A single cell view of the spatiotemporal nitrogen response in roots

项目摘要/摘要：本提案旨在研究调节营养控制的机制 通过利用新的单细胞表达数据来研究侧根(LR)的发育。最终目的是要发展 氮素吸收和氮素利用效率(NUE)提高的植物。这将有助于改进应用程序 过量的氮肥是水和空气污染的主要原因，对人类健康造成负面影响。植物根部 可以感觉到土壤中的硝酸盐水平，这会诱导LR寻找营养。LRS启动胚胎后发育 从根的周周层中分化的细胞，这些细胞去分化形成“创始细胞” 上升到胚胎后的LRS分生组织。因此，LRs对于根的发育可塑性和营养是至关重要的 收购。之前使用绿色荧光蛋白标记的细胞系的研究发现，根中的N-反应是细胞类型特有的。 然而，这些研究缺乏必要的单细胞分辨率和时间成分来确定N- 调节细胞类型的过渡状态(即从周柱细胞到创始细胞)。这项建议旨在：i)使用单一- 细胞N-反应时间序列数据，以创建发育轨迹，模拟从周周到周周的转变 方正细胞和ii)识别/验证调节LR启动的转录因子，以响应N-感觉。要做到这一点，单身- 细胞N-反应转录组数据将从N处理的拟南芥根中分析(目标1)。至 确定此数据集中的Tf？靶基因关系，将使用特定于细胞的N-Response数据来学习 基因调控网络(GRN)使用一种名为OutPredict的基于时间的机器学习算法。接下来， GRN中预测的TfA靶基因相互作用将使用TARGET进行验证，TARGET是一种基于根细胞的Tf 扰动分析(目标2)。最后，候选因子在LR发育和N-调节中的作用。 将在PLANTA中使用Tf突变体在表型分析中验证摄取(目标3)。在硅胶预选赛中 概述的方法应用于现有的单细胞数据的分析，从拟南芥的根，交叉与N- 来自全根的反应转录组数据被鉴定：1.一个创始细胞的假时间轨迹和N- 响应性转录因子(目标1)，2.预测转铁蛋白A靶基因相互作用的创始细胞N-反应GRN，以及3.A 方正细胞轨迹的N反应因子初步清单(表2)。这些初步的TF将是 对它们在N反应LR发育和15N吸收中的作用进行表型(目标3)。这项提议将是第一个 在PLANTA中收集单细胞N-应答数据并使用发育轨迹模拟LR发育 接近了。这项提议还提供了单细胞数据分析和机器的新的计算培训 GRN到PI的学习方法，这将补充她在植物分子生物学方面的实验技能， 并为她作为独立研究员的职业生涯提供了一个多方面的研究基础。这个项目 也将受益于赞助商环境，证明成功地向PI提供新的培训，以及， 建议的单小区和网络建模的理想实验室和协作者环境 实验。

Project Summary/Abstract: This proposal seeks to investigate the mechanisms that mediate nutrient control of lateral root (LR) development by exploiting new single-cell expression data. The ultimate goal is to develop plants with enhanced nitrogen (N)-uptake and N-use-efficiency (NUE). This would help ameliorate the application of excess N-fertilizer - a main cause of water and air pollution that negatively impacts human health. Plant roots can sense levels of nitrate in the soil, which induces LR foraging for nutrients. LRs initiate post-embryonically from differentiated cells in the pericycle layer of the root, which de-differentiate to form “founder cells” that give rise to the post-embryonic LRs meristems. Thus, LRs are vital for root developmental plasticity and nutrient acquisition. Previous studies using GFP-marked cell lines, found that N-responses in roots are cell-type specific. However, those studies lacked the single-cell resolution and time-component necessary to identify the N- regulation of transitional states of cell-types (i.e. pericycle-to-founder cell). This proposal aims to: i) use single- cell N-response time-series data to create developmental trajectories that model the transition from pericycle-to- founder cells and ii) identify/validate TFs that regulate LR initiation in response to N-sensing. To do this, single- cell N-response transcriptome data will be analyzed from N-treated Arabidopsis thaliana roots (Aim 1). To determine the TFàtarget gene relationships in this dataset, cell-specific N-response data will be used to learn gene regulatory networks (GRNs) using a time-based machine learning algorithm called OutPredict. Next, the predicted TFàtarget gene interactions in the GRN will be validated using TARGET, a root cell-based TF perturbation assay (Aim 2). Finally, the function of candidate TFs in N-regulation of LR development and N- uptake will be validated in planta using TF mutants in phenotyping assays (Aim 3). In a preliminary in silico analysis of the outlined approach applied to existing single-cell data from Arabidopsis roots, intersected with N- responsive transcriptome data from whole roots identified: 1. A founder cell pseudotime trajectory and N- responsive TFs (Aim 1), 2. A founder cell N-response GRN that predicts TFàtarget gene interactions, and 3. A preliminary list of N-response TFs from founder cell trajectories (Table 2). These preliminary TFs will be phenotyped for their role in N-responsive LR development and 15N-uptake (Aim 3). This proposal will be the first to collect single-cell N-response data in planta and model LR development using developmental trajectory approaches. This proposal also provides new computational training in single cell-data analysis and machine learning methods for GRN to the PI, which will complement her experimental skills-set in plant molecular biology, and provide her with a multi-faceted research foundation for a career as an independent researcher. This project will also benefit from a sponsor environment with proven success for providing new training to the PI, as well as, the ideal laboratory and collaborator environment for the for the proposed single-cell and network modeling experiments.