Genome-Wide Identification of Potential mRNAs in Drought Response in Wheat (Triticum aestivum L.)

Abstract

Plant cell metabolism inevitably forms an important drought-responsive mechanism, which halts crop productivity. Globally, more than 30% of the total harvested area was affected by dehydration. RNA-seq technology has enabled biologists to identify stress-responsive genes in relatively quick times. However, one shortcoming of this technology is the inconsistent data generation compared to other parts of the world. So, we have tried, here, to generate a consensus by analyzing meta-transcriptomic data available in the public microarray database GEO NCBI. In this way, the aim was set, here, to identify stress genes commonly identified as differentially expressed (p < 0.05) then followed by downstream analyses. The search term “Drought in wheat” resulted in 233 microarray experiments from the GEO NCBI database. After discarding empty datasets containing no expression data, the large-scale meta-transcriptome analytics and one sample proportional test were carried out (Bonferroni adjusted p < 0.05) to reveal a set of 11 drought-responsive genes on a global scale. The annotation of these genes revealed that the transcription factor activity of RNA polymerase II and sequence-specific DNA-binding mechanism had a significant role during the drought response in wheat. Similarly, the primary root differentiation zone annotations, controlled by TraesCS5A02G456300 and TraesCS7B02G243600 genes, were found as top-enriched terms (p < 0.05 and Q < 0.05). The resultant standard drought genes, glycosyltransferase; Arabidopsis thaliana KNOTTED-like; bHLH family protein; Probable helicase MAGATAMA 3; SBP family protein; Cytochrome c oxidase subunit 2; Trihelix family protein; Mic1 domain-containing protein; ERF family protein; HD-ZIP I protein; and ERF family protein, are important in terms of their worldwide proved link with stress. From a future perspective, this study could be important in a breeding program contributing to increased crop yield. Moreover, the wheat varieties could be identified as drought-resistant/sensitive based on the nature of gene expression levels.

Keywords: RNA seq; bread wheat; drought; genomics; meta data.

Figure 1

Flow chart diagram shown for screening of differentially expressed (DE) genes. Redundant genes were collected using expression datasets which were differentially expressed in each dataset.

Figure 2

Venn diagram elaborating shared number of genes across the combinations of microarray datasets. Sharing of DE genes between two datasets are shown in gold color, while sharp pink color represents three combination outcomes, in terms of sharing the number of DE genes.

Figure 3

Chromosomal map depicted the SDGs distribution over their respective chromosomes. Out of all wheat chromosomes, all SDG-bearing chromosomes are included in the map. Gene places on chromosomes were made set according to their genomic locations. The chromosomes were drawn as blue bars.

Figure 4

Heatmap and hierarchical clustering for gene expression matrices under well-watered and drought-stress conditions in 11 SDGs are shown as heatmaps. Color ranges are set between blue and red in order to visualize low to high gene expression, respectively. Clustering analysis of GSE47090, GSE70443, GSE45563 and GSE87325 showed two main horizontal groups of drought and controlled samples. Whereas, vertical scale is used to show all SDGs common in all gene expression datasets. Drought gene with GI accession 25550165 tends to be underexpressed in all experiments. Similarly, 313690563 gene in drought conditions had a trend towards overexpression in all gene expression datasets.

Figure 5

Principal component analysis biplot of gene expression of 11 genes studied in well-watered and drought conditions. PCA plot reveals the variation in the form of principal components (PC) 1 and 2. PC1 and PC2 drawn as Dim1 and Dim2 on horizontal and vertical axis, respectively, accounting for the variation up to 91.4% and 5.8%, respectively. Same genes are placed in plot based on their expression in drought and control conditions.

Figure 6

Drought proteins were annotated using reference genome of wheat from IWGSC. Predicted biological mechanisms indicated that auxin-related response, autophagy, electron transport chain and transcription regulations are prominent in this study.

Figure 7

Cnetplot created using plant ontology enriched terms (Adjusted p < 0.05). Out of all enriched terms epidermis, root hair cell, lead margin, fruit dehiscence zone and primary root differentiation zone terms are plotted by default which was associated with drought in wheat. The terms primary root differentiation zone, epidermis and root hair are governed by the DE genes are TraesCS5A02G456300 and TraesCS7B02G243600.