Microspore embryogenesis induction by mannitol and TSA results in a complex regulation of epigenetic dynamics and gene expression in bread wheat

Abstract

Reprogramming of microspores development towards embryogenesis mediated by stress treatment constitutes the basis of doubled haploid production. Recently, compounds that alter histone post-translational modifications (PTMs) have been reported to enhance microspore embryogenesis (ME), by altering histones acetylation or methylation. However, epigenetic mechanisms underlying ME induction efficiency are poorly understood. In this study, the epigenetic dynamics and the expression of genes associated with histone PTMs and ME induction were studied in two bread wheat cultivars with different ME response. Microspores isolated at 0, 3 and 5 days, treated with 0.7M mannitol (MAN) and 0.7M mannitol plus 0.4µM trichostatin A (TSA), which induced ME more efficiently, were analyzed. An additional control of gametophytic development was included. Microspores epigenetic state at the onset of ME induction was distinctive between cultivars by the ratio of H3 variants and their acetylated forms, the localization and percentage of labeled microspores with H3K9ac, H4K5ac, H4K16ac, H3K9me2 and H3K27me3, and the expression of genes related to pollen development. These results indicated that microspores of the high responding cultivar could be at a less advanced stage in pollen development. MAN and TSA resulted in a hyperacetylation of H3.2, with a greater effect of TSA. Histone PTMs were differentially affected by both treatments, with acetylation being most concerned. The effect of TSA was observed in the H4K5ac localization pattern at 3dT in the mid-low responding cultivar. Three gene networks linked to ME response were identified. TaHDT1, TaHAG2, TaYAO, TaNFD6-A, TabZIPF1 and TaAGO802-B, associated with pollen development, were down-regulated. TaHDA15, TaHAG3, TaHAM, TaYUC11D, Ta-2B-LBD16 TaMS1 and TaDRM3 constituted a network implicated in morphological changes by auxin signaling and cell wall modification up-regulated at 3dT. The last network included TaHDA18, TaHAC1, TaHAC4, TaABI5, TaATG18fD, TaSDG1a-7A and was related to ABA and ethylene hormone signaling pathways, DNA methylation and autophagy processes, reaching the highest expression at 5dT. The results indicated that TSA mainly modified the regulation of genes related to pollen and auxin signaling. This study represents a breakthrough in identifying the epigenetic dynamics and the molecular mechanisms governing ME induction efficiency, with relevance to recalcitrant wheat genotypes and other crops.

Keywords: bread wheat; epigenetics; gene expression; histones; mannitol; microspore embryogenesis; trichostatin A.

Figure 1

Effect of MAN and TSA treatments on histones H3 and H4 acetylation. (A, B) Immunoblot of histones H3 and H4 and pan-acetylated H3 and H4 (H3ac and H4ac, respectively). (C, D) Quantitative analysis of the H3.1/H3.2 and H3.1ac/H3.2ac ratios, respectively. (E, F) Quantitative analysis of the H3.1 and H3.2 acetylation ratio (G) Quantitative analysis of the H4 acetylation ratio. Microspores before stress treatment (0dT), after 3 and 5 days with mannitol treatment (3dT-MAN, 5dT-MAN), and after 3 and 5 days with mannitol plus TSA treatment (3dT-TSA, 5dT-TSA) from Pavon (on the left) and Caramba (on the right). Bars represent the standard errors of the means. Histone H3.1/H3.2, H3.1ac/H3.2ac, H3.1ac/H3.1, H3.2ac/H3.2 and H4ac/H4 ratios with the same letter within each ratio are not significantly different (P˂0.05) according to the Duncan test.

Figure 2

Effect of MAN and TSA on histone PTMs. (A, B) Percentage of labeled microspores and nuclei, respectively, with the acetylation marks H3K9ac (dark blue), H4K5ac (light blue), H4K16ac (green). (C, D) Percentage of labeled microspores and nuclei, respectively, with the methylation marks H3K9me2 (red), H3K27me3 (pink). Microspores before stress treatment (0dT), microspores under gametophyte development (GD), microspores after 3 and 5 days with mannitol treatment (3dT-MAN, 5dT-MAN) and after 3 and 5 days with mannitol plus TSA treatment (3dT-TSA, 5dT-TSA) from Pavon (on the left) and Caramba (on the right). Percentages of labeled microspore and nuclei with the same letter within each epigenetic mark are not significantly different (P˂0.05) according to the Duncan test.

Figure 3

Immunolocalization of histone PTMs in microspores before stress treatment (0dT), microspores under gametophytic development (GD), microspores after 3 and 5 days with mannitol treatment (3dT-MAN, 5dT-MAN) and after 3 and 5 days with mannitol plus TSA treatment (3dT-TSA, 5dT-TSA) from Pavon (A) and Caramba (B). H3K9ac, H4K5ac, H4K16ac, H3K9me2 and H3K27me3 antibodies labeling in green, DAPI in blue. Scale bar = 20 µm.

Figure 4

qRT-PCR analysis of histone deacetylases genes (TaHDA15, TaHDA18, TaHDT1, TaHDT2, TaTAF14) and histone acetyltransferases genes (TaHAG2, TaHAG3, TaHAM, TaHAC1, TaHAC4) in microspore before stress treatment (0dT), microspores under gametophytic development (GD), microspores after 3 and 5 days with mannitol treatment (3dT-MAN, 5dT-MAN) and after 3 and 5 days with mannitol plus TSA treatment (3dT-TSA, 5dT-TSA) from Pavon (on the left) and Caramba (on the right). Data were normalized using Ta.27771 and Ta.2291 as endogenous control genes. Bars represent the standard errors of the means. Relative expression values with the same letter are not significantly different (P<0.05) according to the Duncan test.

Figure 5

qRT-PCR analysis of key genes involved in microspore embryogenesis induction. Microspores before stress treatment (0dT), microspores under gametophytic development (GD), microspores after 3 and 5 days with mannitol treatment (3dT-MAN, 5dT-MAN) and after 3 and 5 days with mannitol plus TSA treatment (3dT-TSA, 5dT-TSA) from Pavon (on the left) and Caramba (on the right). Data were normalized using Ta.27771 and Ta.2291 as endogenous genes control. Bars represent the standard errors of the means. Relative expression values with the same letter are not significantly different (P˂0.05) according to the Duncan test.

Figure 6

(A) Pearson correlation coefficients of histone deacetylases, histone acetyltransferases and key genes involved in ME induction, considering all samples of the two cultivars. (B) Principal component analysis (PCA) plot of the 20 genes in Pavon and Caramba. Microspores before stress treatment (0dT), microspores under gametophytic development (GD), microspores after3 and 5 days with mannitol treatment (3dT-MAN, 5dT-MAN) and after 3 and 5 days with mannitol plus TSA treatment (3dT-TSA, 5dT-TSA).