Validation of a high-confidence regulatory network for gene-to-NUE phenotype in field-grown rice

Authors: SHANKS, CARLY - New York University; HUANG, JI - New York University; CHENG, CHIA-YI - National Taiwan University; SHIH, HUNG-JUI - New York University; Brooks, Matthew; ALVAREZ, JOSE - New York University; ARAUS, VIVIANA - New York University; SWIFT, JOSEPH - Salk Institute Of Biological Studies; HENRY, AMELIA - International Rice Research Institute; CORUZZI, GLORIA - New York University

Submitted to: Frontiers in Plant Science
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 11/1/2022
Publication Date: 11/25/2022
Citation: Shanks, C.M., Huang, J., Cheng, C., Shih, H., Brooks, M.D., Alvarez, J.M., Araus, V., Swift, J., Henry, A., Coruzzi, G.M. 2022. Validation of a high-confidence regulatory network for gene-to-NUE phenotype in field-grown rice. Frontiers in Plant Science. https://doi.org/10.3389/fpls.2022.1006044.
DOI: https://doi.org/10.3389/fpls.2022.1006044

Interpretive Summary: Water and nitrogen availability are two of the most important factors for plant growth and in order to address global climate change the development of crop varieties with increased nitrogen and water use efficiency is necessary. Rice is a staple crop for 3.5 billion people and in this study we used trait and gene expression data from 19 varieties of rice grown in a 2x2 matrix of controlled nitrogen and water conditions to identify 18 transcription factors that predicted to be important for grain yield. Importantly, the expression of these transcription factors is dependent on nitrogen-by-water interactions and therefore likely integrate these signals. These results can now be applied to develop/breed rice plants with improved rice yields in marginal, low N-input, drought-prone soils – which are increasing in the face of climate change.

Technical Abstract: Nitrogen (N) and Water (W) - two resources critical for crop productivity – are becoming increasingly limited in soils globally. To address this issue, we aim to uncover the gene regulatory networks (GRNs) that regulate nitrogen use efficiency (NUE) - as a function of water availability - in Oryza sativa, a staple for 3.5 billion people. In this study, we infer and validate GRNs that correlate with rice NUE phenotypes affected by N-by-W availability in the field. We did this by exploiting RNA-seq and crop phenotype data from 19 rice varieties grown in a 2x2 N-by-W matrix in the field. First, to identify gene-to-NUE field phenotypes, we analyzed these datasets using weighted gene co-expression network analysis (WGCNA). This identified two network modules ("skyblue" & "grey60") highly correlated with NUE grain yield (NUEg). Next, we focused on 90 TFs contained in these two NUEg modules and predicted their genome-wide targets using the N-and/or-W response datasets using a random forest network inference approach (GENIE3). Next, to validate the GENIE3 TF'target gene predictions, we performed Precision/Recall Analysis (AUPR) using nine datasets for three TFs validated in planta. This analysis sets a precision threshold of 0.31, used to "prune" the GENIE3 network for high-confidence TF'target gene edges, comprising 88 TFs and 5,716 N-and/or-W response genes. Next, we ranked these 88 TFs based on their significant influence on NUEg target genes responsive to N and/or W signaling. This resulted in a list of 18 prioritized TFs that regulate 551 NUEg target genes responsive to N and/or W signals. We validated the direct regulated targets of two of these candidate NUEg TFs in a plant cell-based TF assay called TARGET, for which we also had in planta data for comparison. Gene ontology analysis revealed that 6/18 NUEg TFs - OsbZIP23 (LOC_Os02g52780), Oshox22 (LOC_Os04g45810), LOB39 (LOC_Os03g41330), Oshox13 (LOC_Os03g08960), LOC_Os11g38870, and LOC_Os06g14670 - regulate genes annotated for N and/or W signaling. Our results show that OsbZIP23 and Oshox22, known regulators of drought tolerance, also coordinate W-responses with NUEg. This validated network can aid in developing/breeding rice with improved yield on marginal, low N-input, drought-prone soils.