Identification of novel genes potentially involved in somatic embryogenesis in chicory (Cichorium intybus L.)

Abstract

Background: In our laboratory we use cultured chicory (Cichorium intybus) explants as a model to investigate cell reactivation and somatic embryogenesis and have produced 2 chicory genotypes (K59, C15) sharing a similar genetic background. K59 is a responsive genotype (embryogenic) capable of undergoing complete cell reactivation i.e. cell de- and re-differentiation leading to somatic embryogenesis (SE), whereas C15 is a non-responsive genotype (non-embryogenic) and is unable to undergo SE. Previous studies 1 showed that the use of the beta-D-glucosyl Yariv reagent (beta-GlcY) that specifically binds arabinogalactan-proteins (AGPs) blocked somatic embryo production in chicory root explants. This observation indicates that beta-GlcY is a useful tool for investigating somatic embryogenesis (SE) in chicory. In addition, a putative AGP (DT212818) encoding gene was previously found to be significantly up-regulated in the embryogenic K59 chicory genotype as compared to the non-embryogenic C15 genotype suggesting that this AGP could be involved in chicory re-differentiation 2. In order to improve our understanding of the molecular and cellular regulation underlying SE in chicory, we undertook a detailed cytological study of cell reactivation events in K59 and C15 genotypes, and used microarray profiling to compare gene expression in these 2 genotypes. In addition we also used beta-GlcY to block SE in order to identify genes potentially involved in this process.

Results: Microscopy confirmed that only the K59, but not the C15 genotype underwent complete cell reactivation leading to SE formation. beta-GlcY-treatment of explants blocked in vitro SE induction, but not cell reactivation, and induced cell wall modifications. Microarray analyses revealed that 78 genes were differentially expressed between induced K59 and C15 genotypes. The expression profiles of 19 genes were modified by beta-GlcY-treatment. Eight genes were both differentially expressed between K59 and C15 genotypes during SE induction and transcriptionally affected by beta-GlcY-treatment: AGP (DT212818), 26 S proteasome AAA ATPase subunit 6 (RPT6), remorin (REM), metallothionein-1 (MT1), two non-specific lipid transfer proteins genes (SDI-9 and DEA1), 3-hydroxy-3-methylglutaryl-CoA reductase (HMG-CoA reductase), and snakin 2 (SN2). These results suggest that the 8 genes, including the previously-identified AGP gene (DT212818), could be involved in cell fate determination events leading to SE commitment in chicory.

Conclusion: The use of two different chicory genotypes differing in their responsiveness to SE induction, together with beta-GlcY-treatment represented an efficient tool to discriminate cell reactivation from the SE morphogenetic pathway. Such an approach, together with microarray analyses, permitted us to identify several putative key genes related to the SE morphogenetic pathway in chicory.

Figure 1

Cell reactivation and somatic embryogenesis in leaf explants from two chicory genotypes. A: Sections of K59 leaf explants from in vitro plantlets at d0. B: Early reactivating cells. C: Late reactivating cells. D: The second phase of the cell reactivation (d4). E: Cell divisions (d4). Daughter-cells (1, 2, 3) derived from mitosis of a single FRC. F: Morphogenesis (d11). Somatic embryos (SE) are detected as compact, rectangular-shaped and are composed of dense-embryonary cells. G: Sections of C15 leaf explants from in vitro plantlets at d0. The first phase of cell reactivation in C15 leaf explants proceeds in as similar way to that observed in K59: increase of nuclear volume (H), relocation of plastids around the voluminous nucleus and occurrence of a partial plastid crown (I). Neither fully reactivated cells nor cell division events can be observed. J: Evolution of nuclear- and nucleolar-volumes during cell reactivation in the K59 genotype. The first phase of cell reactivation is characterised by a significant increase in both nuclear volume (white bar, × 40 volume increase) and nucleolar volume (black triangle, × 67) volume increase) as compared to those of differentiated mesophyll cells. The second phase of cell reactivation is characterised by a slight increase of nuclear volume (white bar, × 1.5) and nucleolar volume (black triangle, × 1.9) as compared to the first phase of cell reactivation. Fully reactivated-cell status is given to those dedifferentiated cells able to re-enter the cell cycle in only K59. mc: differentiated mesophyll cells; mv: minor veins; RC: reactivating cells; nu: nucleus; chlp-p: chloroplasts in parietal position; chlp-n: perinuclear crown of chloroplasts; n = 10 semi-thin sections from K59 leaf explants. Bars represent standard error. A-L: Three micrometer-semi-thin sections stained with TBO. Bar = 20 μm.

Figure 2

Characterisation of phenotypes observed in chicory leaf explants from the responsive K59 genotype and the non-responsive C15 genotype, cultured in absence (control condition Y-), or in presence (Y+) of β-Glc Y. A: Leaf explants of K59 at d0. B: Leaf explants of K59 at d11,Y-. Somatic embryos (e) can be observed at explant borders. Roots (r) can be observed during embryo development. C: Leaf explants of K59 at d11,Y+. Yariv reagent penetrates within the leaf blade. No morphogenetic pattern can be detected. D: leaf explants of K59 cultured for 4 days in presence of β-Glc Y and transferred to new medium deprived of β-Glc Y during 7 days (d11, 4Y+). Cut explant borders curl upwards and somatic embryos (e) can be observed. E: Leaf explants of C15 at d0. F: Leaf explants of C15 at d11,Y-. No morphogenetic pattern can be detected. G: Leaf explants of C15 at d11,Y+. Yariv reagent penetrates within the leaf blade. No morphogenetic pattern can be observed. H: leaf explants of C15 cultured for 4 days in presence of β-Glc Y and transferred to a new medium deprived of β-Glc Y during 7 days (d11, 4Y+). No somatic embryos can be detected. Bar = 2 mm.

Figure 3

Effects of the β-1,3-D Yariv treatment during SE induction. A and B: Effects of β-1,3-D Yariv treatment on the occurrence (white columns) of reactivating cells (RC), fully reactivated cells (FRC) and somatic embryos (SE) in K59 explants at d4 and d11. All RC are not able to go through the second phase of cell reactivation to reach FRC status, however, the majority of FRC will form somatic embryos at d11. When β-1,3-D Yariv is applied, the numbers of RC and FRC are significantly decreased, and no somatic embryos are produced at d11. C and D: The Yariv reagent is clearly detected as an orange coloration within leaf tissue and it can be observed surrounding reactivating cells (arrows). Three micrometer semi-thin sections observed under Nomarski phase contrast. Bar = 20 μm. Hatched columns = control conditions (absence of β-1,3-D Yariv). The Student-Newman-Keuls test was applied on data collected from 10 independent slides. For one event, different letters indicate significant differences. Bars represent standard errors.

Figure 4

Venn diagram depicting gene expression profiles in K59 and C15 genotypes following SE induction. The circle E (embryogenic) represents the number of genes specifically expressed in K59 genotype (up-regulated or down-regulated in contrast with C15) or differentially expressed (with opposite profiles in K59 and C15) after 4 days of in vitro induction of SE (K59 d4 vs. K59 d0). The circle NE (non-embryogenic) represents the number of genes specifically or differentially expressed in the C15 genotype after 4 days of in vitro induction of SE (C15 d4 vs. C15 d0). The β-Glc Y ring represents the number of genes (in both K59 and C15 genotypes) whose expression is modified after 4 days of in vitro induction in presence of β-Glc Y (K59 or C15 4 d in vitro induction in presence of β-GlcY vs. K59 or C15 4 d in vitro induction in absence of β-GlcY). u: up-regulated; d: down-regulated.

Figure 5

Ternary clusters representing expression profiles of genes affected by β-GlcY-treatment. A: genes specifically or differentially expressed in K59 and C15 genotypes, and whose expression is affected by β-GlcY; B: genes whose expression is affected by β-GlcY-treatment but is not modified during cell reactivation in K59 and C15 genotypes. Columns represents K59 d4 vs. K59 d0 (a); C15 d4 vs. C15 d0 (b); K59 d4 in presence of β-GlcY vs. K59 d4 in absence of β-GlcY (c); C15 d4 in presence of β-GlcY vs. C15 d4 in absence of β-GlcY (d). Red colour indicates gene induction, whereas green colour indicates repression. Induction corresponds to a log2 ratio ≥1, and repression to a log2 ratio ≤ -1. The column (e) corresponds to chicory genes (black circle) that have previously been described in the literature as being associated with wound-/pathogen responses in Arabidopsis thaliana.