Strigolactones Might Regulate Ovule Development after Fertilization in Xanthoceras sorbifolium

Abstract

Strigolactones (SLs) were recently defined as a novel class of plant hormones that act as key regulators of diverse developmental processes and environmental responses. Much research has focused on SL biosynthesis and signaling in roots and shoots, but little is known about whether SLs are produced in early developing seeds and about their roles in ovule development after fertilization. This study revealed that the fertilized ovules and early developing pericarp in Xanthoceras sorbifolium produced minute amounts of two strigolactones: 5-deoxystrigol and strigol. Their content decreased in the plants with the addition of exogenous phosphate (Pi) compared to those without the Pi treatment. The exogenous application of an SL analog (GR24) and a specific inhibitor of SL biosynthesis (TIS108) affected early seed development and fruit set. In the Xanthoceras genome, we identified 69 potential homologs of genes involved in SL biological synthesis and signaling. Using RNA-seq to characterize the expression of these genes in the fertilized ovules, 37 genes were found to express differently in the fertilized ovules that were aborting compared to the normally developing ovules. A transcriptome analysis also revealed that in normally developing ovules after fertilization, 12 potential invertase genes were actively expressed. Hexoses (glucose and fructose) accumulated at high concentrations in normally developing ovules during syncytial endosperm development. In contrast, a low ratio of hexose and sucrose levels was detected in aborting ovules with a high strigolactone content. XsD14 virus-induced gene silencing (VIGS) increased the hexose content in fertilized ovules and induced the proliferation of endosperm free nuclei, thereby promoting early seed development and fruit set. We propose that the crosstalk between sugar and strigolactone signals may be an important part of a system that accurately regulates the abortion of ovules after fertilization. This study is useful for understanding the mechanisms underlying ovule abortion, which will serve as a guide for genetic or chemical approaches to promote seed yield in Xanthoceras.

Keywords: Xanthoceras sorbifolium; fertilized ovules; invertase; strigolactone signals; sugar.

Figure 1

Chemical structures of two natural strigolactones (SLs). Both strigol and 5-deoxystrigol contain a methylbutenolide moiety (D-ring) connected to a variable tricyclic lactone (ABC-ring), which are termed canonical SLs.

Figure 2

Strigolactone content in the fertilized ovule and the pericarp in the Xanthoceras plants under phosphate (Pi)-deficient conditions (control) and with Pi treatment. The values are the means of three biological replicates with standard error bars. Different letters represent significant differences at p < 0.05 (ANOVA, Tukey’s HSD test). 5DS: 5-deoxystrigol.

Figure 3

Effect of exogenous phosphate (Pi) application on expression level of the Pi starvation-induced gene PHO2 in fertilized ovules of the Xanthoceras plants without Pi application and with Pi application, which were normalized to the expression level of actin-2 as a reference. The values are the means of five biological replicates with standard error bars. Asterisks indicate significant difference at p < 0.05 (ANOVA, Tukey’s HSD test).

Figure 4

Histological sections of the ovules after fertilization in Xanthoceras sorbifolium. (A): A portion of a transverse resin section of the normal ovules with low SL levels, showing a rapidly expanding embryo sac and a normally developing syncytial endosperm; (B): A portion of a transverse resin section of the abnormal ovules with high SL contents, showing that development of the embryo sac was arrested and syncytial endosperm degenerated. (C): A many-celled proembryo was formed at the micropylar end of embryo sac in the normally developing ovule 18 days after pollination (DAP). (D). The longitudinal section of the XsD14-silenced ovule 7 DAP, showing delayed degeneration. EM: embryo, END: endosperm, ES: embryo sac, FNE: free nuclear endosperm, II: inner integument, NU: nucellus, OI: outer integument.

Figure 5

Response of fruit set to treatment with exogenous GR24, a synthetic strigolactone analog, and TIS108, a specific inhibitor of SL biosynthesis, in inflorescences of Xanthoceras sorbifolium plants of different genotypes. The same concentration of acetone was used as mock treatment. The values are the means of three biological replicates with standard error bars. Different letters represent significant differences at p < 0.05 (ANOVA, Tukey’s HSD test).

Figure 6

Expression analysis of the genes encoding Xanthoceras homologs of ACC oxidase (1-aminocyclopropane-1-carboxylic acid oxidase, ACC oxidase, ACO) ACO2 and ACO3 and encoding vacuolar processing enzyme (VPE) VPE2 and VPE4 in the fertilized ovules of the inflorescences with the GR24 or control treatment. The values are the means of three biological replicates with standard error bars. Asterisks indicate significant difference at p < 0.05 (ANOVA, Tukey’s HSD test).

Figure 7

Quantitative real-time PCR (qRT-PCR) verification of expression of the selected SL biosynthesis-related candidate genes in Xanthoceras fertilized ovules and comparisons between the qRT-PCR and RNA-Seq data analysis. The CCD7 gene was not expressed at detectable levels in the qRT-PCR analysis. The values are the means of three biological replicates with standard error bars. Asterisks indicate significant difference at p < 0.05 (ANOVA, Tukey’s HSD test).

Figure 8

Phylogenetic analysis of DWARF14 (D14) homologs. Maximum likelihood and Bayesian inference phylogenies were produced from an alignment of D14 hydrolases from Arabidopsis, rice, petunia, and Xanthoceras. A bootstrap analysis was performed using 1000 replications.

Figure 9

Quantitative real-time PCR (qRT-PCR) verification of expression of the selected SL perception- and signaling-related candidate genes in Xanthoceras fertilized ovules and comparisons between the qRT-PCR and RNA-seq data. The values are the means of three biological replicates with standard error bars. Asterisks indicate significant difference at p < 0.05 (ANOVA, Tukey’s HSD test).

Figure 10

Quantitative real-time PCR (qRT-PCR) verification of expression of the selected invertase candidate genes in Xanthoceras fertilized ovules and comparisons between the qRT-PCR and RNA-seq data. The values are the means of three biological replicates with standard error bars. Asterisks indicate significant difference at p < 0.05 (ANOVA, Tukey’s HSD test).

Figure 11

Expression levels of CCD8 and NCED3-2 genes in the Xanthoceras fertilized ovules with sucrose application and control treatment, respectively, relative to the housekeeping gene actin-2. The values are the means of three biological replicates with standard error bars. Asterisks indicate significant difference at p < 0.05 (ANOVA, Tukey’s HSD test).

Figure 12

Invertase activity and relative expression levels of CWIN2 gene in the Xanthoceras fertilized ovules with a synthetic strigolactone analog GR24 treatment and control treatment, respectively. The values are the means of three biological replicates with standard error bars. Asterisks indicate significant difference at p < 0.05 (ANOVA, Tukey’s HSD test).

Figure 13

Effect of VIGS-mediated XsD14 gene silencing on fruit set in terminal inflorescences of 4 Xanthoceras sorbifolium genotypes. The values are the means of five biological replicates with standard error bars. Different letters represent significant differences at p < 0.05 (ANOVA, Tukey’s HSD test).

Figure 14

Expression levels of XsD14 gene in the fertilized ovules with VIGS-mediated XsD14 gene silencing treatment and control treatment in relation to the expression levels of actin-2. The values are the means of three biological replicates with standard error bars. Different letters represent significant differences at p < 0.05 (ANOVA, Tukey’s HSD test).

Figure 15

Soluble sugar content in the fertilized ovules with VIGS-mediated XsD14 gene silencing treatment and control treatment. The values are the means of three biological replicates with standard error bars. Different letters represent significant differences at p < 0.05 (ANOVA, Tukey’s HSD test).