IC4R016-RNA-Seq-2013-24215296

Project Title

• Transcriptomes of isolated Oryza sativa gametes characterized by deep sequencing: evidence for distinct sex-dependent chromatin and epigenetic states before fertilization

The Background of This Project

• The formation of a zygote by the fusion of egg and sperm involves the two gametic transcriptomes. In flowering plants, the embryo sac embedded within the ovule contains the egg cell, whereas the pollen grain contains two sperm cells inside a supporting vegetative cell. The difficulties of collecting isolated gametes and consequent low recovery of RNA have restricted in-depth analysis of gametic transcriptomes in flowering plants. The researchers isolated living egg cells, sperm cells and pollen vegetative cells from Oryza sativa (rice), and identified transcripts for approximately 36 000 genes by deep sequencing.

Plant Culture & Treatment

• Egg cells were isolated using flowers harvested at anthesis from rice plants (O. sativa L. ssp. japonica, cultivar ‘Kitaake’) grown in glasshouses under ambient light during the summer.
• Sperm cells and Ve were isolated from near anthesis anthers of vigorous, visibly disease-free, field grown rice at Dale Bumpers National Rice Research Center in Stuttgart, AR, USA. A differential centrifugation isolation method was used to separate Sp and Ve from whole anthers.
• RNA from the Sp and Ve was extracted using the QIAGEN RNeasy Plant Mini Kit (Qiagen, http://www.qiagen.com) following the manufacturer’s instructions, including the optional on-column DNAse treatment.
• The barcoded FASTQ files of the reads were mapped to rice RGAP 7.0 genomic reference sequences using TOPHAT 2.0.5 with default settings, except that the ‘microexon’ search option was switched on, and minimum and maximum intron sizes were set at 20 and 15000 bases, respectively.

Illumina sequencing

• Illumina sequencing by the RNA-seq protocol was performed on RNA samples from three biological replicates for each of the three cell types.Samples were multiplexed and run with six samples per lane on the Illumina HiSeq 2000 DNA sequencer (Illumina, http://www.illumina.com).

Research Findings

• Using six possible contrasts on the fitted data, approximately threequarters of all the genes assessed among the three cells were considered as differentially expressed (DE), with a false discovery rate (FDR) of 5% (Appendix S4; Figure S1F).For the pairwise comparisons between the three cells, the null hypothesis of the absence of differential expression could not be rejected for 14 236, 20 650 and 16 627 genes,corresponding to EC versus Sp, Sp versus Ve, and EC versus Ve, respectively. The higher number of genes in common for the Sp and Ve comparison is perhaps not surprising, as Sp and VEs constitute the male gametophytes, and, importantly, are also physically associated with one another (Mogensen, 1992). From the results of the six contrasts, the genes were grouped into six differential expression sets (see Experimental procedures) and are presented in a Venn diagram (Figure 2a).

'Figure 2. (a) Venn diagram demonstrating differential gene expression at the gene or gene ontology (GO) level. The numbers of differentially expressed genes that are significantly overexpressed in each of six possible places are shown: three individual cell types (EC, Sp and Ve) and three cell–cell combinations(EC–Sp, Sp–Ve and Ve–EC).(b) GO terms that are uniquely enriched within each of these gene sets are color coded for GO categories: biological process (red);cellular component (green); and molecular function (blue). '

• In general, as shown in Figure 3, we found that the histone variants expressed in the EC were distinct from those expressed in the Sp and Ve, with the latter histone variants showing some similarities.

'Figure 3. Heat map of histone variants. Numbers in cells represent tpm values derived from RNA-seq of egg cells (EC), sperm cells (Sp) and vegetative cells(Ve), in triplicate. Orthology to Arabidopsis (At*) is derived from two histone trees'

• The Polycomb Repressive Complex 2 (PRC2) acts as an H3K27 methyltransferase. This complex contains four protein subunits: MSI1, FIE1 or FIE2, EZ1 or CLF, and EMF2A or EMF2B. In our transcriptome data, we found that the genes encoding all four components of PRC2 (Luo et al., 2009;Gleason and Kramer, 2012) are expressed in both gametes at maturity. The EC shows expression of all possible subunits of the PRC2 except for FIE1 (Figure 4a). The sperm PRC2 appeared to be more specialized, with a putative PRC2 consisting of MSI1, FIE2, CLF and EMF2B (Figure 4a).There are 20 JMJ genes in rice with a range of putative targets on histone H3 (Chen et al., 2011). JMJ proteins seem to be widely expressed in both gametes, with slightly lower expression for some families in the Ve (Figure 4b).The researchers found that all four rice LDL genes are expressed in the egg, whereas the sperm had moderate expression of LDL3 (Figure 4c).

'Figure 4. Heat map of histone modification genes. Genes in the Polycomb Repressive Complex 2 (PRC2) (a), Jumonji (JMJ) histone demethylases (b) and LSDlike (LDL) histone demethylases (c) are shown. Numbers in cells represent transcripts per million (tpm) values derived from RNA-seq of egg cells (EC), sperm cells (Sp) and vegetative cells (Ve), in triplicate.'

Labs working on this Project

• Department of Plant Biology,University of California, Davis, CA 95616, USA,
• Department of Microbiology and Plant Biology, University of Oklahoma, Norman, OK 73019, USA,
• School of Life Sciences, Lanzhou University, Lanzhou 730000, China, and
• Department of Plant Sciences, University of California, Davis, CA 95616, USA

Corresponding Author

• Sarah N. Anderson:srussell@ou.edu & Venkatesan Sundaresan:sundar@ucdavis.edu