Transcriptomic analysis of differentially expressed genes during anther development in genetic male sterile and wild type cotton by digital gene-expression profiling

Abstract

Background: Cotton (Gossypium hirsutum) anther development involves a diverse range of gene interactions between sporophytic and gametophytic tissues. However, only a small number of genes are known to be specifically involved in this developmental process and the molecular mechanism of the genetic male sterility (GMS) is still poorly understand. To fully explore the global gene expression during cotton anther development and identify genes related to male sterility, a digital gene expression (DGE) analysis was adopted.

Results: Six DGE libraries were constructed from the cotton anthers of the wild type (WT) and GMS mutant (in the WT background) in three stages of anther development, resulting in 21,503 to 37,352 genes detected in WT and GMS mutant anthers. Compared with the fertile isogenic WT, 9,595 (30% of the expressed genes), 10,407 (25%), and 3,139 (10%) genes were differentially expressed at the meiosis, tetrad, and uninucleate microspore stages of GMS mutant anthers, respectively. Using both DGE experiments and real-time quantitative RT-PCR, the expression of many key genes required for anther development were suppressed in the meiosis stage and the uninucleate microspore stage in anthers of the mutant, but these genes were activated in the tetrad stage of anthers in the mutant. These genes were associated predominantly with hormone synthesis, sucrose and starch metabolism, the pentose phosphate pathway, glycolysis, flavonoid metabolism, and histone protein synthesis. In addition, several genes that participate in DNA methylation, cell wall loosening, programmed cell death, and reactive oxygen species generation/scavenging were activated during the three anther developmental stages in the mutant.

Conclusions: Compared to the same anther developmental stage of the WT, many key genes involved in various aspects of anther development show a reverse gene expression pattern in the GMS mutant, which indicates that diverse gene regulation pathways are involved in the GMS mutant anther development. These findings provide the first insights into the mechanism that leads to genetic male sterility in cotton and contributes to a better understanding of the regulatory network involved in anther development in cotton.

Figure 1

Flowers and pollen grains of the wild type (WT) and GMS mutant. Floral phenotypes of G. hirsutum ‘Dong A’ (WT: A) and GMS mutant of ‘Dong A’ (WT: B). Pollen grains stained with 2% I₂-KI at 0 DPA of the WT (C) and the GMS mutant anther (D). Scale bar = 100 μm.

Figure 2

TEM analysis of anthers in the WT and GMS mutant. Transverse sections of the (A-C) WT and (D-F) GMS mutant tapetum at the (A, D) meiosis, (B, E) tetrad, and (C, F) uninucleate microspore stages. Extine development in (G-I) WT and (J-L) GMS mutant microspores at the (G, J) meiosis, (H, K) tetrad, and (I, L) uninucleate microspore stages. Outer wall of anther epidermal cells in the (M-O) WT and (P-R) GMS mutant at the (M, P) meiosis, (N, Q) tetrad, and (O, R) uninucleate microspore stages. T: Tapetal layer; AT: abnormal tapetal layer; UB: Ubisch body; NUB: no Ubisch body; Ex: extine; AE: anther epidermal cuticle. Bars = 2 μm (A-L), 1 μm (M-R).

Figure 3

Transcriptome analysis of the WT and GMS mutant anthers. (a) Transcriptome sizes at three stages of the WT and GMS mutant anthers. (b), (c) Venn diagram showing the overlaps between anther stages of WT and GMS mutant (combined according to similarities in development). The number below each stage designation is the total transcripts detected in that stage(s). (d) Analysis of the progression of transcriptome changes during anther development of WT and GMS mutant anthers. The approximately 22,131 and 16,145 transcripts shared by three stages of WT and GMS mutant anthers are not shown, respectively. Numbers above the x-axis represent transcripts present in the indicated stage that are: stage specific (fufous); not present in the prior stage but shared with another stage (orange); or shared with the prior stage but missing in at least one other stage (yellow). Numbers below the x-axis represent transcripts present in the prior stage that are not detected in the current stage (from the category with green and blue color).

Figure 4

Differentially expressed genes across all libraries. All genes mapped to the reference sequence were examined for differences in their expression across the different libraries. (a), (b) and (c) Genes expression levels at meiosis, tetrad and uninucleate microspore stages of the WT and GMS mutant anthers, respectively. (d) The number of differentially expressed genes at three stages of WT and mutant anthers. F-1 and S-1: meiosis stage of WT and GMS mutant anthers; F-2 and S-2: tetrad stage of WT and GMS mutant anthers; F-3 and S-3: uninucleate microspore stage of WT and GMS mutant anthers.

Figure 5

Total soluble sugar content in WT and GMS mutant anthers. Data represent the mean and standard error from three replications. F-1 and S-1: meiosis stage of WT and GMS mutant anthers; F-2 and S-2: tetrad stage of WT and GMS mutant anthers; F-3 and S-3: uninucleate microspore stage of WT and GMS mutant anthers. FW: Fresh weight.

Figure 6

Quantitative RT-PCR validation of tag-mapped genes associated with cotton anther development. TPM, Transcription per million mapped reads. F-1 and S-1: meiosis stage of WT and GMS mutant anthers; F-2 and S-2: tetrad stage of WT and GMS mutant anthers; F-3 and S-3: uninucleate microspore stage of WT and GMS mutant anthers. Relative expression levels were calculated using 18S RNA as an internal control.