Conservation and divergence of transcriptomic and epigenomic variation in maize hybrids
Recent genome-wide studies suggested that in addition to genetic variations, epigenetic variations may also be associated with differential gene expression and growth vigor in plant hybrids. Maize is an ideal model system for the study of epigenetic variations in hybrids given the significant heterotic performance, the well-known complexity of the genome, and the rich history in epigenetic studies. However, integrated comparative transcriptomic and epigenomic analyses in different organs of maize hybrids remain largely unexplored.
Here, we generated integrated maps of transcriptomes and epigenomes of shoots and roots of two maize inbred lines and their reciprocal hybrids, and globally surveyed the epigenetic variations and their relationships with transcriptional divergence between different organs and genotypes. We observed that whereas histone modifications vary both between organs and between genotypes, DNA methylation patterns are more distinguishable between genotypes than between organs. Histone modifications were associated with transcriptomic divergence between organs and between hybrids and parents. Further, we show that genes up-regulated in both shoots and roots of hybrids were significantly enriched in the nucleosome assembly pathway. Interestingly, 22- and 24-nt siRNAs were shown to be derived from distinct transposable elements, and for different transposable elements in both shoots and roots, the differences in siRNA activity between hybrids and patents were primarily driven by different siRNA species.
These results suggest that despite variations in specific genes or genomic loci, similar mechanisms may account for the genome-wide epigenetic regulation of gene activity and transposon stability in different organs of maize hybrids.
Keywordsmaize hybrids transcriptome epigenome
Chromatin immunoprecipitation sequencing
Expressed sequence tag
First filial generation
Histone H3 tri-methylated at lysine 36
Histone H3 tri-methylated at lysine 4
Histone H3 acetylated at lysine 9
Model-based analysis of ChIP-seq
- McrBC-seq Mcr
Polymerase chain reaction
RNA-directed DNA methylation
Aligned reads per kilobase exon model (or genic region: or genomic region) per million mapped reads
Small interfering RNA
Transcription start site
Transcription termination site.
The chromatin states and genome activity in eukaryotes are regulated by a variety of different epigenetic mechanisms, mainly DNA methylation, histone modifications, and the RNA interference pathway [1, 2, 3]. DNA methylation, that is, the addition of a methyl group to a cytosine by DNA methyltransferases, is primarily regarded as a relatively stable repressive epigenetic marker, which maintains genome stability by suppressing the activity of transposons and other repetitive sequences [4, 5]. Recent studies have also indicated an additional potential role for DNA methylation in regulating the expression of protein-coding genes [6, 7, 8]. In some cases, the methylated cytosines can be removed by DNA glycosylase . Histone modifications, which are post-translational modifications of histone proteins at their N-terminal tails, provide a dynamic and reversible mechanism to regulate gene expression in response to diverse endogenous and exogenous stimuli [9, 10]. Lysine acetylation (for example, H3K9ac) and some lysine methylations (for example, H3K4me3 and H3K36me3) are associated with transcriptional activation of genes [8, 9, 10, 11]. Small RNAs, especially small interfering RNAs (siRNAs), provide another layer of epigenetic regulatory mechanism, and repeat-associated siRNAs are known to be involved in the maintenance of genome stability by RNA-directed DNA methylation (RdDM) [1, 12, 13].
Recently, genome-wide studies of these epigenetic components using high-throughput approaches have identified complex networks of their variations during plant evolution and development. For example, some studies characterized the patterns of DNA methylation, histone modifications, and small RNAs (sRNAs) in various tissues or organs of plants [7, 14, 15], whereas other studies compared the DNA-methylation patterns between different plant species or different genotypes [16, 17, 18, 19, 20]. Briefly, these studies revealed the conservation and divergence of epigenetic components in different plant organs or genotypes. Moreover, some studies also discovered the genome-wide epigenetic variations and their potential relationship with altered chromatin states and changed gene activity in plant hybrids. Fox example, the global DNA-methylation variations and their potential association with altered gene expression in hybrids have been extensively discussed [8, 21, 22, 23, 24, 25]. Further, sRNAs were also found to show extensive variation in hybrids of Arabidopsis [21, 24, 25, 26], rice [8, 23, 27] maize , wheat , and yellow poplar , and altered siRNA levels were presumably associated with changed DNA methylation in hybrids through the RdDM pathway [21, 23, 24, 25].
Maize is an appropriate model organism for studying global genetic and epigenetic variation in plants because of its exceptionally complex genome (in particular the high content of transposable elements (TEs)) and is a particularly rich source of epigenetic discoveries . Recent studies have explored the global transcriptional variations in maize hybrids in various tissues or organs, including embryos [32, 33], endosperms , immature ears , leaves , stem meristem , seedling shoots [37, 38], and roots [39, 40]. However, the mechanisms underlying these variations and their conservation and divergence between organs remain elusive. In addition, global epigenetic profiling in maize showed that siRNA populations vary following hybridization, and are associated with regulated transposons in the seedling shoot apex and developing ear of maize hybrids , and that DNA methylation is associated with allelic expression of imprinted genes in the endosperm of maize hybrids . Therefore, it is necessary to comprehensively determine both the genome-wide epigenetic variations and their association with transcriptional divergence between different organs of maize hybrids.
In this study, we generated highly integrated maps of transcriptomes and epigenomes in shoots and roots of two maize inbred lines and their reciprocal hybrids, and obtained a comprehensive view of the variation in gene expression, DNA methylation, histone modifications and sRNAs between organs and genotypes. In general, histone modifications are associated with differential gene expression between organs and between hybrids and parents, whereas, siRNAs and DNA methylation are mainly associated with regulated TEs and other repetitive elements, and thus may change the chromatin states in hybrids. Despite the variation in specific genes or genomic loci, similar global trends of transcriptomes and epigenomes were seen in both shoots and roots of reciprocal hybrids. Our data therefore may serve as a useful resource to better understand the epigenetic basis of gene action in different organs and different genetic backgrounds.
Transcriptomic and epigenomic profiling in shoots and roots of two maize inbred lines and their reciprocal hybrids
We used Illumina high-throughput sequencing approaches to generate integrated maps of mRNA and sRNA transcriptomes, DNA methylomes and genome-wide distribution of three representative histone modifications (H3K4me3, H3K9ac, and H3K36me3) in two maize inbred lines (B73 and Mo17) and their reciprocal hybrids (B73 ´ Mo17 and Mo17 ´ B73). Shoots and roots of both hybrids and parental lines from 14-day-old seedlings were used for all experiments in this study. Illumina sequencing libraries for mRNA sequencing (mRNA-seq), chromatin immunoprecipitation sequencing (ChIP-seq), McrBC sequencing (McrBC-seq), and sRNA sequencing (sRNA-seq) were constructed as previously described [3, 7, 8]. All sequencing reads were aligned to the reference genome of the maize inbred line B73 (ZmB73_RefGen_v2)  using Bowtie software  (see Additional file 1, Table S1).
Next, we analyzed the number and percentage of genes associated with expression or epigenetic modifications in shoots and roots. The levels of different epigenetic modifications on each gene were evaluated by directly counting the numbers of reads located in the genic region, and a threshold of read coverage defined by randomization (P<0.01) was used to identify genes with each modification. Generally, we obtained a similar number and percentage for genes with particular types of epigenetic modifications in shoots and roots (Figure 2c). Of 39,423 annotated genes in the maize genome, 22,696 to 28,786 (58.3% to 73.0%) and 24,248 to 25,532 (61.5% to 64.8%) genes in shoots and roots, respectively, contained histone modifications (H3K4me3, H3K9ac, and H3K36me3). However, only 1,243 (3.1%) and 1,276 (3.2%) genes contained DNA methylation in shoots and roots, respectively (Figure 2c). Together, we generated integrated maps of epigenomes and transcriptomes in shoots and roots of two maize inbred lines and their reciprocal hybrids (see Additional file 2, Figure S4 for a representative region on chromosome 1 showing the integrated maps).
Patterns of variation in gene expression and epigenetic modifications between organs and between genotypes
To survey the global trends of transcriptional and epigenetic variation in different organs of maize hybrids, we performed genome-wide pairwise comparisons of gene expression and each epigenetic modification between organs and between genotypes. For each gene, the relative levels of its expression and epigenetic modifications were normalized to aligned reads per kilobase exon model per million mapped reads (RPKM) and aligned reads per kilobase genic (or genomic) region per million mapped reads (RPKM), respectively. To reduce the bias resulting from different sequencing coverages, only genes or genomic regions with detected transcripts or epigenetic modifications in both compared samples were included in a pairwise comparison. Genes or genomic regions showing significant discrepancy (P<0.05) in expression or epigenetic modifications between biological replicates were excluded from further analyses.
Because the majority (81.8%) of methylated regions are located in intergenic regions in the maize genome (Figure 2b), we also performed hierarchical clustering using detected methylated regions. This analysis identified the same patterns of variation between organs and between genotypes as those found for genic DNA methylation (Figure 3b), thus indicating that patterns of variation in DNA methylation are affected more substantially by genotypes than by organs (Figure 3c). Consistent with this conclusion, a recent study using seedling shoots of the maize inbred lines B73 and Mo17 also suggested extensive variation in DNA methylation . Moreover, further inspection of our data showed that genic DNA methylation in shoots and roots of hybrids exhibited B73-like patterns (Figure 3b), suggesting that parental difference in this repressive epigenetic mark is the main contributor to changed DNA methylation in hybrids, a phenomenon also reported in Arabidopsis hybrids .
Histone modifications are associated with differential gene expression between organs and between hybrids and parents
Next, we explored the relationships between epigenomic variation and transcriptomic diversity between shoots and roots of the maize hybrids. Because the transcripts of most methylated genes (64.2% and 66.4% in shoots and roots, respectively) were undetectable, and because very few genes showed simultaneous variations in gene expression and DNA methylation between organs or between hybrids and parents, genic DNA methylation was excluded from further analysis in this part of the study.
Genes that are upregulated in shoots and roots of hybrids are significantly enriched in the nucleosome assembly pathway
To exploit the biological implication of differential gene expression or epigenetic modifications in different organs of maize hybrids, we examined the functional categories of genes in each pattern of variation. The modes of gene action in hybrids are classified as additive and non-additive, and the latter is further subdivided into upregulation or downregulation relative to the mid-parent value (MPV). To ensure the reliability of the data, we used only sequencing reads that mapped uniquely to the genome of both B73 and Mo17, and used a significance level of P<0.001 to identify the differentially expressed or modified genes between hybrids and parents. To exclude discrepancies between reciprocal hybrids, only genes showing the same pattern of variation in reciprocal hybrids were included in the analyses. Moreover, to reduce the bias resulting from different sequencing coverages, and to make the results comparable between organs, only genes whose transcripts or epigenetic modifications were detected in both shoots and roots of both hybrids and parents were included in the functional analysis. These genes were then subjected to GO analysis using agriGO software . With respect to each epigenetic mark, no biological pathway was significantly enriched in genes showing additive or non-additive epigenetic modifications in hybrids.
We further analyzed the biological functions of genes showing organ-specific non-additive expression, and of genes showing the same pattern of non-additive expression in both organs. Of 255 and 339 genes upregulated only in shoots or roots of reciprocal hybrids, respectively, no biological pathway was significantly enriched. However, the nucleosome assembly pathway was significantly enriched for genes upregulated in both organs (P= 6.8 ´ 10-16, FDR = 7.62 ´ 10-14) (Figure 5b). In addition, no biological pathway was enriched for genes showing either organ-specific downregulation, or in genes downregulated in both organs. These data suggest that upregulation of gene expression may be associated with the nucleosome assembly pathway, and that this association may be a common regulatory mechanism in both shoots and roots of maize hybrids.
Parental alleles contribute similarly to biased expression in both organs of reciprocal hybrids
Our previous study in rice showed that there was no significant parent-of-origin effect for the action of parental alleles in hybrids . In the current study, we also examined whether this conclusion is true or not in maize hybrids. Of 354 and 249 genes with biased expression in shoots and roots of B73 ´ Mo17 and Mo17 ´ B73, respectively, 333 (94.1%) and 222 (89.2%) exhibited the same direction of biased expression in shoots and roots of both hybrids, respectively. Further quantitative analysis showed that in both shoots and roots, there is a strong positive correlation of differential allelic expression between reciprocal hybrids (Figure 6b). These data suggest that, similar to rice, there is no obvious parent-of-origin effect in shoots and roots of maize hybrids that is responsible for the allelic bias.
To investigate whether there are organ-specific effects of allelic expression in maize hybrids, we identified genes showing biased expression in both shoots and roots of hybrids, and compared the direction of biased expression between organs. We found that, of 170 genes with biased expression in both shoots and roots of B73 ´ Mo17, 146 (85.9%) exhibited the same direction of biased expression in both organs of hybrids. Similarly, of 284 genes with biased expression in both organs of Mo17 ´ B73, 261 (91.9%) exhibited the same direction of biased expression. Further quantitative analysis identified a high positive correlation of differential allelic expression between shoots and roots of hybrids (Figure 6c; see Additional file 2, Figure S8), suggesting that the regulatory mechanisms of allelic bias in these organs may be the same.
Small interfering RNAs of 22 and 24 nucleotides in length are derived from distinct transposable elements and are differentially accumulated between hybrids and parents
We also examined sRNA transcriptomes in maize hybrids and their parental lines using sRNA-seq. After removing the adapter sequences and sequencing reads associated with rRNAs, tRNAs, and small nuclear and nucleolar RNAs, sRNA-seq reads were aligned to the reference genome of the maize inbred line B73 (ZmB73_RefGen_v2) . We obtained only a small number of mapped reads from two sRNA-seq libraries (derived from shoots of Mo17 ´ B73 and roots of B73 ´ Mo17), therefore these two libraries were excluded from further analyses.
The sRNA-seq reads corresponding to the precursors of each known maize microRNA (miRNA) were used to characterize miRNA expression in maize hybrids and their parental lines (see Additional file 3, Table S3). The remaining sRNA reads from all libraries were pooled and used to identify 21 nt, 22 nt, and 24 nt siRNA clusters by clustering adjacent sRNA reads of 21 nt, 22 nt, and 24 nt in length, respectively. An siRNA cluster was defined as a region containing a minimum of six sRNA reads, each separated from the nearest neighbor by a maximum of 200 nt. When associated with gene annotations, the identified 21 nt, 22 nt, and 24 nt siRNA clusters were all enriched in a 2 kb area upstream or downstream of the transcribed gene regions (see Additional file 2, Figure S9), similar to the observations in rice  and Arabidopsis , although with different total abundance.
To investigate how siRNAs correlate with DNA methylation and TEs in maize, we analyzed the co-occurrence of TEs with siRNA clusters or DNA methylation across the maize genome. We calculated the number of siRNA clusters or methylated regions with 50% minimum length overlapping with different classes of annotated TEs, and compared the proportion of each class of TEs relative to that in the whole genome. We found that among the 21 nt siRNA clusters co-occurring with TEs, no TE classes were obviously enriched. However, among the 22 nt siRNA clusters co-occurring with TEs, the Copia and Gypsy elements of class I RNA TEs were significantly over-represented (32.2% and 40.1%, respectively; P<0.001, c2 test), whereas the hAT, CACTA, and PIF/Harbinger elements of class II DNA TEs were significantly under-represented (1.8%, 6.7% and 1.6% respectively; P<0.001, c2 test) (Figure 7b). By contrast, among the 24 nt siRNA clusters co-occurring with TEs, Copia and Gypsy were significantly under-represented (11.5% and 9.9%, respectively; P<0.001, c2 test), whereas hAT, CACTA, and PIF/Harbinger were all significantly over-represented (16.5%, 13.4% and 13.3%, respectively; P<0.001, c2 test) (Figure 7b). These data show that 22 nt siRNAs tend to be produced from Copia and Gypsy elements of class I RNA TEs, whereas 24 nt siRNAs tend to be produced from hAT, CACTA, and PIF/Harbinger elements of class II DNA TEs, suggesting that different siRNA species are derived from distinct TE classes.
We also found that among methylated DNA regions co-occurring with TEs, the Gypsy element of class I RNA TEs was significantly over-represented (63.0%; P<0.001, c2 test), whereas Copia, hAT, CACTA, and PIF/Harbinger elements were all under-represented (Figure 7b), suggesting that Gypsy elements are highly methylated in the maize genome. Similar results were also obtained when we analyzed the siRNA clusters or methylated DNA regions that showed variation in shoots or roots of hybrids relative to their parents. These data suggest that in both organs, variations in siRNA activity in hybrids for hAT, CACTA, and PIF/Harbinger elements of class II DNA TEs are primarily driven by 24 nt siRNAs, whereas the differences in siRNA activity between hybrids and patents for Copia and Gypsy elements of class I RNA TEs are primarily driven by 22 nt siRNAs (Figure 7b).
How the combined genomes of parents are regulated in hybrids so as to generate significant differences in genome activities between hybrids and parents is a fundamental biological question. Recent studies suggest that such differences could be ascribed to epigenetic variations [8, 24, 25, 48]. In addition, similarities and differences in gene expression in distinct organs of maize hybrids, such as seedling shoots [37, 38] and roots [39, 40] have also been reported. In this study, we investigated the global variation in transcriptomes and epigenomes in shoots and roots of the B73 and Mo17 inbred lines and their reciprocal hybrids. Our data showed that for each examined epigenetic component in the maize genome, there were no obvious differences in global distribution patterns between organs and between hybrids and parents. However, expression of specific genes or epigenetic modifications at specific genomic loci exhibited significant quantitative variation between hybrids and parents, and between different organs. Our data showed that the patterns of variation in gene expression and each epigenetic modification were distinct. Although many genes showed variation in expression in hybrids, the global patterns of gene expression showed more extensive variation between organs than between hybrids and parents (Figures 3a,c). Conversely, variation in DNA-methylation patterns was more extensive between genotypes than between organs (Figures 3b,c), suggesting a limited contribution of DNA methylation to maize development. Because only a small number of genes are identified with DNA methylation in their transcribed regions (Figure 2b) and very few genes showed simultaneous variation in gene expression and DNA methylation between hybrids and parents, the extensive variation in DNA methylation in maize hybrids were mainly associated with the activity of TEs (especially the Gypsy elements) and therefore would be expected to affect the genomic stability of hybrids (Figure 7b). Moreover, we found that histone modifications varied extensively both between organs and between genotypes (Figures 3a,c; see Additional file 2, Figure S5), and were associated with differential gene expression between organs and between hybrids and parents (Figure 4; see Additional file 2, Figure S6). These results suggest that histone modifications, which are strongly associated with transcribed regions (Figure 2b), play important roles in expression divergence both between organs and between genotypes. By contrast, DNA methylation, which is largely associated with intergenic regions (Figure 2b), may play specific roles in driving the variation in stability and activity of the hybrid genomes by altering the chromatin states.
Many studies have attempted to identify specific gene sets or pathways responsible for hybrid vigor in plants by investigating differential gene expression between hybrids and their parental inbred lines [49, 50]. In the current study, we found that no biological pathway was enriched for genes showing additive or downregulated expression in shoots and roots of reciprocal hybrids; however, genes upregulated in shoots and roots of hybrids were significantly enriched in the nucleosome assembly pathway (Figure 5b; see Additional file 2, Figure S7). As a fundamental biological process required for chromosome replication and maintenance, nucleosome assembly is closely coupled with cell division, and is strongly upregulated during the S-phase of the cell cycle. In addition, it also contributes to the inheritance of chromatin states, and influences the regulation of gene activity and other processes that act on DNA [51, 52]. Because the nucleosome assembly pathway was enriched only for genes upregulated in both organs of hybrids, and because no biological pathway was enriched for genes showing organ-specific expression in hybrids, our data suggest that the enrichment of the nucleosome assembly pathway is likely to be one of the common molecular events in both shoots and roots of maize hybrids. A recent study showed that a dramatic reduction in 24 nt sRNAs strongly affected the expression of genes responsible for chromatin modifications , thus the enrichment of the nucleosome assembly pathway shown in the current study might be a molecular response to the large-scale changes in sRNA profiles and the RdDM pathway. However, whether this pathway is associated with hybrid vigor awaits further investigation. By contrast, no biological pathway was found to be enriched for genes showing downregulated expression in both organs of reciprocal hybrids. This may be a reflection of the fact that downregulated genes, and upregulated genes other than those involved in the nucleosome assembly pathway, are associated with various biological functions resulting from genome-wide genetic variations in hybrids.
Several recent studies have explored the differences in sRNA transcriptomes between hybrids and parents [8, 21, 23, 24, 25, 26, 27, 28, 29, 30]. In the current study, we found that much more siRNA clusters were downregulated (11,558) than upregulated (2,911) in both organs of maize hybrids, consistent with the observation that siRNAs tend to be downregulated in hybrids [8, 21, 25, 28]. Because siRNAs are involved in transcriptional silencing of TEs through the RdDM pathway , we therefore investigated the relationships between siRNAs, DNA methylation, and different TE classes in the maize genome. A recent study showed that 21 and 22 nt siRNAs are derived from distinct retrotransposon families, and are differentially accumulated between the maize inbred lines B73 and Mo17 and their hybrids . Another study also identified a distinct size preference of sRNAs resulting from different TE families, and suggested the 22 nt sRNAs as a major component in the silencing of most TE families in soybean . Our data also show that different TE classes tend to produce distinct siRNA species (Figure 7b). This observation suggests divergent effects of different TEs on chromatin states, as reported in a recent study, which reported that there are family-specific attributes for the effects of TEs on neighboring chromatin .
In addition, we found that whereas DNA methylation was found to be highly enriched in heterochromatic regions, siRNA clusters were not obviously enriched (for 22 nt siRNAs), or even devoid (for 21 and 24 nt siRNAs) in these regions (Figure 7a; see Additional file 2, Figure S10). However, a positive correlation between 22 nt and 24 nt siRNAs and DNA-methylation levels at the same genomic loci were also seen (Figure 7c; see Additional file 2, Figure S11). A possible explanation for these observations may be that both siRNA-dependent and siRNA-independent pathways are responsible for methylation of TEs in the maize genome, or alternatively, TE sequences in heterochromatic regions acquire DNA methylation through spreading from adjacent siRNA-targeted regions .
In this study, we investigated the conservation and divergence of transcriptomic and epigenomic variations in shoots and roots of two maize inbred lines and their reciprocal hybrids. The global distribution patterns of epigenetic components between parents and hybrids contained extensive variations in the levels of DNA methylation, histone modifications, and siRNA transcription, which are conserved between shoots and roots. These diverse epigenetic variations potentially make important contributions to altered genome activity in different organs of hybrids compared with their parents by modulating chromatin states so as to accommodate hybridization. Confirmation of these results awaits further studies exploring integrated transcriptomic and epigenomic profiling with more extensive sequencing in more organs of more hybrids.
Materials and methods
Plant materials and growth conditions
The inbred lines B73 and Mo17 of maize (Zea mays) and their reciprocal F1 hybrids (B73 ´ Mo17 and Mo17 ´ B73) were used in this study. Seeds were grown in soil under controlled environmental conditions (15 hours of light at 25°C, and 9 hours dark at 20°C) in a growth chamber. After 14 days, seedling shoots and roots were harvested, then, frozen in liquid nitrogen and stored at -80°C for isolation of DNA and total RNA, or processed directly for ChIP assays after harvesting.
Sample preparation and sequencing library construction
Three independent biological replicates, each consisting of three pooled shoots or roots of the hybrids and parental lines, were used for constructing mRNA-seq, McrBC-seq, ChIP-seq, and sRNA-seq libraries, and each library was sequenced in a single lane as described previously [3, 7, 8]. Briefly, total RNAs were isolated using TRIzol reagent (Invitrogen Corp., Carlsbad, CA, USA) and treated with RNase-free DNase I (New England Biolabs, Ipswich, MA, USA) to remove any contaminating genomic DNA. mRNA extraction was performed using Dynabeads oligo(dT) (Dynal; Invitrogen Corp.). Double-stranded cDNAs were synthesized using reverse transcriptase (Superscript II; Invitrogen Corp.) and random hexamer primers. The cDNAs were then fragmented by nebulization, and the standard Illumina protocol was followed thereafter to create the mRNA-seq libraries. Genomic DNAs were isolated using a commercial kit (DNeasy Plant Maxi Kit; Qiagen Inc., Valencia, CA, USA). Isolated genomic DNAs were then digested with McrBC (New England Biolabs) followed by gel purification to enrich methylated genomic DNAs. The McrBC-seq libraries were generated using the standard Illumina protocol. The ChIP-seq libraries were generated by immunoprecipitating chromatin with antibodies against H3K4me3 (Abcam, Cambridge, UK), H3K9ac (Upstate Biotechnology, Lake Placid, NY, USA), or H3K36me3 (Abcam), as described previously . The eluted ChIP DNAs from the three ChIP reactions were pooled to generate ChIP-seq libraries for Illumina sequencing, following the manufacturer's protocol. sRNAs were gel-purified from total RNAs, and were subsequently ligated with 3' and 5' adapters, followed by reverse transcription using a 3' reverse transcriptase primer. The cDNAs were then amplified by PCR using primers specific to sRNAs . After gel purification, the sRNA-seq libraries were subjected to Illumina sequencing following the manufacturer's protocol.
The original datasets have become public in the NIH GEO database under the accession [GEO: GSE43142].
Data processing and analyses
For the methods used in the analysis of data from mRNA-seq, McrBC-seq, ChIP-seq and sRNA-seq, see Additional file 4.
This work was supported by grants from the National Basic Research Program of China (973 Program) (2012CB910900), the National Science Foundation (NSF) Plant Genome Program (DBI0922604), the National Institutes of Health (GM047850), the NSF 2010 program (MCB-0929100), the Ministry of Science and Technology of China (2009DFB30030), the Ministry of Agriculture of China (2009ZX08012-021B), the National High Technology Research and Development Program of China (863 Program: 2012AA10A304), the Ministry of Agriculture of China (948 Program: 2011-G2B), and the Peking-Tsinghua Center for Life Sciences. We are also grateful to Dr Cynthia Nezames and Cynthia Chan for proofreading the manuscript.
- 7.Wang X, Elling AA, Li X, Li N, Peng Z, He G, Sun H, Qi Y, Liu XS, Deng XW: Genome-wide and organ-specific landscapes of epigenetic modifications and their relationships to mRNA and small RNA transcriptomes in maize. Plant Cell. 2009, 21: 1053-1069. 10.1105/tpc.109.065714.CrossRefPubMedPubMedCentralGoogle Scholar
- 14.Li X, Wang X, He K, Ma Y, Su N, He H, Stolc V, Tongprasit W, Jin W, Jiang J, Terzaghi W, Li S, Deng XW: High-resolution mapping of epigenetic modifications of the rice genome uncovers interplay between DNA methylation, histone methylation, and gene expression. Plant Cell. 2008, 20: 259-276. 10.1105/tpc.107.056879.CrossRefPubMedPubMedCentralGoogle Scholar
- 15.Zabala G, Campos E, Varala KK, Bloomfield S, Jones SI, Win H, Tuteja JH, Calla B, Clough SJ, Hudson M, Vodkin LO: Divergent patterns of endogenous small RNA populations from seed and vegetative tissues of Glycine max. BMC Plant Biol. 2012, 12: 177-10.1186/1471-2229-12-177.CrossRefPubMedPubMedCentralGoogle Scholar
- 16.Vaughn MW, Tanurdzic M, Lippman Z, Jiang H, Carrasquillo R, Rabinowicz PD, Dedhia N, McCombie WR, Agier N, Bulski A, Colot V, Doerge RW, Martienssen RA: Epigenetic natural variation in Arabidopsis thaliana. PLoS Biol. 2007, 5: e174-10.1371/journal.pbio.0050174.CrossRefPubMedPubMedCentralGoogle Scholar
- 18.Feng S, Cokus SJ, Zhang X, Chen PY, Bostick M, Goll MG, Hetzel J, Jain J, Strauss SH, Halpern ME, Ukomadu C, Sadler KC, Pradhan S, Pellegrini M, Jacobsen SE: Conservation and divergence of methylation patterning in plants and animals. Proc Natl Acad Sci USA. 2010, 107: 8689-8694. 10.1073/pnas.1002720107.CrossRefPubMedPubMedCentralGoogle Scholar
- 20.Eichten SR, Swanson-Wagner RA, Schnable JC, Waters AJ, Hermanson PJ, Liu S, Yeh CT, Jia Y, Gendler K, Freeling M, Schnable PS, Vaughn MW, Springer NM: Heritable epigenetic variation among maize inbreds. PLoS Genet. 2011, 7: e1002372-10.1371/journal.pgen.1002372.CrossRefPubMedPubMedCentralGoogle Scholar
- 22.Waters AJ, Makarevitch I, Eichten SR, Swanson-Wagner RA, Yeh CT, Xu W, Schnable PS, Vaughn MW, Gehring M, Springer NM: Parent-of-origin effects on gene expression and DNA methylation in the maize endosperm. Plant Cell. 2011, 23: 4221-4233. 10.1105/tpc.111.092668.CrossRefPubMedPubMedCentralGoogle Scholar
- 25.Shen H, He H, Li J, Chen W, Wang X, Guo L, Peng Z, He G, Zhong S, Qi Y, Terzaghi W, Deng XW: Genome-wide analysis of DNA methylation and gene expression changes in two Arabidopsis ecotypes and their reciprocal hybrids. Plant Cell. 2012, 24: 875-892. 10.1105/tpc.111.094870.CrossRefPubMedPubMedCentralGoogle Scholar
- 26.Ha M, Lu J, Tian L, Ramachandran V, Kasschau KD, Chapman EJ, Carrington JC, Chen X, Wang XJ, Chen ZJ: Small RNAs serve as a genetic buffer against genomic shock in Arabidopsis interspecific hybrids and allopolyploids. Proc Natl Acad Sci USA. 2009, 106: 17835-17840. 10.1073/pnas.0907003106.CrossRefPubMedPubMedCentralGoogle Scholar
- 30.Li T, Chen J, Qiu S, Zhang Y, Wang P, Yang L, Lu Y, Shi J: Deep sequencing and microarray hybridization identify conserved and species-specific microRNAs during somatic embryogenesis in hybrid Yellow Poplar. PLoS One. 2012, 7: e43451-10.1371/journal.pone.0043451.CrossRefPubMedPubMedCentralGoogle Scholar
- 31.Schnable PS, Ware D, Fulton RS, Stein JC, Wei F, Pasternak S, Liang C, Zhang J, Fulton L, Graves TA, Minx P, Reily AD, Courtney L, Kruchowski SS, Tomlinson C, Strong C, Delehaunty K, Fronick C, Courtney B, Rock SM, Belter E, Du F, Kim K, Abbott RM, Cotton M, Levy A, Marchetto P, Ochoa K, Jackson SM, Gillam B, et al: The B73 maize genome: complexity, diversity, and dynamics. Science. 2009, 326: 1112-1115. 10.1126/science.1178534.CrossRefPubMedGoogle Scholar
- 33.Jahnke S, Sarholz B, Thiemann A, Kuhr V, Gutierrez-Marcos JF, Geiger HH, Piepho HP, Scholten S: Heterosis in early seed development: a comparative study of F1 embryo and endosperm tissues 6 days after fertilization. Theor Appl Genet. 2010, 120: 389-400. 10.1007/s00122-009-1207-y.CrossRefPubMedGoogle Scholar
- 37.Swanson-Wagner RA, Jia Y, DeCook R, Borsuk LA, Nettleton D, Schnable PS: All possible modes of gene action are observed in a global comparison of gene expression in a maize F1 hybrid and its inbred parents. Proc Natl Acad Sci USA. 2006, 103: 6805-6810. 10.1073/pnas.0510430103.CrossRefPubMedPubMedCentralGoogle Scholar
- 39.Hoecker N, Keller B, Muthreich N, Chollet D, Descombes P, Piepho HP, Hochholdinger F: Comparison of maize (Zea mays L.) F1-hybrid and parental inbred line primary root transcriptomes suggests organ-specific patterns of nonadditive gene expression and conserved expression trends. Genetics. 2008, 179: 1275-1283. 10.1534/genetics.108.088278.CrossRefPubMedPubMedCentralGoogle Scholar
- 46.Guo M, Yang S, Rupe M, Hu B, Bickel DR, Arthur L, Smith O: Genome-wide allele-specific expression analysis using massively parallel signature sequencing (MPSS) reveals cis- and trans-effects on gene expression in maize hybrid meristem tissue. Plant Mol Biol. 2008, 66: 551-563. 10.1007/s11103-008-9290-z.CrossRefPubMedGoogle Scholar
- 53.Jia Y, Lisch DR, Ohtsu K, Scanlon MJ, Nettleton D, Schnable PS: Loss of RNA-dependent RNA polymerase 2 (RDR2) function causes widespread and unexpected changes in the expression of transposons, genes, and 24-nt small RNAs. PLoS Genet. 2009, 5: e1000737-10.1371/journal.pgen.1000737.CrossRefPubMedPubMedCentralGoogle Scholar
- 55.Eichten SR, Ellis NA, Makarevitch I, Yeh C-T, Gent JI, Guo L, McGinnis KM, Zhang X, Schnable PS, Vaughn MW, Dawe RK, Springer NM: Spreading of heterochromatin is limited to specific families of maize retrotransposons. PLoS Genet. 2012, 8: e1003127-10.1371/journal.pgen.1003127.CrossRefPubMedPubMedCentralGoogle Scholar
This article is published under license to BioMed Central Ltd. This is an open access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.