Unravelling miRNA regulation in yield of rice (Oryza sativa) based on differential network model

Jihong Hu, Tao Zeng, Qiongmei Xia, Qian Qian, Congdang Yang, Yi Ding, Luonan Chen, Wen WangScientific 2018 · DOI

TL;DR

This paper introduces GRN-DET, a novel algorithm for identifying key regulatory microRNAs (miRNAs) involved in rice yield by analyzing differential gene regulatory networks. It successfully identifies specific miRNAs and their roles in crucial developmental stages, offering new insights into yield improvement strategies.

What Problem Does This Paper Address?

Improving crop yield is a critical challenge, and microRNAs (miRNAs) are known to play regulatory roles in plant development and agronomic traits. However, previous studies have largely focused on individual miRNAs or gene expression levels, lacking a systematic, network-level understanding of how miRNAs differentially regulate yield in high- versus low-yield rice varieties. Specifically, there was a need for a method that could identify key regulatory miRNAs even when their expression levels don't change significantly, but their regulatory networks do. This gap hinders the development of targeted strategies for crop improvement.

Methods

The core innovation of this paper is the Gene Co-expression Network differential edge-like transformation (GRN-DET) algorithm. Unlike traditional methods that rely on large sample sizes and compare overall expression levels, GRN-DET focuses on "edge-like correlations" within individual samples. This allows for the construction of differential networks even with limited samples.

Here's a breakdown of the key methodological aspects:

  • Differential Edge-like Transformation (DET): This is the foundational technique. It transforms raw gene expression data into "edge-like correlations" between gene pairs within a single sample. These correlations capture the dynamic relationships between genes, even if their individual expression levels are stable.
  • GRN-DET: This integrates DET with known miRNA-gene and gene-transcription factor (TF) interaction data. It constructs miRNA-gene-TF co-regulation networks. The "differential" aspect comes from analyzing the variances of miRNA expression and the co-variances of miRNA-mRNA interactions across different conditions (e.g., high vs. low yield).
  • Network Construction: The method builds tissue-specific networks based on these edge-like correlations. The strength of the interaction (edge weight) is determined by these correlations.
  • Key miRNA Identification: miRNAs are identified as "key" if they exhibit differential regulation (i.e., changes in their regulatory networks) between conditions, even if their expression levels are not significantly different. This is quantified using a "PCC-induced key-associated score."
  • Phytohormone Pathway Analysis: The identified regulatory networks are analyzed for enrichment in known phytohormone signaling pathways (auxin and brassinosteroid) to understand their broader biological context.

Datasets

The study utilizes several datasets:

  • Public Small RNA and RNA-seq Data: Data originally used to investigate rice grain filling, specifically from superior and inferior spikelets of Oryza sativa spp. japonica cv. 'Xinfeng 2'. This dataset was used for initial validation of the GRN-DET method.
  • In-house Small RNA and Transcriptome Sequencing Data: This was generated from the rice variety IR64, comparing an ultra-high yield location (Taoyuan) with a natural yield control location (Jinghong). Data was collected from three developmental stages: tillering, panicle branching, and grain filling.
  • Literature-derived Data: A collection of 150 reported yield-associated miRNAs and their targets from existing literature and degradome data was used to guide the network construction and validation.

Key Insights and Ideas

  • Network-level regulation is crucial: The study highlights that significant regulatory changes can occur at the network level, even without substantial changes in individual miRNA expression. This challenges the sole reliance on differential expression analysis for identifying regulatory molecules.
  • DET enables single-sample network analysis: The development of DET is a significant methodological advancement, allowing for the construction and comparison of regulatory networks from limited sample sizes, which is common in many biological studies.
  • miRNAs orchestrate complex pathways: The identified miRNAs are not acting in isolation but are part of intricate networks that interact with transcription factors and influence key developmental processes.
  • Phytohormone crosstalk is central to yield: The research reveals that miRNAs involved in auxin and brassinosteroid signaling pathways play a significant role in regulating rice yield across different developmental stages. This suggests that manipulating these hormonal pathways via miRNAs could be a strategy for yield improvement.

Key Results

  • Identification of Key Regulatory miRNAs: GRN-DET successfully identified several miRNAs, including osa-miR171 and osa-miR1432, as key regulators of rice yield, particularly in panicle branching and grain filling.
  • Discovery of Sample-Specific miRNAs (SmiRNAs): The method identified SmiRNAs that exhibit differential regulation in their networks between different conditions or tissues, even if their expression levels were not significantly different.
  • Validation of GRN-DET: The method's effectiveness was validated using public datasets, where it identified known yield-related miRNAs and revealed novel regulatory insights.
  • Phytohormone Pathway Involvement: The analysis confirmed the involvement of miRNAs in auxin and brassinosteroid signaling pathways, which are known to influence plant growth and development, thus impacting yield.
  • Cross-talk between miRNAs and Hormones: The study suggests complex interactions between miRNAs and phytohormones, such as auxin and BR, contributing to yield regulation.

Limitations and Future Work

  • Limitations of DET: While DET allows for single-sample analysis, the paper acknowledges that it is a linear correlation-based method. The authors suggest that future work could explore non-linear correlation methods to capture more complex regulatory relationships.
  • Experimental Validation: While some miRNAs were validated using qRT-PCR, further experimental validation of the identified regulatory networks and the functional roles of the key miRNAs in planta would strengthen the findings.
  • Broader Application: The authors suggest that the GRN-DET method could be applied to other crops and biological processes where understanding differential regulatory networks is crucial.
  • Integration with other 'omics': Future research could integrate GRN-DET with other 'omics' data (e.g., epigenomics, proteomics) to build more comprehensive regulatory models.

Summary updated

Abstract

Rice (Oryza sativa L.) is one of the essential staple food crops and tillering, panicle branching and grain filling are three important traits determining the grain yield. Although miRNAs have been reported being regulating yield, no study has systematically investigated how miRNAs differentially function in high and low yield rice, in particular at a network level. This abundance of data from high-throughput sequencing provides an effective solution for systematic identification of regulatory miRNAs using developed algorithms in plants. We here present a novel algorithm, Gene Co-expression Network differential edge-like transformation (GRN-DET), which can identify key regulatory miRNAs in plant development. Based on the small RNA and RNA-seq data, miRNA-gene-TF co-regulation networks were constructed for yield of rice. Using GRN-DET, the key regulatory miRNAs for rice yield were characterized by the differential expression variances of miRNAs and co-variances of miRNA-mRNA, including osa-miR171 and osa-miR1432. Phytohormone cross-talks (auxin and brassinosteroid) were also revealed by these co-expression networks for the yield of rice.
BibTeX
@inproceedings{hu2018unravelling,
  author = {Jihong Hu and Tao Zeng and Qiongmei Xia and Qian Qian and Congdang Yang and Yi Ding and Luonan Chen and Wen Wang},
  title = {{Unravelling miRNA regulation in yield of rice (Oryza sativa) based on differential network model}},
  booktitle = {Scientific},
  year = {2018},
  doi = {10.1038/s41598-018-26438-w},
}
Figure