Characterization of bHLH/HLH genes that are involved in brassinosteroid (BR) signaling in fiber development of cotton (Gossypium hirsutum)

Abstract

Background: Basic helix-loop-helix/helix-loop-helix (bHLH/HLH) transcription factors play important roles in plant development. Many reports have suggested that bHLH/HLH proteins participate in brassinosteroid (BR) hormone signaling pathways to promote cell elongation. Cotton fibers are single-cells and derived from seed surface. To explore the roles of bHLH/HLH proteins in cotton fiber development progress by modulating BR signaling pathway, we performed a systematic analysis of the bHLH/HLH gene family in upland cotton (Gossypium hirsutum) genome.

Results: In this study, we identified 437 bHLH/HLH genes in upland cotton (G. hirsutum) genome. Phylogenetic analysis revealed that GhbHLH/HLH proteins were split into twenty six clades in the tree. These GhbHLH/HLH genes are distributed unevenly in different chromosomes of cotton genome. Segmental duplication is the predominant gene duplication event and the major contributor for amplification of GhbHLH/HLH gene family. The GhbHLH/HLHs within the same group have conserved exon/intron pattern and their encoding proteins show conserved motif composition. Based on transcriptome data, we identified 77 GhbHLH/HLH candidates that are expressed at relatively high levels in cotton fibers. As adding exogenous BR (brassinolide, BL) or brassinazole (Brz, a BR biosynthesis inhibitor), expressions of these GhbHLH/HLH genes were up-regulated or down-regulated in cotton fibers. Furthermore, overexpression of GhbHLH282 (one of the BR-response genes) in Arabidopsis not only promoted the plant growth, but also changed plant response to BR signaling.

Conclusion: Collectively, these data suggested that these GhbHLH/HLH genes may participate in BR signaling transduction during cotton fiber development. Thus, our results may provide a valuable reference data as the basis for further studying the roles of these bHLH/HLH genes in cotton fiber development.

Keywords: Brassinosteroid (BR) signaling; Cotton (Gossypium hirsutum); Fiber development; Gene expression; Phylogenetic analysis; bHLH/HLH transcription factor.

Fig. 1

Phylogenetic relationship of cotton bHLH/HLH transcription factors. MEGA 6.0 software was employed to construct an unrooted phylogenetic tree based on alignments of 437 bHLH/HLH domains identified from upland cotton (G. hirsutum) using the NeighborJoining (NJ) method with the following parameters: The number of differences model, pairwise deletion and 1000 bootstraps. Subfamilies are collapsed and represented as grey dots with area proportional to member numbers of each subfamily

Fig. 2

Chromosomal distribution and duplication of cotton (G. hirsutum) bHLH/HLH genes. The scale is in megabases (Mb). The chromosome number is indicated at the top of each chromosome. The paralogous bHLH genes are linked with a red line

Fig. 3

Characterization of cotton (G. hirsutum) bHLH/HLH genes and their encoding proteins. a The conserved distribution of exons and introns in GhbHLH/HLH genes. The yellow columns represent exons and black lines indicate introns. b Conserved motifs of GhbHLH/HLH proteins. The motifs 1–15 were identified using MEME program, and each motif is shown with a specific color. The composition elements of all the 15 motifs are shown in Additional file 1: Figure S2

Fig. 4

Quantitative RT-PCR analysis of expression of GhCPD and GhDWF4 genes in cotton fibers treated with BL and Brz. Cotton bolls (9 DPA) were in vitro cultured in liquid BT medium without or with 100 nM 2, 4-epibrassinolide (eBL) or brassinazole2001 (Brz) at 30 °C in darkness for 3 h. Transcript levels of GhCPD and GhDWF4 were determined by quantitative RT-PCR, using GhUBI1 (EU604080) as a quantification control, and expression levels of the genes in the untreated samples (0 nM BL or Brz) were set to 1. Data were processed with Microsoft Excel. Mean values and standard deviations are shown from three independent experiments. Two asterisks represent there was very significant difference (P < 0.01) in gene expression level between the treated sample and the untreated sample (control). DPA, day post anthesis

Fig. 5

Heatmap representation for expression profiles of cotton (G. hirsutum) bHLH/HLH genes in 9 DPA fibers under BL and Brz treatments. Transcript levels of GhbHLH/HLH genes were determined by quantitative RT-PCR, using GhUBI1 as a quantification control. Data were processed with Microsoft Excel. The values were normalized and the average of three biological replicates was used to generate log2 expression value. Then log2-transformed values were subjected to R sofware (15.2) for expression analysis. Heatmaps for gene expression patterns were generated by the online program Omicsharea (http://www.omicshare.com/tools/Home/Soft/heatmap). DPA, day post anthesis

Fig. 6

Heatmap representation for expression profiles of cotton (G. hirsutum) bHLH/HLH genes in cotton fibers. Transcript levels of the bHLH/HLH genes were determined by quantitative RT-PCR, using GhUBI1 as a quantification control. Data were processed with Microsoft Excel. The values were normalized and the average of three biological replicates was used to generate log2 expression value. Then log2-transformed values were subjected to R sofware (15.2) for expression analysis. Heatmaps for gene expression patterns were generated by the online program Omicsharea (http://www.omicshare.com/tools/Home/Soft/heatmap). 3DPA – 21DPA, 3 to 21 DPA (day post anthesis) fibers of cotton

Fig. 7

Phenotypic analysis of GhbHLH282-overexpressing transgenic Arabidopsis plants in responses to 2, 4-epibrassinolide (BL). a-c Subcellular localization of GhbHLH282 protein. a Fluorescence of eGFP:GhbHLH282 fusion proteins in the tobacco (Nicotiana Benthamiana) foliar epidermis cells. b Nuclear DAPI staining of the same cell in a. c The images of a and b were merged over the bright-field. d Semiquantitative RT-PCR analysis of GhbHLH282 expression in the transgenic plants and wild type controls. Transcript levels of GhbHLH282 were determined by Semiquantitative RT-PCR analysis, using AtACTIN2 as a quantification control. e The phenotypes of GhbHLH282-overexpressing seedlings and wild type grown on half-strength MS medium without BL for 7 days. f The phenotypes of GhbHLH282-overexpressing seedlings and wild type grown on half-strength MS medium with 1000 nM BL for 7 days. g Statistical analysis of petiole length of the GhbHLH282-overexpressing transgenic lines and wild type. The petiole growth of GhbHLH282-overexpressing plants showed the decreased sensitivity to BL, compared with mock control (0 nM BL), when treated with 10, 100 and 1000 nM BL. h Statistical analysis of root length of the GhbHLH282-overexpressing transgenic lines and wild type. The root growth of GhbHLH282-overexpressing plants showed the increased sensitivity to BL, compared with mock control (0 nM BL), when treated with 10, 100 and 1000 nM BL. Mean values and standard deviations are shown from three independent experiments. One or two asterisks represent there was significant (P < 0.05) or very significant (P < 0.01) difference in petiole length or root length between the BL-treated sample and the untreated sample (control), respectively. WT, wild type; L1, L4 and L6, three GhbHLH282 transgenic lines