Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2017 May;214(3):1213-1229.
doi: 10.1111/nph.14448. Epub 2017 Feb 10.

DNA Methylation and Gene Expression Regulation Associated With Vascularization in Sorghum Bicolor

Affiliations
Free PMC article

DNA Methylation and Gene Expression Regulation Associated With Vascularization in Sorghum Bicolor

Gina M Turco et al. New Phytol. .
Free PMC article

Abstract

Plant secondary cell walls constitute the majority of plant biomass. They are predominantly found in xylem cells, which are derived from vascular initials during vascularization. Little is known about these processes in grass species despite their emerging importance as biomass feedstocks. The targeted biofuel crop Sorghum bicolor has a sequenced and well-annotated genome, making it an ideal monocot model for addressing vascularization and biomass deposition. Here we generated tissue-specific transcriptome and DNA methylome data from sorghum shoots, roots and developing root vascular and nonvascular tissues. Many genes associated with vascular development in other species show enriched expression in developing vasculature. However, several transcription factor families varied in vascular expression in sorghum compared with Arabidopsis and maize. Furthermore, differential expression of genes associated with DNA methylation were identified between vascular and nonvascular tissues, implying that changes in DNA methylation are a feature of sorghum root vascularization, which we confirmed using tissue-specific DNA methylome data. Roots treated with a DNA methylation inhibitor also showed a significant decrease in root length. Tissues and organs can be discriminated based on their genomic methylation patterns and methylation context. Consequently, tissue-specific changes in DNA methylation are part of the normal developmental process.

Keywords: biofuel; cell type-specific; epigenetics; sorghum (Sorghum bicolor); transcriptome.

Figures

Figure 1
Figure 1
Transcriptional profiling of sorghum tissues. (a) Upper panel, 3‐d‐old sorghum root and shoot tissues and respective shoot cross0section. Bar, 100 µm. Lower panel, whole sorghum root and root cross section with 400 µm scale. (b) Laser capture microdissection (LCM) capturing the root vascular tissue from meristematic zone through the elongation zone. The left panel shows the whole root, the middle panel has the vascular section captured, and the right panel has both the vascular and nonvascular sections captured. (c) Selected gene ontology (GO) terms enriched for differentially expressed genes up‐regulated between root and shoot, and between root vascular and nonvascular tissues. A darker color blue indicates a more significant P‐value. All GO terms have a P‐value < 0.005 in at least one of the tissue types.
Figure 2
Figure 2
Conservation and divergence of vascular‐expressed genes in monocots and dicots. (a) Number of vascular‐enriched genes shared and unique to sorghum, maize and Arabidopsis. (b) Selected gene ontology (GO) term enrichment for vascular‐expressed genes between species; darker blue corresponds to a lower P‐value. (c) Expansion and conservation of annotated transcription factor families expressed in vascular tissue. Enrichment of a family in vascular tissue is indicated by a darker color red. A Fisher exact test was used to test for vascular enrichment among species where the number of vascular‐expressed transcription factors for each family over the total number of transcription factors for the family was compared with the number of vascular‐expressed genes in that species over the total number of genes in the genome. (d) Auxin response factor (ARF) transcription factor family phylogeny across sorghum, maize and Arabidopsis. Filled symbols represent vascular‐enriched genes within the ARF family. Sorghum and maize have an expansion of ARFs in comparison to Arabidopsis.
Figure 3
Figure 3
Identification of genes associated with vascular development in sorghum through coexpression analysis. Top 50 genes coexpressed with Bmr2,VND7 and CESA4 (a, b and c, respectively). Coexpression was determined by correlating the slope of the line for each gene's reads per kilobase per million (RPKM) expression value across tissues against the reference gene's slope. Genes with a Pearson correlation ≥ 0.9 were considered coexpressed. Sorghum genes with known functions in secondary cell wall biosynthesis have been colored according to function. TF, transcription factor.
Figure 4
Figure 4
Sorghum methylome and genome architecture. Sodium bisulfite sequencing data across three biological replicates were merged and used in calling methylated regions for each tissue type. (a) Sorghum epigenome density plot for CG, CHG and CHH methylation contexts in vascular tissue. A darker color blue corresponds to a higher percentage methylation within a 10 000 bp window. Repetitive elements (RE) and gene density are reported on the inner tracks, where a darker color red corresponds to a higher density. (b) DNA methylation patterns averaged across genic regions for each methylation context in vascular tissue. Vascular methylation patterns shown here are representative of all tissue types. (c) Weighted means of genome‐wide average methylation in each methylation context and each sample. (d) Sorghum seedlings grown on DNA methylation inhibitor 5‐azacytidine. Ten randomly selected plants 4 d postgermination (DPG) grown on either 100 μM 5‐azacytidine dissolved in dimethyl sulfoxide (DMSO) or on DMSO (untreated). (e) Quantification of root lengths of sorghum seedlings grown on either 100 μM 5‐azacytidine (AZA) or DMSO. (e) **, < 0.005 for replicate × treatment interaction term determined using a two‐way ANOVA fit to root length ~ treatment × replicate. The error bars represent ± SE between samples (= 30).
Figure 5
Figure 5
Distribution of methylation across genomic features. (a) Principal component (PC) analysis on methylation state (methylated or unmethylated) for all potential cytosines of each methylation context shared between the four tissue types (root, shoot, vascular and nonvascular) for three biological replicates. (b) Methylation density in each context, CG, CHG and CHH, averaged in all genes with mapping coverage of ≥ 4× in root, shoot, vascular and nonvascular samples. The bottom panel denotes significant differences in average methylation of the genomic region (upstream, gene body, downstream) between samples (root, shoot, vascular, nonvascular), where the darkest blue represents P‐values ≤ 0.0001, the second darkest are P‐values ranging from 0.0001 to 0.001, and the lightest blue are P‐values in the range 0.001–0.01 determined by a Tukey test.
Figure 6
Figure 6
Methylation in genic regions varies with tissue type and tissue‐specific expression level. Average methylation levels of genic regions for each expression level group from each sample. Genes were categorized into four groups based on their quartile of expression level in each sample, where group 1 is the quartile containing genes of lowest expression level and group 4 is the quartile containing genes with the highest expression level. The methylation level of each group in each sample is shown for each context; CG (a), CHG (b) and CHH (c). The bottom panel denotes significant differences in average methylation of the genomic region (upstream, gene body, downstream) between samples (root, shoot, vascular, nonvascular), where the darkest blue represents P‐values ≤ 0.0001, the second darkest are P‐values ranging from 0.0001 to 0.001, and the lightest blue are P‐values in the range 0.001–0.01 determined by a Tukey test.
Figure 7
Figure 7
Differentially methylated regions between vascular and nonvascular tissues. Selected differentially methylated regions (DMRs) that overlap with differentially expressed genes. All three methylation contexts were merged to identify DMRs. The first few panels colored cytosines as either red (methylated) or blue (unmethylated) depending on their methylation status. Regions boxed in red represent DMRs found between vascular and nonvascular samples. The following panels show the number of RNA‐Seq reads (in gray) mapping to each gene in vascular and nonvascular tissues. The gene model and the direction of the gene are indicated by the turquoise track and white arrowheads. Differences in read count between vascular and nonvascular tissues are differentially expressed following edgeR analysis. Gene annotations are shown below in green, and exons are represented by larger squares. Both the expansin precursor Sb06g026480 and gibberellin‐regulated gene Sb10g009640 are differentially expressed.

Similar articles

See all similar articles

Cited by 3 articles

References

    1. Bao N, Lye KW, Barton MK. 2004. MicroRNA binding sites in Arabidopsis class III HD‐ZIP mRNAs are required for methylation of the template chromosome. Developmental Cell 7: 653–662. - PubMed
    1. Belmonte MF, Kirkbride RC, Stone SL, Pelletier JM, Bui AQ, Yeung EC, Hashimoto M, Fei J, Harada CM, Munoz MD et al 2013. Comprehensive developmental profiles of gene activity in regions and subregions of the Arabidopsis seed. Proceedings of the National Academy of Sciences, USA 110: E435–E444. - PMC - PubMed
    1. Bonawitz ND, Kim JI, Tobimatsu Y, Ciesielski PN, Anderson NA, Ximenes E, Maeda J, Ralph J, Donohoe BS, Ladisch M et al 2014. Disruption of Mediator rescues the stunted growth of a lignin‐deficient Arabidopsis mutant. Nature 509: 376–380. - PubMed
    1. Brady SM, Orlando DA, Lee J‐Y, Wang JY, Koch J, Dinneny JR, Mace D, Ohler U, Benfey PN. 2007a. A high‐resolution root spatiotemporal map reveals dominant expression patterns. Science 318: 801–806. - PubMed
    1. Brady SM, Song S, Dhugga KS, Rafalski JA, Benfey PN. 2007b. Combining expression and comparative evolutionary analysis. The COBRA gene family. Plant Physiology 143: 172–187. - PMC - PubMed

Associated data

Feedback