Expansion of the circadian transcriptome in Brassica rapa and genome-wide diversification of paralog expression patterns.

An important challenge of crop improvement strategies is assigning function to paralogs in polyploid crops. Here we describe the circadian transcriptome in the polyploid crop Brassica rapa. Strikingly, almost three-quarters of the expressed genes exhibited circadian rhythmicity. Genetic redundancy resulting from whole genome duplication is thought to facilitate evolutionary change through sub- and neo-functionalization among paralogous gene pairs. We observed genome-wide expansion of the circadian expression phase among retained paralogous pairs. Using gene regulatory network models, we compared transcription factor targets between B. rapa and Arabidopsis circadian networks to reveal evidence for divergence between B. rapa paralogs that may be driven in part by variation in conserved non-coding sequences (CNS). Additionally, differential drought response among retained paralogous pairs suggests further functional diversification. These findings support the rapid expansion and divergence of the transcriptional network in a polyploid crop and offer a new approach for assessing paralog activity at the transcript level.

Like animals, plants have internal biological clocks that allow them to adapt to daily and yearly changes, such as day-night cycles or seasons turning. Unlike animals, however, plants cannot move when their environment becomes different, so they need to be able to weather these changes by adjusting which genes they switch on and off. To do this, plants keep track of how long days are using external cues such as light or temperature. One of the effects of climate change is that these cues become less reliable, making it harder for plants to adapt to their environment and survive. This is a potential problem for crop species, like Brassica rapa. This plant has many edible forms, including Chinese cabbage, oilseed, pak choi, and turnip. It is also a close relative of the well-studied model plant, Arabidopsis. Since evolving away from Arabidopsis, the genome of B. rapa tripled, meaning it has one, two, or three copies of each gene. This has allowed the extra gene copies to mutate and adapt to different purposes. The question is, what impact has this genome expansion had on the plant's biological clock? One way to find out is to perform RNA-sequencing experiments, which record the genes a plant is using at any one time. Here, Greenham, Sartor et al. report the results of a series of RNA-sequencing experiments performed every two hours across two days. Plants were first exposed to light-dark or temperature cycles and then samples were taken when the plants were in constant light and temperature. This revealed which genes B. rapa turned on and off in response to signals from the internal biological clock. It turns out that the biological clock of B. rapa controls close to three quarters of its genes. These genes showed distinct phases, increasing or decreasing in regular patterns. But the different copies of duplicated and triplicated genes did not necessarily all behave in the same way. Many of the copies had different rhythms, and some increased and decreased in patterns totally opposite to their counterparts. Not only did the daily patterns differ, but responses to stressors like drought were also altered. Comparing these patterns to the patterns seen in Arabidopsis revealed that often, one B. rapa gene behaved just like its Arabidopsis equivalent, while its copies had evolved new behaviors. The different behaviors of the copies of each gene in B. rapa relative to its biological clock allow this plant to grow in different environments with varying temperatures and day lengths. Understanding how these adaptations work opens new avenues of research into how plants detect and respond to environmental signals. This could help to guide future work into targeting genes to improve crop growth and stress resilience.

Identification of the expressome by machine learning on omics data.

Accurate annotation of plant genomes remains complex due to the presence of many pseudogenes arising from whole-genome duplication-generated redundancy or the capture and movement of gene fragments by transposable elements. Machine learning on genome-wide epigenetic marks, informed by transcriptomic and proteomic training data, could be used to improve annotations through classification of all putative protein-coding genes as either constitutively silent or able to be expressed. Expressed genes were subclassified as able to express both mRNAs and proteins or only RNAs, and CG gene body methylation was associated only with the former subclass. More than 60,000 protein-coding genes have been annotated in the reference genome of maize inbred B73. About two-thirds of these genes are transcribed and are designated the filtered gene set (FGS). Classification of genes by our trained random forest algorithm was accurate and relied only on histone modifications or DNA methylation patterns within the gene body; promoter methylation was unimportant. Other inbred lines are known to transcribe significantly different sets of genes, indicating that the FGS is specific to B73. We accurately classified the sets of transcribed genes in additional inbred lines, arising from inbred-specific DNA methylation patterns. This approach highlights the potential of using chromatin information to improve annotations of functional genes.

Analysis of differential gene expression and alternative splicing is significantly influenced by choice of reference genome.

RNA-seq analysis has enabled the evaluation of transcriptional changes in many species including nonmodel organisms. However, in most species only a single reference genome is available and RNA-seq reads from highly divergent varieties are typically aligned to this reference. Here, we quantify the impacts of the choice of mapping genome in rice where three high-quality reference genomes are available. We aligned RNA-seq data from a popular productive rice variety to three different reference genomes and found that the identification of differentially expressed genes differed depending on which reference genome was used for mapping. Furthermore, the ability to detect differentially used transcript isoforms was profoundly affected by the choice of reference genome: Only 30% of the differentially used splicing features were detected when reads were mapped to the more commonly used, but more distantly related reference genome. This demonstrated that gene expression and splicing analysis varies considerably depending on the mapping reference genome, and that analysis of individuals that are distantly related to an available reference genome may be improved by acquisition of new genomic reference material. We observed that these differences in transcriptome analysis are, in part, due to the presence of single nucleotide polymorphisms between the sequenced individual and each respective reference genome, as well as annotation differences between the reference genomes that exist even between syntenic orthologs. We conclude that even between two closely related genomes of similar quality, using the reference genome that is most closely related to the species being sampled significantly improves transcriptome analysis.

Improving Gene Regulatory Network Inference by Incorporating Rates of Transcriptional Changes.

Organisms respond to changes in their environment through transcriptional regulatory networks (TRNs). The regulatory hierarchy of these networks can be inferred from expression data. Computational approaches to identify TRNs can be applied in any species where quality RNA can be acquired, However, ChIP-Seq and similar validation methods are challenging to employ in non-model species. Improving the accuracy of computational inference methods can significantly reduce the cost and time of subsequent validation experiments. We have developed ExRANGES, an approach that improves the ability to computationally infer TRN from time series expression data. ExRANGES utilizes both the rate of change in expression and the absolute expression level to identify TRN connections. We evaluated ExRANGES in five data sets from different model systems. ExRANGES improved the identification of experimentally validated transcription factor targets for all species tested, even in unevenly spaced and sparse data sets. This improved ability to predict known regulator-target relationships enhances the utility of network inference approaches in non-model species where experimental validation is challenging. We integrated ExRANGES with two different network construction approaches and it has been implemented as an R package available here: http://github.com/DohertyLab/ExRANGES . To install the package type: devtools::install_github("DohertyLab/ExRANGES").

Integration of omic networks in a developmental atlas of maize.

Coexpression networks and gene regulatory networks (GRNs) are emerging as important tools for predicting functional roles of individual genes at a system-wide scale. To enable network reconstructions, we built a large-scale gene expression atlas composed of 62,547 messenger RNAs (mRNAs), 17,862 nonmodified proteins, and 6227 phosphoproteins harboring 31,595 phosphorylation sites quantified across maize development. Networks in which nodes are genes connected on the basis of highly correlated expression patterns of mRNAs were very different from networks that were based on coexpression of proteins. Roughly 85% of highly interconnected hubs were not conserved in expression between RNA and protein networks. However, networks from either data type were enriched in similar ontological categories and were effective in predicting known regulatory relationships. Integration of mRNA, protein, and phosphoprotein data sets greatly improved the predictive power of GRNs.

Maize death acids, 9-lipoxygenase-derived cyclopente(a)nones, display activity as cytotoxic phytoalexins and transcriptional mediators.

Plant damage promotes the interaction of lipoxygenases (LOXs) with fatty acids yielding 9-hydroperoxides, 13-hydroperoxides, and complex arrays of oxylipins. The action of 13-LOX on linolenic acid enables production of 12-oxo-phytodienoic acid (12-OPDA) and its downstream products, termed "jasmonates." As signals, jasmonates have related yet distinct roles in the regulation of plant resistance against insect and pathogen attack. A similar pathway involving 9-LOX activity on linolenic and linoleic acid leads to the 12-OPDA positional isomer, 10-oxo-11-phytodienoic acid (10-OPDA) and 10-oxo-11-phytoenoic acid (10-OPEA), respectively; however, physiological roles for 9-LOX cyclopentenones have remained unclear. In developing maize (Zea mays) leaves, southern leaf blight (Cochliobolus heterostrophus) infection results in dying necrotic tissue and the localized accumulation of 10-OPEA, 10-OPDA, and a series of related 14- and 12-carbon metabolites, collectively termed "death acids." 10-OPEA accumulation becomes wound inducible within fungal-infected tissues and at physiologically relevant concentrations acts as a phytoalexin by suppressing the growth of fungi and herbivores including Aspergillus flavus, Fusarium verticillioides, and Helicoverpa zea. Unlike previously established maize phytoalexins, 10-OPEA and 10-OPDA display significant phytotoxicity. Both 12-OPDA and 10-OPEA promote the transcription of defense genes encoding glutathione S transferases, cytochrome P450s, and pathogenesis-related proteins. In contrast, 10-OPEA only weakly promotes the accumulation of multiple protease inhibitor transcripts. Consistent with a role in dying tissue, 10-OPEA application promotes cysteine protease activation and cell death, which is inhibited by overexpression of the cysteine protease inhibitor maize cystatin-9. Unlike jasmonates, functions for 10-OPEA and associated death acids are consistent with specialized roles in local defense reactions.