Analysis of the dynamic nitrogen response of root and shoot transcripts in arabidopsis thaliana to gain insights on long-distance nitrogen signaling interactions

Genetic advances account for less than 2% of global annual crop yield growth, with nitrogen (N) fertilizer use comprising the remaining 98% and its use expected to increase to 138Tg to meet crop demand by 2030. This overapplication of N fertilizer results in direct economic and environmental consequences as plants only take up 30-50% of the available soil N, with the rest being lost to the environment. To combat this, many studies have focused on improving the N use efficiency of plants through understanding the mechanisms and pathways involved in the N-responsive long-distance signaling pathways between roots and shoots. These studies, however, often fall short of integrating data across time and space due to various biological constraints, while others attempt to use time series models not designed for biological systems. Here, I propose a new time series model that is suitable for biological systems, accounting for these constraints. This model was applied to unevenly spaced, multivariate time-series data from root and shoot tissue in Arabidopsis thaliana in response to a N signal. From 2,173 shoot and 568 root differentially expressed genes, the model predicted 3,078 significant granger-causal interactions. Of these, 2,012 interactions have a root causal gene while 1,066 interactions have a shoot casual gene. Of the total 1,007 different causal genes from either organ, 384 have been known or predicted to produce a mobile gene product, possibly involved in N signaling. The interactions were then globally explored using a bioinformatics pipeline that included gene ontology term analysis, network analysis, transcription factor binding, as well as exploring causal genes involved in known N-responsive signaling pathways and interactions. Further, an A. thaliana grafting method is put forward to validate selected bioinformatically-supported predictions. Future directions are then discussed with respect to using the time series model to integrate shoot metabolite data to root/shoot transcriptomic data to identify possible N-responsive gene-metabolite relationships.

Name thesis:degree_name

M.S.

Level thesis:degree_level

Thesis

Discipline thesis:degree_discipline

Plant Biology

Grantor

University of Illinois at Urbana-Champaign

Year dc:date

2020

Author dc:creator

• Heerah, Sachin A

Contributors dc:contributor

• Marshall-Colón, Amy
• Ainsworth, Lisa
• Briskin, Donald
• Guerrier, Stéphane

• nitrogen
• time series
• signaling

Statement dc:rights

• Copyright 2020 Sachin Heerah

Language dc:language

en

Handle dc:identifier

http://hdl.handle.net/2142/108639

OAI identifier oai:identifier

oai:www.ideals.illinois.edu:2142/108639