Transcriptomic analysis highlights epigenetic and transcriptional regulation during zygotic embryo development of Pinus pinaster

Abstract

Background: It is during embryogenesis that the plant body plan is established and the meristems responsible for all post-embryonic growth are specified. The molecular mechanisms governing conifer embryogenesis are still largely unknown. Their elucidation may contribute valuable information to clarify if the distinct features of embryo development in angiosperms and gymnosperms result from differential gene regulation. To address this issue, we have performed the first transcriptomic analysis of zygotic embryo development in a conifer species (Pinus pinaster) focusing our study in particular on regulatory genes playing important roles during plant embryo development, namely epigenetic regulators and transcription factors.

Results: Microarray analysis of P. pinaster zygotic embryogenesis was performed at five periods of embryo development from early developing to mature embryos. Our results show that most changes in transcript levels occurred in the first and the last embryo stage-to-stage transitions, namely early to pre-cotyledonary embryo and cotyledonary to mature embryo. An analysis of functional categories for genes that were differentially expressed through embryogenesis highlighted several epigenetic regulation mechanisms. While putative orthologs of transcripts associated with mechanisms that target transposable elements and repetitive sequences were strongly expressed in early embryogenesis, PRC2-mediated repression of genes seemed more relevant during late embryogenesis. On the other hand, functions related to sRNA pathways appeared differentially regulated across all stages of embryo development with a prevalence of miRNA functions in mid to late embryogenesis. Identification of putative transcription factor genes differentially regulated between consecutive embryo stages was strongly suggestive of the relevance of auxin responses and regulation of auxin carriers during early embryogenesis. Such responses could be involved in establishing embryo patterning. Later in development, transcripts with homology to genes acting on modulation of auxin flow and determination of adaxial-abaxial polarity were up-regulated, as were putative orthologs of genes required for meristem formation and function as well as establishment of organ boundaries. Comparative analysis with A. thaliana embryogenesis also highlighted genes involved in auxin-mediated responses, as well as epigenetic regulation, indicating highly correlated transcript profiles between the two species.

Conclusions: This is the first report of a time-course transcriptomic analysis of zygotic embryogenesis in a conifer. Taken together our results show that epigenetic regulation and transcriptional control related to auxin transport and response are critical during early to mid stages of pine embryogenesis and that important events during embryogenesis seem to be coordinated by putative orthologs of major developmental regulators in angiosperms.

Figure 1

Microarray hybridization. (A) Staging system (T0 to T7) used for Pinus pinaster zygotic embryo development [15], showing how samples were divided into five developmental groups/time-points representing early embryogenesis (T0 to T2), pre- cotyledonary (T3 and T4), early cotyledonary (T4B), late embryogenesis (T5) and mature embryo (T7). Bar: T0 and T1 = 300 μm; T2, T3 and T4 = 400 μm; T4B = 800 μm; T5, T6 and T5 = 0.1 cm. Three biological replicates of each sample harvested on Day 0 to Day 25 and two technical replicates were used for hybridization with the reference sample, which consisted of a pool containing equal amounts of total RNA from all five time points. (B) Cluster analysis of the thirty replicates generated using MeV with Pearson correlation and average linkage.

Figure 2

Clustering of functional categories that showed similar expression profiles during P. pinaster zygotic embryogenesis. Functional categories include gene ontologies, plant ontologies, enzyme codes, pathways, families and structures annotations. The expression of a category is represented by the median expression values of the genes annotated within that category. Dots indicate the distribution (mean and quartiles) of the median values for the categories included in each profile/cluster. Functional categories are listed in the inset.

Figure 3

Clustering of differentially expressed genes according to their expression profiles during P. pinaster zygotic embryogenesis, and representative gene ontology terms in each cluster. Transcripts having similar expression profiles, which were differentially expressed in time according to maSigPro analysis, were clustered together, and a representative median expression profile was inferred from the expression of all the genes in each cluster. For each cluster, the number of transcripts in each gene ontology term is indicated in a bar graph.

Figure 4

Venn diagrams of genes regulated between two consecutive stages. The number of genes showing a fold-change >2 between consecutive stages is shown. For each transition, genes showing increasing (A) or decreasing (B) expression between consecutive stages are represented in different diagrams.

Figure 5

Validation of microarray transcript profiles. Fold-changes for selected transcripts obtained by microarray analysis and RT-qPCR are shown for each developmental time point. Insets represent the profiles based on microarray M-values intensity ratios (triangles) or RT-qPCR relative expression normalized values (squares).