Source: IOWA STATE UNIVERSITY submitted to NRP
THE GENOMIC BASIS OF ROOT ARCHITECTURAL PHENOTYPES FOR IMPROVED NITROGEN CAPTURE
Sponsoring Institution
National Institute of Food and Agriculture
Project Status
COMPLETE
Funding Source
Reporting Frequency
Annual
Accession No.
1028575
Grant No.
2022-67012-37220
Cumulative Award Amt.
$225,000.00
Proposal No.
2021-08393
Multistate No.
(N/A)
Project Start Date
May 1, 2022
Project End Date
Jul 31, 2024
Grant Year
2022
Program Code
[A1152]- Physiology of Agricultural Plants
Recipient Organization
IOWA STATE UNIVERSITY
2229 Lincoln Way
AMES,IA 50011
Performing Department
GDCB
Non Technical Summary
Nitrogen (N) is a key nutrient required for plant growth. In intensively farmed regions, such as Iowa, many farmers apply N fertilizer to their fields to ensure that maize has plentiful N to sustain high yields. However, the productionand excess use of N fertilizer can pollute our air and waterways. In contrast, much of the world's agricultural systems lack sufficient N to sustain maize production. Therefore, developing maize lines that can grow well with less N are beneficial to agriculture worldwide.Since N is most often acquired by the roots in the soil, many root traits have been shown to improve N acquisition, particularly when the amount of N is reduced. For instance, roots that grow at a steep angle and enable deep rooting more quickly improve N acquisition. Though, little is known about how this trait is controlled.Fortunately, we can make use of the existing diversity within maize to tackle this problem. Much of the genotypic diversity we observe in maize is derived from transposable elements (TEs), which are small bits of DNA that are able to replicate and move around the genome. TEs have long been thought to be "junk DNA" accumulating over evolutionary time. New evidence has shown that some TEs are responsive to environmental stress and modulate the activity of neighboring genes potentially inducing changes in phenotype. Therefore, this project seeks to harness the potential of TEs as regulators of stress responses in order to develop maize lines that optimize N capture through better designed root systems.
Animal Health Component
20%
Research Effort Categories
Basic
60%
Applied
20%
Developmental
20%
Classification

Knowledge Area (KA)Subject of Investigation (SOI)Field of Science (FOS)Percent
2011510106065%
2031510102035%
Goals / Objectives
Our overarching goal is to bridge the gaps between the plant genome, transcriptome, and phenome by studying how transposable elements may regulate transcriptional and phenotypic responses to variable nitrogen availability.Goal 1: Assess the effect of N stress on root transcriptional networksBuild RNAseq libraries to measure transcript abundance over time in response to nitrogen stress in two maize lines with contrasting root growth angleIdentify genes responsive to nitrogen deficiency or differentially regulated by root phenotypeConstruct gene co-expression and regulatory networks to infer nitrogen stress effects on gene-gene relationshipsGoal 2:Relate TE variation in genomes to changes in gene expressionQuantify TE expression under high and low nitrogenCorrelate TE expression with differentially expressed genesCharacterize the activity of key nitrogen-responsive TEs or TE familiesCreate maize mutant where key TE or TE family is knocked out using CRISPR/Cas9Evaluate maize mutant under nitrogen stress in the greenhouse or field to evaluate transcriptional, phenotypic, and functional differences?Goal 3:Integrate genome, transcriptome, and phenome to predict phenotypesPhenotype root growth angle across a NAM family in the fieldQuantify expression of key genes and TEs that contribute to root phenotypes and N stress responses in the NAM familyDevelop a predictive model for root phenotypes and N-stress responses based on genome-scale TE presence/absence data
Project Methods
Methods:Build RNAseq libraries and RNA sequencing: Maize root tissue will be collected from 20-day-old seedling grown in sand culture with nutrient solution containing either high nitrogen (N) (15ppm) or low N (0.5 ppm) concentrations. Two genotypes will be evaluated: B73 and Oh7B. Three different tissue types will be collected from second and third whorl roots: the root tip (most distal 1.2 cm), the young root tissue from which lateral roots have not yet emerged, and the mature root tissue that contains lateral roots. Total RNA will be isolated from these samples using a Qiagen RNeasy Plant Mini Prep kit. A NEBNext Ulta II RNA kit will be used to prepare libraries for RNA sequencing.RNA sequencing will be performed at the Iowa State DNA Facility using an Illumina HiSeq 3000 with a target of >5 million reads per sample. Clean reads will be aligned to B73 to quantify transcripts. A secondary analysis will instead align all reads to Oh7B.Differential expression analysis: I will identify differentially expressed genes and transposable elements (TEs) using the package "DESeq" and R. Raw reads will be normalized to reads per million. Differentially expressed genes/TEs will be defined as those that exhibited at least a 10-fold change in expression (i.e. absolute values of the log(change) is equal to 1). Gene ontology enrichment using AgriGO will be used to infer the functional roles of differentially expressed genes.Network construction: To examine the interactivity of genes associated with root phenotypes and limiting N, I will construct a weighted gene co-expression network using the package "WGCNA" in R. I will assess differences in network topology and module membership between each genotype, N treatment, and root tissue type. I will also examine modules containing known N-responsive genes. Separate networks will also be constructed for each genotype to parse treatment-specific responses.I will also construct a gene regulatory network to test the role of TFs regulating root phenotypes. Our group is still evaluating the best tools to build regulatory networks, but a strong candidate is GENIST, which combines clustering with dynamic Bayesian network inference.Generation and evaluation of CRISPR/Cas9 maize mutants: Using the differential expression and network data, I will identify TE candidates that are associated with responses to N stress. For 2-4 of our top candidates, I will knockout the activity of targeted TEs using CRISPR/Cas9. For candidates that are individual TEs, we will design guide RNAs that target a unique sequence of the individual TE to compromise its activity. For candidates that are a TE family, we will design guide RNAs that target the target site duplication specific to that family and knockout the activity of multiple TEs across the genome.Maize mutants will be evaluated in a controlled environment in the greenhouse or in Iowa State's transgenic field for responses to N stress. Leaf and seed N content, shoot biomass, leaf greenness, and photosynthesis rates are some metrics that will be recorded to assess physiological function. I will also quantify the expression of known N-responsive genes and other novel N-responsive genes selected from previous research. If funds allow, transcriptomic and network analyses may be repeated in these mutants.Field design: Maize genotypes will be grown under high and low N conditions in the field in a split-plot design. The high N treatment will follow yearly N fertilizer recommendations for field corn in central Iowa while the low N treatment will not have any supplemental fertilizer applied. Two replicates of each genotype will be planted in each treatment. Each replicate consists of a 15 ft long, single-rowed plot. Each plot contains 20 individuals planted at 9 in spacing and 2 in depth. Small-scale experiments will be jab planted while large-scale experiments (>80 genotypes) will be planted using a cone planter.Plant performance measures: Data will be collected throughout the whole field season to assess N stress effects on plant growth and development. Plant height and leaf greenness will be regularly checked every 2 weeks. We will collect flowering notes. Shoot biomass may be collected at anthesis. Yield will be collected from 3 plants per plot at the end of the season to calculate total yield and 100 seed weight.Root phenotyping: Root crowns will be collected using "shovelomics". In short, a standard spade is used to excavate the volume of soil that is one spade head deep and in radius around the base of the plant. The root crown is soaked and washed with low pressure water to remove remaining dirt and imaged in a photo booth. ImageJ is used to measure root growth angle, which is the angle at which the roots grow relative to the soil surface.Building a model to predict root phenotypes and N stress responses: I will integrate all transcriptomic, genomic, and phenomic data to build a partial least squares regression model to predict root responses to N stress. TE presence/absence variation will first be quantified in the NAM founders using scripts published in Anderson et al. (2019). These scripts will facilitate pairwise comparisons between each NAM founder and B73. First, TE annotations will be lifted from B73v4 to B73v5. Second, 400 bp flanking sequences centered on the start and stop points of each TE will be collected and locally aligned on a NAM founder genome. "Present" TEs are those where the full flank mapped successfully to the NAM founder in a given search window. "Absent" TEs are those where only the outside 200 bp portions of the flank successfully map within the search window. "Unresolved" TEs are those that were unable to be determined as present or absent. This data will be used to identify structural variants.To build a model, I will integrate the structural variant and differential expression data to predict root phenotypes. PLS regression will create latent variables to best predict our phenotypic outcomes, for which several genotypic traits may contribute to. 80% of our field data will be used to train the model while the remaining 20% will test our model with 5-fold cross-validation.Efforts: TEs had long been called "junk DNA", but this research will demonstrate the important role that TEs potentially play in plant genomes. This work may be shown in educational settings to provide tangible examples on the function of TEs. This project as a whole will also provide experiential learning opportunities for undergraduate research assistants employed as part of this grant and public school teachers that participate in hands-on activities in various labs across Iowa State University.Evaluation:*key milestonesTotal RNA isolated from greenhouse samples is high quality.RNA sequencing achieved >5 million reads per sample with adequate coverage (at least 10x).Generate count tables of all samples.*Differentially expressed genes, TEs, and TE families are identified.Gene networks show topological differences between genotypes, root tissue types, and N treatments.*Identify 2-4 TE candidates based on differential expression and network analyses.*Develop and evaluate maize mutant lines where TE activity is knocked out.*Phenotype roots of B73 x Oh7B NAM family in the field under high and low N conditions.Quantify TE presence/absence variation in the NAM founders.Re-format NAM family genotype data to B73- or Oh7B-like.*Train and test PLS regression model.

Progress 05/01/22 to 07/30/24

Outputs
Target Audience:The target audience was primarily fellow genetics and genomics researchers in the maize community. As a result of our research, we hope to have generated some data resources that others in the community may use for their own explorations and hypothesis generation. This project also served as a means of training for young scientists. For this past year, one undergraduate student was employed on this project to assist with data collection and analysis. This undergraduate student then transitioned to a lab technician for 5 months after their graduation. We also mentored an elementary school science and math teacher to gain hands-on research experience. While no funds from this project were used to support that teacher, the teacher assisted on field research objectives and data collection. Changes/Problems:The major changes and problems we encountered dealt with time. When proposing this project, we were perhaps a little ambitious in the number of objectives we could feasibly accomplish in the span on two years. Some of these objectives, like those dealing with developing a gene edited mutant, were just not able to be accomplished in the span of this project. Also, because of personellimitations, we had to simplify some of our field experiments, particularly on the phenotyping work. So, we adapted by identifying a best representative whorl for root growth angle phenotyping in our larger germplasm collections rather than phenotyping the entire root crown. We also simplified the transcriptome studies by removing time as a variable. We realized that time was not closely related to our hypotheses and so we increased the number of maize lines we analyzed from 2 to 14 total to better capture variation in transcript abundance to phenotypic variation. What opportunities for training and professional development has the project provided?This project was the primary financial support for PD Stephanie Klein, and so supported her training and professional development by covering conference registration and travel costs. The grant also supported the training and development of two undergraduate students who assisted on project goals. They were able to build fundamental research skills, like data collection and management; gain specific training on concepts related to the project, like root and plant phenotyping; and had opportunities to present research at undergraduate research symposia held at Iowa State University or the Maize Genetics Meeting. How have the results been disseminated to communities of interest?The results have primarily been shared through oral and poster presentations at conferences. During the course of thisproject, there were2poster presentations and 1 oral presentation at two Maize Genetics meetings, 1 poster presentation at the Gordon Research Conference on quantitative genetics and genomics, and 1 oral presentation at the International Society of Root Research 12th Symposium. PD Klein has also been able to share research at 2 virtual seminars through the Zeavolution series and 1 invited seminar at UC-Davis. Results will also be shared in written publications: a review was published in 2022 while another research manuscript is still in preparation. What do you plan to do during the next reporting period to accomplish the goals? Nothing Reported

Impacts
What was accomplished under these goals? Goal 1.1: We built RNAseq libraries to measure transcript abundance in response to nitrogen stress in fourteen maize lines with contrasting root growth angle. We did not measure this over time as we adapted our experimental design, which no longer included time as an independent variable. Goal 1.2: We identified hundreds of genes that are differentially expressed based on either nitrogen deficiency or genotype-based differences. This is a list that can be mined by the greater community, but we have only focused on a handful as the strongest candidates. Goal 1.3: We did not construct gene co-expression and regulatory network to infer nitrogen stress effects on gene-gene relationships. While this kind of research is certainly possible given the transcriptome datasets we generated, but we did not have enough time during the timeline of this project to complete this objective. Also, our focus shifted to more of a genomics-based approach and candidate identification and away from a networks-based exploration. Goal 2.1: While much of our project began with intending to focus on TEs and TE expression, our research focus began to shift more towards gene activity. Though we made some interesting discoveries for TEs. We did not quantify TE expression under high and low nitrogen, but our transcriptome libraries can be used to quantify this in another project. Goal 2.2: Because we did not quantify TE expression, we could not correlated TE expression with differentially expressed genes. Instead, we looked at TE presence/absence variation in genomic regions surrounding differentially expressed genes based on contrasting phenotypes. Goal 2.3: Because we did not quantify TE expression, we did not identify key nitrogen-responsive TEs or TE families. Instead, we looked at which TE superfamilies were enriched or depleted in genomics regions associated with contrasting root phenotypes. So, this again was attributed to TE presence/absence variation. Goal 2.4: We did not create a maize mutant where key TE or TE family is knocked out using CRISPR/Cas9. This was mostly due to the timeline of the project. We did not identify strong candidates until late in the project. However, we hope this is something that can be pursued in future research. Goal 2.5: Because we did not generate a mutant, we did not validate a mutant under nitrogen stress. Goal 3.1: We successfully quantified root growth angle across the NAM founders and 5 additional check lines for three field seasons. We also phenotyped the root growth angle across a NAM family (~125 recombinant inbred lines) in the field. Goal 3.2: Using 12 key recombinant inbred lines with contrasting root phenotypes, we generated full RNAseq libraries that can then be used to evaluate the expression of candidate genes associated with the contrasting root growth angle. Goal 3.3: We did not develop a predictive model how we intitially conceived using presence/absence data. Instead, we used QTL mapping to identify genomic regions associated with contrasting root phenotypes which we could then integrate with our transcriptome data. We identified 1 QTL and identified a narrower region that is likely contributing to these contrasting phenotypes.

Publications


    Progress 05/01/23 to 04/30/24

    Outputs
    Target Audience:Our target audience has been primarily focused on other biological researchers in the field, especially those with ties to plant breeders that may see our results and apply similar methods to breeding pipelines for maize or other crops. Changes/Problems:We have had to make some adjustments based on resources available to us in the lab. The largest adjust has been changing our primary focus to a mapping-based approach rather than building a predictive model based on presence/absence data. This adjustment has been made primarily because of the quality of data available to us. We also had to adjust our RNA-sequencing strategy and placed a stronger focus on structural variation. We experienced mixed success with our nitrogen stress assays in the greenhouse, and so put a stronger emphasis on structural variation as a means of understanding the genomic basis of root phenotypes under variable environmental conditions. Hence, we invested more time in the bioinformatics pipeline development. Plus, this pipeline became more possible to achieve given new data that had been created by a collaborator. What opportunities for training and professional development has the project provided?With this work, I have been able to mentor two undergrads who gained experience with plant phenotyping, field experimental design and data collection, R coding, data visualization, and root biology. I have also been able to mentor a public school teacher in lab where we designed classroom activities geared toward 5th/6th graders integrating plant biology and math activities. How have the results been disseminated to communities of interest?During 2023, I shared my progress at multiple conferences: The Gordon Research Conference on Quantitative Genetics (poster presentation) The American Society of Plant Biologists Midwest Regional Meeting (poster presentation) The Maize Genetics Meeting (poster presentation) What do you plan to do during the next reporting period to accomplish the goals?Most of the work now in data analysis and writing. I have developed analytical pipelines with preliminary data that I will make use over the next few months.

    Impacts
    What was accomplished under these goals? Under goal 1, we identified multiple genotypes based on our field data that warrant investigation for transcriptional analysis. We built RNAseq libraries for 14 maize genotypes grown under high and low nitrogen conditions with variable root responses to nitrogen stress. We are still in progress of analyzing said sequencing libraries. Under goal 2, we have adapted our plans to relate TE variation in genomes to changes in gene expression. We instead created a computational tool allowing us to conduct comparative genomics work that enables to identify TEs and features that are shared between two maize genotypes. This gives us an understanding of structural variation over which we can lay our transcript data to identify TEs or other features that are potentially differentially transcribed. Under goal 3, we have used our field phenomics data to inform our transcriptome and genomics work. We completed phenotyping of a NAM family in the field and used the data to identify quantitative trait loci (QTL) associated with root growth angle. We have successfully identified a primary region of interest within which we will study genomic structural variation and transcript abundance. We are currently quantifying expression of key genes and TEs that contribute to root phenotypes and N stress responses in the NAM family.

    Publications


      Progress 05/01/22 to 04/30/23

      Outputs
      Target Audience:Our target audience for our work so far are scientists in the plant biology community. We have generated data and code that the greater community can use to supplement their research projects or use for hypothesis testing. We have not yet made this data publicly available, however, we would share it with collaborators if asked. Changes/Problems:We intially encountered some coding errors that delayed our transcript quantification from the RNA libraries. However, we have corrected those errors and since made progress, though our timeline is slightly delayed. What opportunities for training and professional development has the project provided?This grant has supported my technical training in new field of research, particularly the RNA-sequencing and analysis work. I have been able to share my finding in three seminars.This grant has also supported my professional development by providing funds for conference travel to the annual Maize Genetics Meeting andthe upcoming Gordon Research Seminar and Conference on Quantitative Genetics and Genomics. How have the results been disseminated to communities of interest?Primarily through seminars and poster presentations. My supervisor and I also wrote a short perspective presenting information that is th cornerstone of my research project: how we can use TEs to modulate transcriptional networks. What do you plan to do during the next reporting period to accomplish the goals?Move forward with the plan. We have generated a wealth of data over the last year and built some of the analytical framework to interpet it. Now is the time to continue analyzing data and replicating our fieldwork plans. For our computational work, we will make use of Iowa State's high performance computing cluster in order to perform our analyses or make do with R on our personal computers. We have developed a phenotyping pipeline that brute force phenotypes the root growth angle from the field and we've devised a reliable protocol to induce nitrogen stress in our field and greenhouse studies.

      Impacts
      What was accomplished under these goals? We slightly modified some of our goals, but are in keeping with what was proposed. Our accomplished goals were to: 1.1 Build RNAseq libraries to measure transcript abundance in three root tissuesin two maize lines with contrasting responses to nitrogen stress in root growth angle 1.2 Identify gene responsive to nitrogen deficiency or differentially regulated by root phenotype: This is still ongoing 2.1 Quantify TE expression under high and low nitrogen: This is still ongoing 3.1 Phenotype root growth angle across a NAM family in the field: We complete our first season of this work and will repeat this summer.

      Publications

      • Type: Journal Articles Status: Published Year Published: 2022 Citation: Stephanie P. Klein, Sarah N. Anderson, The evolution and function of transposons in epigenetic regulation in response to the environment, Current Opinion in Plant Biology, Volume 69, 2022