NutriNet: A Network Inspired Approach to Improving Nutrient Use Efficiency (NUE) in Crop Plants

Pi：Gloria Coruzzi(纽约大学)Copis：Dennis Shasha(纽约大学)，Stephen Moose(伊利诺伊大学香槟分校)，Sandine Ruffel和Gabriel Krouk(INRA，法国蒙彼利埃)高级人员：Manpreet Katari(纽约大学)和W.Richard McCombie(冷泉港实验室)提高作物的养分利用效率(NUE)对于改善未来气候变化的影响以及可持续地提高全球作物产量以满足预计的粮食和能源需求至关重要。NutriNet项目试图识别和比较生物上相连的基因网络，这些基因网络的集体表达模式可以预测拟南芥和玉米中NUE的表型变异。这种跨物种网络启发的方法可以很容易地应用于许多其他重要的经济性状，并适用于其他作物。NutriNet方法的优势包括：i)利用拟南芥基因和蛋白质相互作用的详细数据集，为数据贫乏的作物物种的分析提供信息；ii)识别可应用于分子育种计划的健壮的网络模块。原则证明研究将证明调节氮同化和再动员的网络模块(但不一定是候选基因)既保守又特定于物种的特征。在这个项目中产生的新知识将包括基因发现，调控电路的阐明，以及更好地理解推动作物生产力的营养生理学的分子基础。作为一项实用的成果，将开发网络启发的分子育种工具，有望在选择具有改良NUE的基因类型方面比候选基因方法表现得更好。NutriNet团队将系统生物学、植物生理学和作物基因组学方面的专业知识联系起来，以增加对作物养分利用的基本了解。该项目为纽约和伊利诺伊州的博士后科学家、研究生和本科生提供多学科培训。通过纽约大学生物学家和计算机科学家的共同指导，高中生将被介绍给系统生物学。此外，由于公众对营养高效作物的广泛兴趣，该项目团队将通过在伊利诺伊玉米育种者学校的外联活动吸引受众，利用伊利诺伊大学的开创性努力和悠久的历史，与育种者合作，了解作物对营养的反应和氮素利用的育种。基因组测序、功能基因组学和计算工具的最新进展使人们能够系统地了解关键的生理和发育过程，包括模式植物拟南芥的NUE。然而，将这种“网络知识”从拟南芥转化到作物中，以潜在地增强作物物种在农业上的重要表型，仍然具有挑战性。该项目的目标是开发网络连接的基因模块，通过利用拟南芥网络知识，可以用来预测作物氮素利用效率的结果。该项目通过开发如下新的数据集和分析方法来实现这一目标：1)将NUE的表型变异与新的和现有的营养响应基因表达谱数据相结合，这使得能够开发利用拟南芥和玉米遗传多样性的训练集；2)使用裂根实验设计来识别在根冠N信号中起作用的进化保守的基因机制，该机制可能控制根在土壤中寻找养分；3)使用生物信息学管道定义预测NUE特性的网络模块，将拟南芥的“网络知识”与玉米转录组数据相结合，以生成NutriNet模块，该模块将被用于验证和测试其基于基因表达预测NUE的能力；以及4)使用从NutriNet模块获得的信息从不同的种质池中选择具有最佳NutriNet配置的单个基因型，然后将在实验室(拟南芥)和田间(玉米)评估这些基因的改良的NUE特性。对实验室到田间结果的比较分析将直接评估从拟南芥到玉米的网络知识的“转化”，作为一般原则的证明，这可以应用于其他网络和物种。所有数据和生物资源将根据请求提供，并可通过长期数据和种质储存库获取。

PI: Gloria Coruzzi (New York University)CoPIs: Dennis Shasha (New York University), Stephen Moose (University of Illinois at Urbana-Champaign), Sandrine Ruffel and Gabriel Krouk (INRA, Montpellier, France)Senior Personnel: Manpreet Katari (New York University) and W. Richard McCombie (Cold Spring Harbor Laboratory)Improving nutrient use efficiency (NUE) in crop plants is critical to ameliorating the impacts of future climate change and to sustainably increasing global crop yields to meet projected food and energy demands. The NutriNet project seeks to identify and compare biologically connected gene networks whose collective expression patterns are predictive of phenotypic variation in NUE in Arabidopsis and maize. This cross-species network inspired approach may be readily applied to many other economically important traits, and adapted to other crops. The advantages of the NutriNet approach include: i) exploiting detailed datasets for gene and protein interactions in Arabidopsis, to inform analysis of data poor crop species, and ii) identification of robust network modules that can be applied in molecular breeding programs. Proof-of-principle studies will demonstrate both conserved and species-specific features of network modules (but not necessarily candidate genes) regulating nitrogen assimilation and remobilization. The new knowledge generated in this project will consist of gene discovery, elucidation of regulatory circuits, and a better understanding of the molecular basis for nutrient physiology that drives crop productivity. As a practical deliverable, network-inspired molecular breeding tools will be developed that are expected to perform better than candidate gene approaches in selecting genotypes with improved NUE. The NutriNet team links expertise in systems biology, plant physiology, and crop genomics, to increase the fundamental understanding of crop utilization of nutrients. The project offers multidisciplinary training to postdoctoral scientists, graduate and undergraduate students in New York and Illinois. High school students will be introduced to systems biology through co-mentorship by biologists and computer scientists at NYU. In addition, because of the broad public interest in nutrient-efficient crops, the project team will engage audiences through outreach activities at the Illinois Corn Breeders' school to leverage the pioneering efforts and long history of the University of Illinois in concert with breeders to understand crop responses to nutrients and breeding for nitrogen utilization.Recent advances in genome sequencing, functional genomics, and computational tools enable a systems level understanding of key physiological and developmental processes including NUE in the model plant Arabidopsis thaliana. However, translating this "network knowledge" from Arabidopsis to crops to potentially enhance agriculturally important phenotypes in crop species remains challenging. The goal of this project is to develop network-connected gene modules that can be used to predict the outcome of NUE in crops, by exploiting Arabidopsis network knowledge. The project approaches this goal by developing novel data sets and analytical methods as follows: 1) integrating phenotypic variation for NUE with new and existing data for nutrient-responsive gene expression profiles which allows for the development of a training set that exploits the power of genetic diversity from both Arabidopsis and maize; 2) using a split-root experimental design to identify evolutionarily conserved gene mechanisms that function in root-shoot N-signaling that may control root foraging for nutrients in the soil; 3) defining network modules predictive of NUE traits using a bioinformatics pipeline to combine Arabidopsis "network knowledge" with maize transcriptome data to generate NutriNet modules that will be validated using and tested for their ability to predict NUE based on gene expression; and, 4), using information derived from NutriNet modules to select individual genotypes that possess optimal NutriNet configurations from diverse germplasm pools which will then be evaluated for improved NUE traits in the lab (Arabidopsis) and field (maize). A comparative analysis of lab-to-field results will directly assess the "translation" of network knowledge from Arabidopsis to maize to serve as a general proof-of-principle, which can be applied to other networks and species. All data and biological resources will be available upon request and accessible through long-term data and germplasm repositories.