability to induce heritable edits, which highlights the need to better understand virus–host interactions that enable heritable modifications. Continued innovation aimed at improving efficiency, expanding host range, and enabling reliable heritable editing is poised to unlock new opportunities for both fundamental discovery and crop improvement.

A Rapid Screening Platform for Mini Cas Activity

Radwa Kamel¹, Feng Zhang¹

¹Department of Plant and Microbial Biology, University of Minnesota–Twin Cities, St. Paul, Minnesota, USA

Genome editing in plants has been revolutionized by CRISPR–Cas systems, with CRISPR-Cas9 being the most common and widley used platform. However, its dependence on tissue culture and stable transformation remains a major bottleneck for rapid and scalable applications. Recently, virus-mediated genome editing has emerged as a promising alternative, enabling the direct delivery of compact CRISPR systems into plants and bypassing the need for tissue culture. In particular, the use of mini nucleases offers new opportunities for efficient viral delivery.Despite these advances, several challenges remain, including the identification of optimal nuclease variants, guide RNAs, and target sites, especially given the unique target adjacent motifs of these compact systems. Additionally, traditional validation of editing efficiency in endogenous targets is time-consuming, typically requiring at least 10 days followed by genotyping analyses.

To address these limitations, we developed a rapid and scalable RUBY-based screening system. In this approach, the nuclease, guide RNA, and target sequence are co-delivered within a single construct, enabling direct visualization of editing outcomes through betalain pigment production. This system allows estimation of editing efficiency within 5 days, significantly accelerating the screening of mini nucleases and guide RNAs. Our platform provides a powerful tool for optimizing virus-mediated genome editing and facilitating the rapid evaluation and application of compact CRISPR systems in plants.

Stacking C4-like Anatomical Traits with OsDREB1c Upregulation to Drive Photosynthetic Efficiency and Nitrogen Flux in Oryza sativa 

Erik Myers¹ ([email protected]), Abe Steinberger¹, Ji Hyun Kim¹, Daniela Vlad², Jane Langdale², Daniel Voytas¹

^1. Dept. of Genetics, Cell Biology and Development, University of Minnesota
^2. Dept. of Biology, University of Oxford

The transition from C3 to C4 photosynthesis in rice (Oryza sativa) represents a strategy for increasing crop yields by minimizing photorespiratory loss. While rice naturally utilizes the less efficient C3 pathway, previous research demonstrates that the targeted knockout of three microproteins—OsPEL1, OsPEL2, and OsPEL3—successfully recapitulates a Kranz-like anatomy by increasing chloroplast biogenesis within the bundle sheath cells. Because 50% of leaf blade nitrogen in C3 plants is already committed to photosynthesis, supporting the demands of this enhanced photosynthetic architecture requires highly optimized nitrogen (N) source-sink relations. To address this, we sought to simultaneously knock out the PEL family while upregulating the expression of OsDREB1c, an important transcription factor related to N uptake and allocation. By leveraging two distinct gene-editing reagents—LbCas12a and a prime editor—we successfully generated large deletions in the PEL family and inserted a small 36bp enhancer sequence to increase the endogenous expression of OsDREB1c. The resulting lines exhibit the dark green phenotype reported in the literature and demonstrate a ~15-fold increase in OsDREB1c expression. This work demonstrates the technical feasibility of delivering multiple gene-editing reagents, introducing large T-DNAs (~35kb), and executing complex multiplex edits, providing a foundational platform to investigate whether stacking these structural and metabolic traits can synergistically improve nutrient flux in rice leaves to support C4 photosynthesis.

Acknowledgement: This work is supported by the Gates foundation

References: 
Choi, Heebak, et al. "Editing of rice PSEUDO-ETIOLATION IN LIGHT microProtein genes promotes chloroplast development." The Plant Cell 37.11 (2025): koaf235.
Wei, Shaobo, et al. "A transcriptional regulator that boosts grain yields and shortens the growth duration of rice." Science 377.6604 (2022): eabi8455.

Single-Nucleus and Spatial Transcriptomics Reveal Circadian Regulatory Architecture in the Mature Arabidopsis Leaf

Zachary Myers¹, Alveena Zulfiqar¹, Danielle Schoenecker¹, Ananda Menon¹, and Kathleen Greenham¹

¹ Department of Plant and Microbial Biology, University of Minnesota, Saint Paul, MN, USA

Plant circadian clocks are decentralized, allowing distinct cell types to tune daily transcriptional programs to their physiological roles; however, how these clocks are organized across an intact mature leaf remains poorly understood. Here, we combine single-nucleus RNA sequencing and targeted spatial transcriptomics to define cell-type-resolved and spatially anchored circadian regulation in Arabidopsis. These approaches resolved major leaf cell types and identified widespread rhythmic transcription, including more than 7,400 genes cycling in at least one cluster. Temporal co-expression analysis revealed major circadian expression programs shared across cell types, while also uncovering genes with cell-type-specific phase shifts. Cluster-specific gene regulatory networks showed that core clock regulators, including CCA1 and PRR7, regulate both broadly shared and cell-type-restricted targets associated with photosynthesis, light signaling, auxin, and abscisic acid responses. Targeted spatial transcriptomics validated broad time-of-day effects among clock components, supported single-nucleus marker-based cell-type assignments, and revealed spatially patterned transcription factor expression missed by single-nucleus profiling alone. Together, these data show that circadian regulation in the mature leaf is widespread, spatially structured, and highly dependent on cell-type context.

Acknowledgements: This work was supported by the National Science Foundation (NSF DBI 2042159), the National Institute of Health (NIH NIGMS 5R35GM14837-03), the University of Minnesota Genomics Center, and by resources and staff at the University of Minnesota University Imaging Center (UIC, SCR_020997) and the University of Illinois Chicago Spatial and Genome Technologies Core.

A Sea of Data: Screening for Interactions Between Plant Immune Receptors & Elicitors from Mycotoxigenic Fungi

Brian D. Rutter¹, Tiana Roth¹, Avery Hickcox¹, Mengying Wang¹, and Josiah Mutuku¹

¹2Blades, University of Minnesota – Twin Cities, St. Paul, Minnesota, USA 

Mycotoxigenic fungi pose a serious threat to global food security and human health. In the United States, $3-4 billion is spent annually to mitigate the impacts of mycotoxins on agriculture. Contaminated grains are often destroyed outright or sold at severely marked down prices, leading to substantial economic losses. World-wide, and especially in developing countries, over 4.5 billion people consume food with unregulated mycotoxin levels. The presence of mycotoxins can lead to serious health effects, including cancers, birth defects, organ damage or even death.

2Blades is non-profit organization dedicated to developing new, disease-resistant crops, with groups based at The Sainsbury Institute in Norwich, England and at the University of Minnesota (UMN), St. Paul. The 2Blades UMN Group uses a maize protoplast-based assay to screen its extensive library of plant pattern recognition receptors (PRR), which are genes involved in early pathogen detection. For the past year, the 2Blades UMN Group has screened thousands of interactions between fungal elicitors and the PRR genes and uncovered new and exciting interactions. Some of the genes discovered through screening are currently being transformed into maize. The goal is to develop more resilient crops with durable resistance to mycotoxigenic fungi and put them in the hands of growers.

Acknowledgement: This work is supported by the USDA (58-5062-014), The Foundation for Food & Agriculture Research (FFAR; 23-000769) and The Minnesota Corn Growers Association (MCGA; 6138-24DD).

GopherEye: Multimodal Vision-Language Model for Automated Grape Leaf Disease Diagnosis

Maiia Gareeva¹, Chang Shen¹, Nicholas Padilla¹, Ce Yang¹

¹ Department of Bioproducts and Biosystems Engineering, University of Minnesota Twin Cities, Minneapolis, MN, USA.

Grape leaf diseases such as Downy Mildew and Powdery Mildew pose significant threats to global agricultural progress, moreover accurate diagnosis requires trained plant pathologists, expert evaluation and does not scale efficiently. To address this challenge, we present GopherEye application with a vision-language model (VLM) designed for automated grape leaf disease diagnosis and structured report generation. The system takes a leaf image as input and produces both disease classification and a detailed natural language diagnostic report.

Our approach combines visual feature extraction with language modeling through a two-stage training pipeline, enabling efficient adaptation with limited computational overhead. This design allows the lightweighted model to capture fine-grained visual symptoms and translate them into relevant descriptions.

In stage 1, we adopt vision-language representation learning using a LAVIS BLIP-2 framework, optimizing image-text contrastive, image -text matching, and language modeling objectives.In stage 2, we connect Q-Former to Qwen with a MLP. We perform vision-to-language generative fine-tuning using LoRA adapter on Qwen.

We evaluate GopherEye on the Niphad Grape Leaf Disease Dataset (NGLD) and our own vineyard samples, consisting of annotated images of healthy leaves and disease classes. The model achieves strong classification performance with a macro F1 score of 0.861, while maintaining high precision for major disease categories. Additionally, text generation quality is assessed using ROUGE metrics, demonstrating the ability to generate relevant evidence-based explanations.

Qualitative results show that the model at most correctly identifies key disease indicators, producing structured outputs that include disease prediction, symptom indicators, and supporting evidence. This capability highlights the potential of multimodal AI systems in agricultural decision support. Overall, GopherEye provides a scalable, interpretable, and efficient solution for plant disease diagnosis.

Acknowledgements: This work was supported by the National Science Foundation (NSF) I-Corps Program. Computational resources were provided by the Minnesota Supercomputing Institute (MSI). We gratefully thank Dr. Soon Li Teh (Department of Horticultural Science, University of Minnesota) for providing access to vineyard plants and enabling data collection.
 

Developmental regulators enable rapid and efficient soybean transformation and CRISPR-mediated genome editing

Anshu Alok^{1, 3*}, Vidhyavathi Raman^{1, 3*}, Leonidas D’Agostino², Arjun Ojha Kshetry², Krishan Mohan Rai^{1, 3}, Chunfang Wang^{1, 3}, Samatha Gunapati⁴, Robert M. Stupar^{3, 4}, Gunvant B. Patil², Feng Zhang^{1, 3}

¹Department of Plant and Microbial Biology, University of Minnesota, Saint Paul, MN 55108
²Institute of Genomics for Crop Abiotic Stress Tolerance (IGCAST), Department of Plant and Soil Science, Texas Tech University, Lubbock, Texas, USA
³Center for Precision Plant Genomics, University of Minnesota, Saint Paul, MN 55108
⁴Department of Agronomy and Plant Genetics, University of Minnesota, Saint Paul, MN 55108

Soybean (Glycine max) transformation remains challenging and has not kept pace with rapid advances in genetic engineering technologies due to low efficiency, lengthy timelines, and genotype dependency. Here, we developed a streamlined transformation method by leveraging developmental regulators (DRs) to promote de novo shoot regeneration directly from growing soybean plants. By evaluating multiple DR combinations, our results showed that co-expression of WUSCHEL2 (WUS2) and the gene encoding isopentenyltransferase (IPT) achieved higher transformation efficiencies (14.6% to 22.3%) in Williams 82 and Bert varieties than individual DRs without requiring exogenous hormones or selection agents. Moreover, this method produced heritable transgenic events within 9 to 11 weeks and successfully delivered CRISPR-Cas9 components, generating heritable mutations with 20% efficiency. The temporal transcriptomic and gene regulatory network analyses revealed that WUS2/IPT synergistically modulates stress responses and activates developmental pathways, orchestrating a transition from initial stress adaptation to regenerative programming. Our findings demonstrate that this DR-enabled approach significantly enhances soybean transformation frequency, reduces tissue culture requirements, and offers a promising genome-editing platform for soybean improvement.

Acknowledgement: A.A., V.R., C.W., and F.Z. are supported by the National Science Foundation (IOS-2040218 and IOS-2206920) awards. F.Z. were supported by USDA NIFA award #2021-67013-34565. L.D., A.O.K., and G.P. are supported by the Texas Governor’s University Research (GURI) grant. 

Viral Discovery for Delivery of Gene-editing Tools to Monocots 

Jacob R. Botkin^1,2, Charles Anderson^1,2, and Daniel F. Voytas^1,2 

¹Department of Genetics, Cell Biology and Development, University of Minnesota, St. Paul, MN; ²Center for Plant Precision Genomics, University of Minnesota, St. Paul, MN 

Maize (Zea mays) and sorghum (Sorghum bicolor) rank among the top five cereal crops in the world providing fodder, feed, food, and fuel. Viral delivery of gene-editing components offers a compelling strategy to bypass tissue culture, a significant bottleneck in gene-editing¹. Virus-induced genome editing (VIGE) has been demonstrated in Cas9-expressing dicots as well as monocots, such as barley and wheat². However, viral vectors for maize and sorghum, such as Foxtail mosaic virus (FoMV), do not access germline tissues and mutations are not heritable. In this study, we surveyed over 19,000 public transcriptome records of maize and sorghum to identify viral vector candidates based on tissue tropism. Our objective is to identify viruses that facilitate heritable mutagenesis, allowing for rapid functional genetics studies of maize and sorghum. Following de novo assembly and homology-based identification of viral contigs, we identified 153 and 44 known plant-infecting virus species from maize and sorghum, respectively, with 23 plant-infecting species in both crops. SCMV, MaYMV, BgSMV, MDMV, SMV, and BYSMV were top candidates for VIGE based on tissue tropism. Additionally, we identify 221 and 112 novel virus sequences for maize and sorghum, respectively, that were assigned to orders that contain plant-infecting viruses based on RdRP homology. Of those 26 and 13 novel virus species represent complete genomes that could be further classified based on capsid and replicase proteins for maize and sorghum, respectively. Interestingly, an uncharacterized Virgaviridae-like virus was found in maize ear, embryo, seed, ovule, nucellus, and pericarp tissues. Efforts are ongoing to vectorize and test the capacity of these species for the delivery of gene-editing components. Overall, this study increases the known and unknown viral diversity, and identifies candidate viruses for VIGE of maize and sorghum. 

References

1. Altpeter, F., Springer, N. M., Bartley, L. E., Blechl, A. E., Brutnell, T. P., Citovsky, V., Conrad, L. J., Gelvin, S. B., Jackson, D. P., Kausch, A. P., LeMieux, P. G., Medford, J. I., Orozco-Cárdenas, M. L., Tricoli, D. M., Van Eck, J., Voytas, D. F., Warbot, V., Wang, K., Zhang, Z. J., & Stewart, C. N. (2016). Advancing Crop Transformation in the Era of Genome Editing. The Plant Cell, 28(7), 1510–1520. DOI: 10.1105/tpc.16.00196

2. Liu, D., Ellison, E. E., Myers, E. A., Donahue, L. I., Xuan, S., Swanson, R., Qi, S., Prichard, L. E., Starker, C. G., & Voytas, D. F. (2024). Heritable gene editing in tomato through viral delivery of isopentenyl transferase and single-guide RNAs to latent axillary meristematic cells. Proceedings of the National Academy of Sciences, 121(39), e2406486121. DOI: 10.1073/pnas.2406486121

Visualizing Viral Co-Infections to Expand Cargo Delivery for Plant Gene Editing

Sara Endejan¹, Daniel Voytas¹

¹Department of Plant and Microbial Biology, University of Minnesota, St. Paul, MN, USA 

Viral-Induced Gene Editing (VIGE) represents a promising approach for plant genome engineering, enabling heritable edits without the need for tissue culture. Despite its potential, the broader application of VIGE is constrained by host plant defenses, limited viral cargo capacity, and complex virus–virus interactions. This study aims to identify viral combinations capable of establishing stable co-infections in Nicotiana benthamiana and ultimately facilitating the delivery of genetic cargo that has previously been restricted by size limitations. To investigate viral interactions during co-infection, we developed a visualization platform incorporating the fluorophores AmCyan and mVenus, alongside a novel split-recombinase system with a Ruby-reporter Nicotiana benthamiana line. This system enables visualization of viral distribution, revealing zones of co-infection, interaction, and exclusion within plant tissues. Overlapping infection regions are of particular interest, as they permit the reconstitution of split cargo into functional full-length proteins. Using this approach, we assess the compatibility of viral pairs for coordinated cargo delivery and identify candidates that support efficient co-infection dynamics. Additionally, the clear spatial delineation of infection zones provides a framework to explore strategies for expanding regions of viral activity and enhancing cooperative interactions. Collectively, this work underscores the importance of integrating advanced visualization tools to expand the capabilities of viral-based gene editing systems in plants.

Acknowledgement: This work was supported by the National Science Foundation (project 2126592), PlantSynBio: Chassis design for sustainable production of high value terpenoids in the crop species tomato. 

Precise genetic duplications in gene-edited plants

Savio S. Ferreira^1,2,3, Michael J. Smanski^1,2,3

¹Department of Biochemistry, Molecular Biology and Biophysics, University of Minnesota, St Paul, Minnesota, USA 
²BioTechnology Institute, University of Minnesota, St Paul, Minnesota, USA 
³Center for Precision Plant Genomics, University of Minnesota, St Paul, Minnesota, USA 

Transcription Factors (TFs) are master switches that control the expression of specific genes, ultimately regulating metabolic pathways, growth, development, and stress responses. TFs are composed of a DNA-binding domain (DBD), responsible for binding to the promoter of its targets, and a transactivation domain (TAD), which activates gene expression. Previous studies have demonstrated that transgenic plants harboring synthetic TFs with duplicated TADs can boost TF activity (He et al., 2025), leading to the up-regulation of downstream genes and the enhancement of traits controlled by those TFs. However, because this approach relies on traditional transgenesis, it faces significant regulatory burdens and poor public perception, limiting its utility in commercial breeding programs. To generate transgene-free plants with duplicated TADs, we employed Amplification Editing (AE) (Zhang et al., 2024)—a modified Prime Editing approach—to mimic natural evolutionary DNA duplications without leaving transgenic scars. Transient assays in Arabidopsis protoplasts demonstrated the feasibility of this method, achieving precise duplications for 8 out of 9 tested targets, with duplication sizes ranging from 200 bp to 5 kb. We subsequently generated transgenic Arabidopsis plants expressing the AE components to produce stable duplications of these eight targets. Upon screening approximately 20 T1 plants per target, precise duplications were confirmed in 4 out of 8 targets. These plants were self-pollinated, and their offspring were screened to identify transgene-free individuals with heritable duplications. We have confirmed at least one target with a precise duplication in the transgene-free progeny via both PCR and sequencing; genotyping for the remaining targets is ongoing. Next steps will focus on testing the expression of downstream genes affected by TFs with confirmed TAD duplications. This approach is virtually applicable to any transcription factor (TF) in a crop genome, providing a versatile framework to modernize trait development across a wide range of agricultural species.

Acknowledgements: This research was supported by the DOE Office of Science, Biological and Environmental Research (BER) Program, grant no. DE-SC0023142

References: 
He, C., Liang, Y., Chen, R., Shen, Y., Li, R., Sun, T., et al. (2025) Boosting transcriptional activities by employing repeated activation domains in transcription factors. The Plant Cell, 37, koae315.

Zhang, R., He, Z., Shi, Y., Sun, X., Chen, X., Wang, G., et al. (2024) Amplification editing enables efficient and precise duplication of DNA from short sequence to megabase and chromosomal scale. Cell, 187, 3936-3952.e19.

GopherEye: Multimodal Vision-Language Model for Automated Grape Leaf Disease Diagnosis

Maiia Gareeva¹, Chang Shen¹, Nicholas Padilla¹, Ce Yang¹

¹ Department of Bioproducts and Biosystems Engineering, University of Minnesota Twin Cities, Minneapolis, MN, USA.

Grape leaf diseases such as Downy Mildew and Powdery Mildew pose significant threats to global agricultural progress, moreover accurate diagnosis requires trained plant pathologists, expert evaluation and does not scale efficiently. To address this challenge, we present GopherEye application with a vision-language model (VLM) designed for automated grape leaf disease diagnosis and structured report generation. The system takes a leaf image as input and produces both disease classification and a detailed natural language diagnostic report.

Our approach combines visual feature extraction with language modeling through a two-stage training pipeline, enabling efficient adaptation with limited computational overhead. This design allows the lightweighted model to capture fine-grained visual symptoms and translate them into relevant descriptions.

In stage 1, we adopt vision-language representation learning using a LAVIS BLIP-2 framework, optimizing image-text contrastive, image -text matching, and language modeling objectives.In stage 2, we connect Q-Former to Qwen with a MLP. We perform vision-to-language generative fine-tuning using LoRA adapter on Qwen.

We evaluate GopherEye on the Niphad Grape Leaf Disease Dataset (NGLD) and our own vineyard samples, consisting of annotated images of healthy leaves and disease classes. The model achieves strong classification performance with a macro F1 score of 0.861, while maintaining high precision for major disease categories. Additionally, text generation quality is assessed using ROUGE metrics, demonstrating the ability to generate relevant evidence-based explanations.

Qualitative results show that the model at most correctly identifies key disease indicators, producing structured outputs that include disease prediction, symptom indicators, and supporting evidence. This capability highlights the potential of multimodal AI systems in agricultural decision support. Overall, GopherEye provides a scalable, interpretable, and efficient solution for plant disease diagnosis.

Acknowledgements: This work was supported by the National Science Foundation (NSF) I-Corps Program. Computational resources were provided by the Minnesota Supercomputing Institute (MSI). We gratefully thank Dr. Soon Li Teh (Department of Horticultural Science, University of Minnesota) for providing access to vineyard plants and enabling data collection.
 

AgroGem: A Rapid and Scalable Transient Transformation System for Functional Genetics in Plants

Shengsong Guo^1,2, Oliver Schlegel³, Jitesh Kumar^1,2, Zachary Myers¹, Shahryar Kianian⁴, Kathleen Greenham^1,2, Feng Zhang^1,2

¹Department of Plant and Microbial Biology, University of Minnesota, Saint Paul, MN 55108
²Center for Precision Plant Genomics, University of Minnesota, Saint Paul, MN 55108
³Department of Genetics, Cell Biology and Development, University of Minnesota, Minneapolis, MN 55455
⁴USDA-ARS Cereal Disease Laboratory, St. Paul, MN, 55108

Plant genetic transformation technologies, including both transient and stable approaches, are fundamental for studying gene function and advancing genome engineering in plants. Compared to stable transformation, transient systems offer quick evaluation of gene function and, more recently, for testing genome editing reagents. However, existing transient expression platforms are often limited by variable efficiency, technical complexity, and poor scalability, underscoring the need for a robust and versatile system applicable across plant species. Here, we developed AgroGem, an efficient and scalable Agrobacterium-mediated transient transformation system for Arabidopsis and related Brassicaceae species. By evaluating Agrobacterium strains, geminiviral replicon-based T-DNA vectors, and co-cultivation conditions, we achieved markedly enhanced transformation efficiency and transgene expression. AgroGem enabled high CRISPR-mediated editing frequencies that substantially outperformed existing transient methods. Importantly, CRISPR-mediated mutation spectra at targeted loci closely resembled those generated by stable floral dip transformation and preserved chromatin accessibility-dependent editing patterns across distinct CRISPR-Cas systems. This high-resolution mutation profiling allowed rapid genetic dissection of core non-homologous end joining (NHEJ) genes and uncovered unexpected roles for key factors, such as KU80 and XRCC4, in modulating DNA repair outcomes. In addition, AgroGem supported scalable protein–protein interaction assays and efficient transient transformation across multiple Brassicaceae species. Together, AgroGem establishes a versatile, high-throughput transient platform for genome editing, DNA repair analysis, and functional genetics in plants.

Acknowledgements: J.K. and F.Z. is supported by the National Science Foundation (IOS-2040218 and IOS-2206920) awards. K.G. is supported by the National Science Foundation (IOS-2029549 and DBI-2042159).

Uncovering Temporal Regulation of Cold Acclimation and Its Role in Freeze Tolerance in Arabidopsis Using a 24-Hour RNA Sequencing Time Course

Adelaide Hazen¹, Will Gustafson¹, Ananda Menon¹, Danielle Schoenecker¹, Sam Seaver², and Kathleen Greenham¹

¹Department of Plant and Microbial Biology, University of Minnesota, Twin Cities, Minnesota, USA 
²Argonne National Laboratory, Lemont, Illinois, USA 

Increasingly unpredictable weather patterns, including late spring frosts, have led to large-scale crop damage and loss. In response to abiotic stress, plants employ diel-regulated physiological changes to maximize energy and resource usage. Before cold stress, cooler temperatures signal a cold‑acclimation period that triggers transcriptome and metabolic remodeling in preparation for freezing. These regulatory changes greatly enhance freeze tolerance; however, the temporal dynamics of these shifts remain largely unresolved. In this study, we aimed to capture specifically timed responses during cold acclimation leading to freeze tolerance in Arabidopsis thaliana. First, we screened the freeze tolerance of ten geographically diverse ecotypes to select one freeze tolerant (Tsu-0) and one freeze sensitive (C24) ecotype for an RNA-seq time-course. We collected leaf tissue from three-week-old plants of both ecotypes every 4 h for 24 h under control and cold-acclimation conditions. We found transcriptome-wide changes in diel expression patterns, suggesting a remodeling of the transcriptional network. A close examination of enzymatic reactions in central and specialized metabolism uncovered a significant time of day cold response in proline biosynthesis in Tsu-0 but not C24. A follow up experiment confirmed higher proline levels at specific times of day during cold acclimation in Tsu-0, suggesting a mechanism for its enhanced freeze tolerance. To validate these findings, we introduced the rate-limiting enzyme for proline biosynthesis, P5CS2, under the control of a morning-expressed promoter in the cold-sensitive background (C24). We tested the freeze tolerance of these transgenic lines to validate the impact of specifically-timed expression of P5CS2 and proline accumulation during cold acclimation on overall freeze tolerance. Our findings highlight the critical role of diel timing in shaping plant preparation for freezing stress, underscoring the importance of diel regulation as a key factor in developing crops adapted to climate variability.

Acknowledgements: This work is supported by the National Science Foundation (grant no. NSF IOS-2029549). 

A Rapid Screening Platform for Mini Cas Activity

Radwa Kamel¹, Feng Zhang¹

¹Department of Plant and Microbial Biology, University of Minnesota–Twin Cities, St. Paul, Minnesota, USA

Genome editing in plants has been revolutionized by CRISPR–Cas systems, with CRISPR-Cas9 being the most common and widley used platform. However, its dependence on tissue culture and stable transformation remains a major bottleneck for rapid and scalable applications. Recently, virus-mediated genome editing has emerged as a promising alternative, enabling the direct delivery of compact CRISPR systems into plants and bypassing the need for tissue culture. In particular, the use of mini nucleases offers new opportunities for efficient viral delivery.Despite these advances, several challenges remain, including the identification of optimal nuclease variants, guide RNAs, and target sites, especially given the unique target adjacent motifs of these compact systems. Additionally, traditional validation of editing efficiency in endogenous targets is time-consuming, typically requiring at least 10 days followed by genotyping analyses.

To address these limitations, we developed a rapid and scalable RUBY-based screening system. In this approach, the nuclease, guide RNA, and target sequence are co-delivered within a single construct, enabling direct visualization of editing outcomes through betalain pigment production. This system allows estimation of editing efficiency within 5 days, significantly accelerating the screening of mini nucleases and guide RNAs. Our platform provides a powerful tool for optimizing virus-mediated genome editing and facilitating the rapid evaluation and application of compact CRISPR systems in plants.

A modular imaging toolkit for the moss Physcomitrium patens

Adelaide Mahler¹, Ivan Radin¹

¹Department of Plant and Microbial Biology, University of Minnesota, St. Paul, 55108 MN, USA

Cells are highly dynamic systems undergoing constant organellar rearrangement to accommodate environmental and developmental changes. Visualizing these processes is essential for understanding how subcellular organization underlies the function and fitness of organelles and the whole cell. And yet robust tools for imaging organelles remain limited. 

Here, we present a modular, Golden-Gate based fluorescent imaging toolkit in the model moss, Physcomitrium patens. Built as a six-fragment, high-efficiency assembly system of interchangeable promoters, targeting signals, fluorophores, and plant selectable markers, this library enables rapid generation of customized constructs for subcellular labeling across diverse experimental setups. Given its genetic tractability, simple tissue architecture, and unique evolutionary position, moss (P. patens) is a powerful model system, and we believe this toolkit will help advance cell biology research.  

To minimize biological disruption and artifacts, we utilize minimal targeting signals paired with a suite of constitutive promoters of graded strength, allowing for precise tuning of expression without overexpressing active, full-length proteins. This toolkit supports scarless assembly for markers of all major subcellular compartments, including plasma membrane, nucleus, cytoplasm, endoplasmic reticulum (ER), Golgi and trans-Golgi network (TGN), chloroplasts, mitochondria, peroxisomes, and vacuole lumen, paired with moss codon-optimized fluorophores spanning the visible spectrum. Its modular design allows users to custom-tailor constructs by combining promoters, targeting signals, and colors to fit unique experimental needs. 

Leveraging moss’ efficient homology-directed repair (HDR), constructs can be stably integrated at multiple characterized, nonessential loci, facilitating multicolor experiments. We designed additional vectors for native knock-in labeling of proteins. These can be used to label compartments for which efficient targeting signals are not known (e.g. vacuolar membrane, microtubules). 

Overall, our system is designed as an adaptable, “build-your-own” toolbox with easy addition of core elements (promoters, targeting signals, tags, etc.). It also allows for native tagging and overexpression of your favorite protein. We are excited to share these resources with the plant community to further understanding of nuanced subcellular fluctuations!

Acknowledgements: This work was supported by Internal Funding from the College of Biological Sciences, University of Minnesota. 

Stacking C4-like Anatomical Traits with OsDREB1c Upregulation to Drive Photosynthetic Efficiency and Nitrogen Flux in Oryza sativa 

Erik Myers¹ ([email protected]), Abe Steinberger¹, Ji Hyun Kim¹, Daniela Vlad², Jane Langdale², Daniel Voytas¹

^1. Dept. of Genetics, Cell Biology and Development, University of Minnesota
^2. Dept. of Biology, University of Oxford

The transition from C3 to C4 photosynthesis in rice (Oryza sativa) represents a strategy for increasing crop yields by minimizing photorespiratory loss. While rice naturally utilizes the less efficient C3 pathway, previous research demonstrates that the targeted knockout of three microproteins—OsPEL1, OsPEL2, and OsPEL3—successfully recapitulates a Kranz-like anatomy by increasing chloroplast biogenesis within the bundle sheath cells. Because 50% of leaf blade nitrogen in C3 plants is already committed to photosynthesis, supporting the demands of this enhanced photosynthetic architecture requires highly optimized nitrogen (N) source-sink relations. To address this, we sought to simultaneously knock out the PEL family while upregulating the expression of OsDREB1c, an important transcription factor related to N uptake and allocation. By leveraging two distinct gene-editing reagents—LbCas12a and a prime editor—we successfully generated large deletions in the PEL family and inserted a small 36bp enhancer sequence to increase the endogenous expression of OsDREB1c. The resulting lines exhibit the dark green phenotype reported in the literature and demonstrate a ~15-fold increase in OsDREB1c expression. This work demonstrates the technical feasibility of delivering multiple gene-editing reagents, introducing large T-DNAs (~35kb), and executing complex multiplex edits, providing a foundational platform to investigate whether stacking these structural and metabolic traits can synergistically improve nutrient flux in rice leaves to support C4 photosynthesis.

Acknowledgement: This work is supported by the Gates foundation

References: 
Choi, Heebak, et al. "Editing of rice PSEUDO-ETIOLATION IN LIGHT microProtein genes promotes chloroplast development." The Plant Cell 37.11 (2025): koaf235.
Wei, Shaobo, et al. "A transcriptional regulator that boosts grain yields and shortens the growth duration of rice." Science 377.6604 (2022): eabi8455.

Single-Nucleus and Spatial Transcriptomics Reveal Circadian Regulatory Architecture in the Mature Arabidopsis Leaf

Zachary Myers¹, Alveena Zulfiqar¹, Danielle Schoenecker¹, Ananda Menon¹, and Kathleen Greenham¹

¹ Department of Plant and Microbial Biology, University of Minnesota, Saint Paul, MN, USA

Plant circadian clocks are decentralized, allowing distinct cell types to tune daily transcriptional programs to their physiological roles; however, how these clocks are organized across an intact mature leaf remains poorly understood. Here, we combine single-nucleus RNA sequencing and targeted spatial transcriptomics to define cell-type-resolved and spatially anchored circadian regulation in Arabidopsis. These approaches resolved major leaf cell types and identified widespread rhythmic transcription, including more than 7,400 genes cycling in at least one cluster. Temporal co-expression analysis revealed major circadian expression programs shared across cell types, while also uncovering genes with cell-type-specific phase shifts. Cluster-specific gene regulatory networks showed that core clock regulators, including CCA1 and PRR7, regulate both broadly shared and cell-type-restricted targets associated with photosynthesis, light signaling, auxin, and abscisic acid responses. Targeted spatial transcriptomics validated broad time-of-day effects among clock components, supported single-nucleus marker-based cell-type assignments, and revealed spatially patterned transcription factor expression missed by single-nucleus profiling alone. Together, these data show that circadian regulation in the mature leaf is widespread, spatially structured, and highly dependent on cell-type context.

Acknowledgements: This work was supported by the National Science Foundation (NSF DBI 2042159), the National Institute of Health (NIH NIGMS 5R35GM14837-03), the University of Minnesota Genomics Center, and by resources and staff at the University of Minnesota University Imaging Center (UIC, SCR_020997) and the University of Illinois Chicago Spatial and Genome Technologies Core.

Diel Regulation and Tissue-Specific Partitioning of Glucosinolate Metabolism in Thlaspi arvense

Samantha Pelletier¹, Will Gustafson¹, Danielle Schonecker¹, Ananda Menon¹, Adrian Hegeman¹, and Kathleen Greenham¹

¹Department of Plant and Microbial Biology, University of Minnesota–Twin Cities, St. Paul, Minnesota, USA

The balance between resources going to primary or specialized metabolism is crucial for plant health. An example of this balance is in sulfur metabolism, with sulfur rich glucosinolates (GSLs) accounting for up to 30% of sulfur in Brassicaceae plants. One such plant, Thlaspi arvense (pennycress), is currently being developed as a winter cover crop to be used as a source of biofuel production but is limited in downstream processing for animal fodder due to its high level of GSLs in the seed. The delicate balance of GSL accumulation across tissue types is not understood, although there is vast diversity in GSL type between seed and vegetative tissue in many species. This accumulation is under diel control, resulting in time-of-day changes in abundance. To better understand the role that GSLs play in plant health, we performed a 24-hour time-course on leaf and root tissue to determine time of day accumulation of GSLs across tissue types. The metabolic profiles reflect GSL content that is tissue specific, and dynamic changes to sulfur compounds such as methionine. We also have transcriptomic data that reflect dynamic changes in GSL biosynthesis throughout the day. Additionally, we have data on the metabolic profiles of flower, floral bud, and seed tissues, which establishes clear tissue specific presence of specific GSLs, with only one GSL being present in the seed. Together, this project will uncover the importance of temporal and spatial regulation of specialized metabolism in the context of sulfur metabolism. Ultimately, this work will be essential for the metabolic engineering of T. arvense and other Brassicales species, where plant health can be optimized alongside seed quality.

Acknowledgements: This work is supported by the U.S. Department of Energy, Office of Science, Biological and Environmental Research (grant no. DE-SC0024336) to K.G.

Near real-time crop mapping through multimodal integration of remote sensing observations and crop planting patterns 

Lang Qiao¹, Junxiong Zhou¹, Xuecheng Li¹, and Ce Yang¹ 

¹Bioproducts and Biosystems Engineering Department, University of Minnesota, 1390 Eckles Ave, St. Paul, MN 55108, USA

Accurate early-season crop mapping is critical for agricultural monitoring and yield prediction. However, the effectiveness of crop mapping is hindered by the challenging task of acquiring reliable crop type labels. Although strategies based on historical classifiers and sample transfer have made some progress, the spatio-temporal variability of crop growth and the complexity of multi-year planting patterns remain major challenges for crop mapping. To address this issue, this study proposes a Mixture-of-Experts-based Spectral and Crop-sequence Temporal Network (SCTNet), which integrates heterogeneous data sources to improve the accuracy and robustness of early-season multi-class crop type mapping. Specifically, the proposed SCTNet model consists of two expert branches. The first branch employs a Swin Transformer to learn spatial and spectral-temporal features of crops from current-season optical satellite imagery, while the second branch utilizes an LSTM network to capture long-term dependencies from historical crop planting sequences. A class-aware gating module is then used to dynamically fuse the predictions of both experts for each crop type. The proposed model was validated on five major crop types across eight sites with diverse geographic conditions across the U.S.  The result shows that our method can map five crops with an overall accuracy of 85 % at a relatively early date (early July), although the local heterogeneity of agricultural landscapes potentially impacts the accuracy during the early stages. Our findings highlight the importance of leveraging historical crop planting sequences to complement current-season spectral time series for crop mapping, which contributes to improved accuracy in early-season crop classification and enhanced generalization across domains.

Acknowledgement: We acknowledge the Minnesota Supercomputing Institute (MSI) at the University of Minnesota for providing and maintaining the GPU servers that supported this research.

References: 
Zang, Y., Zhou, J., Chen, X., Liu, T., Shen, M., Yang, W., Zhu, X., Zhang, F., Chen, J., 2026. Integrating classifier transfer and sample transfer strategies for in-season crop mapping based on sample weighting technique. Remote Sensing of Environment 334, 115208. https://doi.org/10.1016/j.rse.2025.115208
You, N., Dong, J., 2020. Examining earliest identifiable timing of crops using all available Sentinel 1/2 imagery and Google Earth Engine. ISPRS Journal of Photogrammetry and Remote Sensing 161, 109–123. https://doi.org/10.1016/j.isprsjprs.2020.01.001
Zhang, H.K., Shen, Y., Zhang, X., Li, J., Yang, Z., Xu, Y., Zhang, C., Di, L., Roy, D.P., 2025. Robust and timely within-season conterminous United States crop type mapping using Landsat Sentinel-2 time series and the transformer architecture. Remote Sensing of Environment 329, 114950. https://doi.org/10.1016/j.rse.2025.114950
Xie, Y., Nhu, A.N., Song, X.-P., Jia, X., Skakun, S., Li, H., Wang, Z., 2025. Accounting for spatial variability with geo-aware random forest: A case study for US major crop mapping. Remote Sensing of Environment 319, 114585. https://doi.org/10.1016/j.rse.2024.114585
Wu, Y., Chen, X., Liao, C., Xu, X., He, Y., Wang, J., Wang, T., 2025. Generating crop type labels from historical annual crop inventory data with an ensemble learning method. Computers and Electronics in Agriculture 237, 110670. https://doi.org/10.1016/j.compag.2025.110670

A Sea of Data: Screening for Interactions Between Plant Immune Receptors & Elicitors from Mycotoxigenic Fungi

Brian D. Rutter¹, Tiana Roth¹, Avery Hickcox¹, Mengying Wang¹, and Josiah Mutuku¹

¹2Blades, University of Minnesota – Twin Cities, St. Paul, Minnesota, USA 

Mycotoxigenic fungi pose a serious threat to global food security and human health. In the United States, $3-4 billion is spent annually to mitigate the impacts of mycotoxins on agriculture. Contaminated grains are often destroyed outright or sold at severely marked down prices, leading to substantial economic losses. World-wide, and especially in developing countries, over 4.5 billion people consume food with unregulated mycotoxin levels. The presence of mycotoxins can lead to serious health effects, including cancers, birth defects, organ damage or even death.

2Blades is non-profit organization dedicated to developing new, disease-resistant crops, with groups based at The Sainsbury Institute in Norwich, England and at the University of Minnesota (UMN), St. Paul. The 2Blades UMN Group uses a maize protoplast-based assay to screen its extensive library of plant pattern recognition receptors (PRR), which are genes involved in early pathogen detection. For the past year, the 2Blades UMN Group has screened thousands of interactions between fungal elicitors and the PRR genes and uncovered new and exciting interactions. Some of the genes discovered through screening are currently being transformed into maize. The goal is to develop more resilient crops with durable resistance to mycotoxigenic fungi and put them in the hands of growers.

Acknowledgement: This work is supported by the USDA (58-5062-014), The Foundation for Food & Agriculture Research (FFAR; 23-000769) and The Minnesota Corn Growers Association (MCGA; 6138-24DD).

Engineering a Broad-Spectrum Geminivirus Vector for Genome Editing

Ashish Srivastava¹, Jitendra Kumar¹, Anshu Alok², Daniel Voytas², Feng Zhang², and Shahryar Kianian³

¹Department of Plant Pathology, University of Minnesota, Saint Paul, MN 55108, USA
²Department of Genetics, Cell Biology and Development and Center for Precision Plant Genomics, University of Minnesota, Saint Paul, MN 55108, USA
³USDA-ARS Cereal Disease Laboratory, St. Paul, MN, 55108 USA

Despite causing significant crop losses, plant virus genomes enable popular tools like VIGS for efficient gene knockdown without stable transformation. This study evaluates the potential of Geminivirus-based Virus-Induced Genome Editing (VIGE) as a robust alternative to stable transformation. Due to their high replicon copy number, geminiviruses facilitate elevated expression of editing components, significantly enhancing editing efficiency and scalability. In this study, we evaluated the plant genome editing systems using various Cas constructs and delivered it using Wheat dwarf India virus (WDIV), and Ageratum yellow leaf curl betasatellite (AYLCB) genome. 

Previous studies in our laboratory demonstrated the efficient delivery of Cas effectors and gRNA in Nicotiana benthamiana. Building upon this, we evaluated the performance of various guide RNA (gRNA) architectures and miniCas combinations in recalcitrant crops using a frame-shift restoration RUBY assay (assay developed by the Zhang Lab). demonstrate the robust potential of the WDIV-satellite molecule system for genome engineering across a broad host range. This work advances virus-mediated genome engineering platforms and establishes a critical foundation for enhancing editing efficiency in cereal and other recalcitrant crop species.

Branching out: How genotype and row spacing shape soybean branch angle

Lara Waldt¹ , Suma Sreekanta¹ , Gary Muehlbauer¹ , Aaron Lorenz¹ 

¹ Department of Agronomy and Plant Genetics, University of Minnesota, Saint Paul, MN, USA 

Soybean shoot architecture plays a critical role in light interception, resource allocation, and yield potential, yet remains underexplored compared to major crops such as wheat and rice. Branch angle is an important trait influencing canopy coverage and structure. This project investigates the impact of row spacing and genotype on soybean shoot morphology. A total of 10 genotypes chosen for diverse branching habits were grown using two row spacings – 30 inches and 15 inches. We assessed the interaction between genotype and row spacing on branch angle. Using an analysis of variance on data collected in 2022 and 2023, we found that both genotype and row spacing had a strong effect on branch angle (p < 0.0001). Specifically, plants grown in wider rows had a significantly wider mean branch angle (40.2°) compared to those in narrow rows (35.7°). Genotype also had a strong effect, with branch angles ranging from 29.2° (PI404188A) to 45.6° (PI612717) across the 10 genotypes. Additionally, year effects were significant (p < 0.01), as the magnitude of the row spacing effect differed between years. The interaction between genotype and row spacing, however, was not significant. Together, these complementary studies advance our understanding of how agronomic practices and genetic factors shape soybean shoot architecture. The findings will support soybean breeding and management strategies aimed at optimizing plant structure for maximum yield. 

Acknowledgements: This work is graciously supported by Minnesota Soybean Research and Promotion Council and USDA National Institute of Food and Agriculture AFRI program. 

DNA repair under heat: DNA Polymerase λ modules heat stress-induced mutagenesis in plants

Clair M. Wootan^1,2, John Lutterman^c, Nathan Springer¹, Xiaosa Xu⁴, Feng Zhang^1,2 

¹ Department of Plant and Microbial Biology, University of Minnesota, Saint Paul, MN 55108  
² Center for Precision Plant Genomics, University of Minnesota, Saint Paul, MN 55108 
³ Department of Plant Biology, University of California, Davis, CA 95616
⁴ Department of Genetics, Cell Biology and Development, University of Minnesota, Minneapolis, MN 55455

Environmental stress can elevate mutation rates, a phenomenon known as stress-induced mutagenesis, yet the molecular mechanisms underlying this process in eukaryotes remain unclear. Increased mutation rates have often been ascribed to DNA damage and replication errors. However, how heat stress effects DNA repair is a matter yet undiscovered. Here we investigate the impact of heat stress on DNA repair and mutagenesis in Arabidopsis thaliana using CRISPR-Cas9 induced double-stranded-breaks, whole genome sequencing, and single-cell transcriptomics. In addition to significantly increasing editing frequency, heat stress also shifted repair towards single base pair insertions. We identify a heat-inducible, error-prone polymerase, DNA Polλ as a key mediator of these altered repair profiles. Given the versatility of DNA Polλ in multiple repair pathways and its capacity to introduce insertions and substitutions, we further examined its role in non-targeted DNA damage occurring during heat stress. Through genome-wide analysis of somatic mutations, we found that heat-induced mutagenesis is dependent on DNA Polλ. Single-cell transcriptomic profiling showed that DNA Polλ expression is tightly regulated and enriched in the central zone of the shoot apical meristem. Together our results establish DNA Polλ as a central mediator of heat stress-induced genetic variation and provide mechanistic insight into how plants integrate environmental signals with DNA repair to promote adaptability.  

Acknowledgement: This work is supported by the National Science Foundation (IOS-2040218 and IOS-2206920) awards and the Bayer Crop Science/University of Minnesota Multi-functional Agricultural Graduate Student Fellowship.