Comparative transcriptome analysis reveals potential regulatory genes involved in the development and strength formation of maize stalks

Abstract

Background: Stalk strength is a critical trait in maize that influences plant architecture, lodging resistance and grain yield. The developmental stage of maize, spanning from the vegetative stage to the reproductive stage, is critical for determining stalk strength. However, the dynamics of the genetic control of this trait remains unclear.

Results: Here, we report a temporal resolution study of the maize stalk transcriptome in one tropical line and one non-stiff-stalk line using 53 transcriptomes collected covering V7 (seventh leaf stage) through silking stage. The time-course transcriptomes were categorized into four phases corresponding to stalk early development, stalk early elongation, stalk late elongation, and stalk maturation. Fuzzy c-means clustering and Gene Ontology (GO) analyses elucidated the chronological sequence of events that occur at four phases of stalk development. Gene Ontology analysis suggests that active cell division occurs in the stalk during Phase I. During Phase II, processes such as cell wall extension, lignin deposition, and vascular cell development are active. In Phase III, lignin metabolic process, secondary cell wall biogenesis, xylan biosynthesis process, cell wall biogenesis, and polysaccharide biosynthetic process contribute to cell wall strengthening. Defense responses, abiotic stresses, and transport of necessary nutrients or substances are active engaged during Phase IV. Kyoto Encyclopedia of Genes and Genomes (KEGG) analysis showed that the two maize lines presented significant gene expression differences in the phenylpropanoid biosynthesis pathway and the flavonoid biosynthesis pathway. Certain differentially expressed genes (DEGs) encoding transcription factors, especially those in the NAC and MYB families, may be involved in stalk development. In addition, six potential regulatory genes associated with stalk strength were identified through weighted gene co-expression network analysis (WGCNA).

Conclusion: The data set provides a high temporal-resolution atlas of gene expression during maize stalk development. These phase-specific genes, differentially expressed genes, and potential regulatory genes reported in this study provide important resources for further studies to elucidate the genetic control of stalk development and stalk strength formation in maize.

Keywords: Maize; Regulatory genes; Stalk development; Stalk strength; Transcriptome.

Fig. 1

Stalk strength phenotypes of maize inbred lines CML323 and W22. (A) Plants of CML323 (left) and W22 (right) at R1 stage; bar = 10 cm. (B) Rind penetrometer resistance evolution in CML323 (n = 3 at each point) and W22 (n = 3 at each point); mean ± SD. VT, tasseling stage; R1, silking stage; R2, blister stage; RPR, rind penetrometer resistance. two-tailed Student’s t-test was used to determine P values. **P < 0.01; ***P < 0.001; NS, not significant

Fig. 2

Hierarchical clustering of dynamically expressed genes spanning V7 to silking (R1) and representative genes that showed temporal expression pattern at the four development phases. (A) Cluster dendrograms of the transcriptomes of CML323 (left panel) and W22 (right panel). C, CML323; W, W22. These selected genes were mainly expressed in phases I (B), II (C), III (D), and IV (E) correspondingly

Fig. 3

Clustering of dynamically expressed genes. (A) Gene expression clusters in CML323. (B) Gene expression clusters in W22. Fuzzy c-means clustering shows the dynamic expression profile of the 9 clusters throughout the studied development stages in each line. C, CML323; W, W22

Fig. 4

Differential expression analysed between CML323 and W22. (A) Number of DEGs between CML323 and W22 in these representative stages. Down and up, down-regulated and up-regulated in CML323 compared with W22. (B) The Venn diagram shows the overlap of DEGs of these 3 stages. (C) KEGG pathway enrichment analysis of DEGs

Fig. 5

Differentially expressed TFs between CML323 and W22. (A) Overview of transcription factors (TFs, ) in the V8, V10, V14 stage. Down and up, purple indicate that the gene is upregulated in CML323, while blue arrows indicate that it is downregulated. (B) Top Differentially expressed in NAC and MYB transcription factors. Red arrows indicate that the gene is upregulated in CML323, while blue arrows indicate that it is downregulated

Fig. 6

Co-expression network analysis of the dynamically expressed genes and identification of critical genes predicting stalk strength formation. (A) Module-phenotype association. Each row represents a colored module, with the number of genes displayed within the colored box of each module. The correlation coefficient between each module and RPR is indicated in red for positive correlations (ranging from 0 to 1) and in green for negative correlations (also ranging from 0 to 1). (B) The Venn diagram shows the overlap of DEGs and hub genes. (C) Interactions between hub genes within the blue module. Six hub genes are shown as orange circles. MYB42 is shown as violet circle. RPR, rind penetrometer resistance

Fig. 7

Validation of differentially expressed genes (DEGs) by qRT-PCR. Four DEGs were up-regulated in CML323 compared with W22 at V10 and V14 stages. The relative expression level of each gene was expressed as the fold change in RNA-seq data and qRT-PCR data, setting values of W22 as 1. Error bars represent the standard deviation (n = 3)