Dynamic re-programming of the cold transcriptome in Arabidopsis

Lead Research Organisation: University of Glasgow
Department Name: College of Medical, Veterinary, Life Sci

Abstract

Abstracts are not currently available in GtR for all funded research. This is normally because the abstract was not required at the time of proposal submission, but may be because it included sensitive information such as personal details.

Technical Summary

Our experimental approaches and bioinformatics developments demonstrate that dynamic changes in the transcriptome depend on both transcription and AS and that AS forms an important part of re-programming of the transcriptome.

We will continue to use ultra-deep RNA-seq of time-courses of plants treated with low temperatures and exploit our developments in RNA-seq analysis. The dynamic transcript-specific profiles allow network modelling at both the gene (transcriptional) and splicing (AS) levels and, significantly, gives the opportunity to integrate these different layers of regulation. The most up-to-date approaches in network analysis will be applied to existing data and the new data (constant light; first 3 h at 4C) allowing causal inferences in the networks to be predicted.

Key hub genes (genes with many connections to other genes) are likely to be important in re-programming the cold transcriptome. Selected genes will be characterised for their physiological (freezing tolerance and acclimation) and molecular phenotypes (validation of network connections - effects on AS target genes of mutants/over-expression lines).

Immediate/early effects on AS may be important to priming the plant (transcriptome) for temperature change before establishing the full low temperature response. We will therefore exploit the extensive expertise in phosphoproteomics at Dundee to investigate immediate/early phosphorylation of splicing factors in response to low temperatures and will investigate specific kinases involved in cold signalling.

Finally, genes with AS events important in establishing the cold transcriptome and tolerant/acclimation are expected to show an altered phenotype when knocked out. Also, the different transcript isoforms (which may code for different protein isoforms) may be responsible for the phenotype. We will express specific AS isoforms in mutant backgrounds to demonstrate the functionality of specific AS isoforms.

Planned Impact

The impact of this work will be the novel information on how gene expression is reprogrammed at the transcript, gene and network levels and how to apply the current technologies to perform such analyses in different plant and crop species. These approaches can be used by plant scientists examining gene function and gene expression of development, responses to external stimuli, metabolic pathways, biotechnology approaches, plant breeding etc.. In particular, the approaches will benefit many areas of crop genomics and biology but translation to crops requires an understanding of the potential and action to focus on generation of the transcriptomic resources. As such the main beneficiaries and users are the research sector, both academic and industrial.

The main challenge to maximising impact is to raise awareness of its potential and utility with the people who are most likely to use it and benefit their research. This needs to be done in a timely fashion so that other researchers can plan and design RNA-seq experiments with the goal of analysing data using comprehensive RTDs and the best available programmes to quantify transcript levels and establish network models.

The main Impact Objectives are to:

- Publicise the value of transcript-specific expression and understanding the dynamic transcriptome
- Encourage the development of RTDs for other plant/crop species.
- Engage with crop scientists and industry

To achieve these objectives:
1) The PIs/Co-Is will ensure community awareness by contacting research groups in the plant community with details of the project and how it will benefit them

2) The PIs/Co-Is will present regular updates of progress at national and international conferences and meetings on plant biology and genomics such as SEB, ICAR conferences etc as well as through more focussed meetings on RNA splicing, gene networks and the circadian clock. The PIs/Co-Is will also present their findings at invited seminars.

4) The PI/Co-Is will develop a strategy for interacting with crop scientists developing genomics approaches and engaging with industry.

5) Tools and resources for RNA-seq and network analysis will be released on the most appropriate websites for rapid uptake; a "cold" expression website will be established where users can examine expression profiles of their gene of interest in our time-course series.

6) Results will be published in timely fashion.

6) Public engagement activities.

7) Training and mentoring of the PDRAs.
 
Description Using ultra-deep RNA-seq of a time-course of plants exposed to cold, we demonstrated massive and rapid changes in both expression and alternative splicing, taking full account of the time-of-day variations in gene expression. We demonstrated the rapid and dynamic induction of changes in gene expression and alternative splicing of thousands of genes in the first few hours after the onset of cold. We have now constructed a novel gene co-expression regulatory network and a novel alternative splicing network. A major effort has been to dissect these networks to understand what they tell us about the complex regulation of expression in response to cold and to identify genes that are likely to be key regulators of plants' adaptation to reduced temperatures. Identifying such regulators will facilitate the improvement of crops by 1) revealing pathways likely to convey stress tolerance and maintain crop yield and 2) enabling targeted manipulation of the cold response.

The novel gene co-expression network was constructed by clustering the expression profiles of over 7,000 differentially expressed (DE) genes into 74 clusters and correlating these with profiles of over 900 individual protein-coding transcripts from DE transcription factor genes (TF regulators). Correlating TF regulators with clusters provides structure to the network. The gene network defined five inter-connected sub-networks. Two of these were associated with the widely studied cold responsive C-REPEAT-BINDING FACTOR/DEHYDRATION RESPONSE ELEMENT-BINDING PROTEIN (CBF/DREB) transcription factors (CBF1-3) and their target genes - the "CBF regulon". The other subnetworks were a novel, Early Cold Response sub-network and sub-networks of down-regulated gene clusters (Repressed group) and up-regulated gene clusters enriched for ribosome biogenesis factors and regulation of gene expression in chloroplasts and mitochondria (Ribo group). The gene network represents a significant advance over what has been published previously in terms of its resolution and separation of groups of gene clusters and their regulators. For example, although the CBFs are essential for the cold response, previous research in the field has identified other key transcription factors and has described CBF-dependent and CBF-independent pathways. The gene network provides a structure to investigate the relationships among such pathways. Taken together, predicted functions of gene clusters in the different subnetworks are consistent with overall biochemical and metabolic changes in the cold response but when superimposed on the network structure, specific functions are associated with different subnetworks. For example, the majority of core clock genes are associated with the two CBF subnetworks consistent with previous data on gating of the cold response by the clock and interaction between clock genes and CBFs. We have also examined enrichment of known TF binding sites among the gene clusters in the network. This has provided further insights into control of the various gene clusters. For example, WRKY and NAC factors are predominantly associated with early responding genes in the ECR as well as repressed genes; ERF/AP2 TFs are predominantly in the CBF sub-networks while MYB TFs have an extensive network in the Repressed gene subnetwork. We propose that current models of CBF regulons are greatly underestimated and show the integral relationship between clock genes and CBF regulated genes.


We have validated the gene network against previously published data on cold response genes (e.g. CFs and other transcription factors). We have mapped over 70 cold sensitive or cold tolerant mutants onto the network demonstrating that genes essential to the cold response and acclimation are distributed across the network reflecting the complexity of the response.


The gene co-expression network is complemented by an alternative splicing network. The splicing network was constructed by calculating the splicing ratio of each alternatively spliced transcript at each of the 26 time-points and clustering the profiles into 45 clusters. Analogous to the gene network, the expression profiles of protein-coding transcripts of over 400 DE splicing factors/RNA-binding protein (SF/RBP) genes were correlated to the splicing ratio cluster profiles to impose structure on the network. A paper describing this network will be submitted shortly.

The splicing network (top 1.5% of correlations) contains 26 clusters with 3,740 transcripts and 110 SF regulators. Two main groups of clusters have: 1) ca. 2,800 transcripts with adaptive changes in AS where cold-induced changes in AS are maintained throughout the cold period and 2) over 900 transcripts with cold-induced rhythmic changes in AS where the splicing ratio is stable at 20°C and rapidly becomes rhythmic at 4°C. The rhythmic AS clusters become rhythmic with peaks of expression at the same time (9 h after cold). This suggests a key cold-induced connection with the clock that switches the AS phenotype.

One of the objectives of the grant which was affected by COVID was to generate RNA-seq time-course data on plants exposed to cold and in constant light to examine the impact of the clock on the cold response and specifically on changes in rhythmic expression and AS. A great deal of effort was invested on finding conditions where seedlings survived cold and constant light (stressful conditions). Initial evidence from RT-qPCR showed that for some of the genes tested, expression remained rhythmic in cold/constant light samples confirming that they are under circadian control. The RNA-seq time-course data has been generated and is being analysed.

The high temporal resolution of our dataset also highlighted early expression and alternative splicing events and identified transcription factors and splicing factors/RNA-binding proteins whose expression changed significantly in the earliest time-points (0-3 h). We generated RNA-seq data to expand the first 3 h of cold treatment into 7 time-points (called immediate-early cold response data). This data has been analysed and has identified some the earliest affected TFs and SF/RBPs. On the basis of information from the main time-course and immediate early data we identified key factors for further studies.

The clock gene, RVE2, was the gene which showed the largest AS response to cold and had an AS switch such that RVE2 was only expressed in the cold. We obtained a mutant of RVE2 and performed initial assays with RT-qPCR. rve2 mutant seedlings have been shown previously to be sensitive to freezing stress both before and after acclimation. We showed that, with cooling, the diurnal profiles of CBF2, CBF3 and PRR5 were significantly affected in the rve2 mutant. These preliminary results clearly demonstrate that RVE2 is involved in regulation of expression of some clock and cold response genes. This is consistent with the induction of the CBF genes requiring the integration of low-temperature and clock-regulatory pathways and we hypothesise that RVE2 is an important integrator of temperature signaling and the clock. We therefore set up an RNA-seq time-course of the rve2 mutant and wild-type plants exposed to cold. One complete bio-rep was completed before the COVID lockdown but the two other bio-reps were lost due to the lockdown.

As a result of the COVID lockdown the University of Glasgow part of the joint grant was extended and was in abeyance for a period. Dr Allan James left the grant to work at the Lighthouse Lab in Glasgow, one of the UK-wide Lighthouse centres for PCR analysis of COVID samples. He was replaced by Dr Emily May Armstrong who worked for one year (October 2020-September 2021), mainly repeating the two bio-reps of the rve2 mutant experiment. Analysis of the data showed that RVE2 target genes comprise a cascade of gene expression initiated shortly after peak RVE2 expression at mid-nocturnal phase and extending into the first cool day. RVE2 functions mainly as a repressor of nuclear target gene expression; several of these nuclear encoded candidate genes have established roles in plants responses to temperature. The importance of RVE2 seems, at least for these nuclear-encoded genes, to lie in repressing a range of cold-responsive genes specifically during the dynamic temperature changes of cool nights. However RVE2 also activates expression of some chloroplast genes. We went on to find that in the rve2-2 mutant photosynthetic capacity was significantly higher than in wild-type plants following chilling, suggesting that RVE2 serves to limit photosynthetic performance at reduced temperatures.

Overall the data show that RVE2 integrates nocturnal temperature with priming photosynthetic capacity in anticipation of the break of day. A paper describing this work is currently under revision.

Experimental work on the Glasgow part of the grant finished on 30/09/2021 and the grant ended on 31/03/2022.

A second splicing factor whose expression and alternative splicing changed very rapidly in response to cold was RBM39A. Various analyses of a rbm39a mutant under the Dundee part of the grant showed that it affected putative target genes.
Exploitation Route The data should be useful for plant breeders in terms of cold tolerance.
Sectors Agriculture, Food and Drink
 
Description Lab blog 
Form Of Engagement Activity Engagement focused website, blog or social media channel
Part Of Official Scheme? No
Geographic Reach International
Primary Audience Public/other audiences
Results and Impact Allan James, researcher co-investigator on my grants BB/G008752/1 and BB/P006868/1, has set up a blog https://abouttimeresearch.com which covers both our collaborative work with John Brown's group at Dundee and Katherine Denby at York and various activities and insights into science in our lab such as the residency and exhibition of artworks by Ally Wallace, which was funded by the Leverhulne Trust Artist in Residence scheme.
Year(s) Of Engagement Activity 2014,2015,2016,2017,2018
URL https://abouttimeresearch.com