Abstract:
Abstract
Background: Somatic embryogenesis (SE) in plants is a process by which embryos are generated directly from
somatic cells, rather than from the fused products of male and female gametes. Despite the detailed expression
analysis of several somatic-to-embryonic marker genes, a comprehensive understanding of SE at a molecular level is
still lacking. The present study was designed to generate high resolution transcriptome datasets for early SE providing
the way for future research to understand the underlying molecular mechanisms that regulate this process. We
sequenced Arabidopsis thaliana somatic embryos collected from three distinct developmental time-points (5, 10
and 15 d after in vitro culture) using the Illumina HiSeq 2000 platform.
Results: This study yielded a total of 426,001,826 sequence reads mapped to 26,520 genes in the A. thaliana
reference genome. Analysis of embryonic cultures after 5 and 10 d showed differential expression of 1,195 genes;
these included 778 genes that were more highly expressed after 5 d as compared to 10 d. Moreover, 1,718 genes
were differentially expressed in embryonic cultures between 10 and 15 d. Our data also showed at least eight
different expression patterns during early SE; the majority of genes are transcriptionally more active in embryos
after 5 d. Comparison of transcriptomes derived from somatic embryos and leaf tissues revealed that at least 4,951
genes are transcriptionally more active in embryos than in the leaf; increased expression of genes involved in DNA
cytosine methylation and histone deacetylation were noted in embryogenic tissues. In silico expression analysis
based on microarray data found that approximately 5% of these genes are transcriptionally more active in somatic
embryos than in actively dividing callus and non-dividing leaf tissues. Moreover, this identified 49 genes expressed
at a higher level in somatic embryos than in other tissues. This included several genes with unknown function, as
well as others related to oxidative and osmotic stress, and auxin signalling.
Conclusions: The transcriptome information provided here will form the foundation for future research on
genetic and epigenetic control of plant embryogenesis at a molecular level. In follow-up studies, these data could
be used to construct a regulatory network for SE; the genes more highly expressed in somatic embryos than in
vegetative tissues can be considered as potential candidates to validate these networks.