Drought stress has transgenerational effects on seeds and seedlings in winter oilseed rape (Brassica napus L.)

Abstract

Background: Drought stress has a negative effect on both seed yield and seed quality in Brassica napus (oilseed rape, canola). Here we show that while drought impairs the maternal plant performance, it also increases the vigour of progeny of stressed maternal plants. We investigated the transgenerational influence of abiotic stress by detailed analysis of yield, seed quality, and seedling performance on a growth-related and metabolic level. Seeds of eight diverse winter oilseed rape genotypes were generated under well-watered and drought stress conditions under controlled-environment conditions in large plant containers.

Results: We found a decrease in seed quality in seeds derived from mother plants that were exposed to drought stress. At the same time, the seeds that developed under stress conditions showed higher seedling vigour compared to non-stressed controls.This effect on seed quality and seedling vigour was found to be independent of maternal plant yield performance.

Conclusions: Drought stress has a positive transgenerational effect on seedling vigour. Three potential causes for stress-induced improvement of seedling vigour are discussed: (1) Heterotic effects caused by a tendency towards a higher outcrossing rate in response to stress; (2) an altered reservoir of seed storage metabolites to which the seedling resorts during early growth, and (3) inter-generational stress memory, formed by stress-induced changes in the epigenome of the seedling.

Keywords: Amino acids; Canola; Drought stress; Fatty acids; Intergenerational stress memory; Metabolite analysis; Rapeseed; Seed germination; Seed quality; Seedling vigour.

Fig. 1

Effect of drought stress on (a) Total seed yield, (b) Thousand seed weight, (c) Number of seeds per silique, (d) Number of siliques per plant and (e) number of seeds per plant of 8 diverse winter oilseed rape genotypes cultivated in a semi-controlled container trial. Bars are means of three replicates with standard errors. Significant differences at p < 0.1′, p < 0.05* and p < 0.01**

Fig. 2

Effect of drought stress on (a) seed oil amount, (b) seed protein amount and (c) concentrations of seed glucosinolates of 8 diverse winter oilseed rape genotypes cultivated in a semi-controlled container trial. Bars are means of three replicates with standard errors. Significant differences at p < 0.05*, p < 0.01** and at p < 0.001***

Fig. 3

Effect of drought stress on the amount of (a) Oleic acid, (b) Linoleic acid and (c) Linolenic acid measured in the seeds of 8 diverse winter oilseed rape genotypes cultivated in a semi-controlled container trial. Bars are means of three replicates with standard errors. Significant differences at p < 0.1′, p < 0.05* and at p < 0.001***

Fig. 4

Effect of maternal drought stress on mean germination time (MGT) of seeds from 8 diverse winter oilseed rape genotypes, cultivated in a semi-controlled container trial: (a) germination performance under 0 MPa and (b) germination performance under moderate osmotic stress (− 0.5 MPa). Boxplots represent performances of 100 seeds each. Significant differences at p < 0.05* and at p < 0.001***

Fig. 5

Effect of maternal drought stress on seedling growth performance of seeds harvested from 8 diverse winter oilseed rape genotypes cultivated in a semi-controlled container trial. Bars are means of three replicates with standard errors. Significant differences at p < 0.1′, p < 0.05*, p < 0.01** and p < 0.001***

Fig. 6

Effect of maternal drought stress on seedling metabolite concentrations of the offspring derived from 8 diverse winter oilseed rape genotypes cultivated in a semi-controlled container trial. Bars are means of three replicates with standard errors. Significant differences at p < 0.1′, p < 0.05*, p < 0.01** and p < 0.001***. Scatter diagrams show correlations between the corresponding relative metabolite concentration (Rel. Conc.) and relative seedling fresh weight (Rel. FW) under drought stress

Fig. 7

Two-dimensional principal component analysis showing (a) the multivariate variation in terms of metabolite concentrations among seeds and seedlings derived from 8 open-pollinated winter oilseed rape genotypes cultivated under control and drought stress conditions in a semi-controlled container trial. All biological replicates are shown. Seedlings from maternal control conditions are shown in blue, seedlings from maternal stress conditions are shown in red. Proximity between samples indicates similarity in metabolite patterns. The axes show the first and second principal component along with the variation explained by them in brackets. (b) The same data set now represented as variable vectors indicating the strength and direction of each seed component and seedling metabolite. Vectors which are perpendicular to each other show independence of the component, vectors pointing in opposite directions show a negative correlation, while vectors with a small angle between them show a positive correlation. The color denotes the weight of the contribution to the total variation in the data set, with higher values meaning a stronger contribution. Amino acids are abbreviated as 3 letter IUPAC code. Fatty acids are named without the extension “acid”. TSW: Thousand seed weight, FW: Fresh weight

Fig. 8

Principal scheme of fatty acid biosynthesis in chloroplasts (green body) and endoplasmatic reticulum (blue body) of rapeseed. Arrows indicate, whether relative amounts of fatty acids have increased or decreased under drought stress in three or more of the 8 observed winter oilseed rape genotypes. FAS: Fatty acid synthase, ACC: Acetyl-CoA-Carboxylase, Δ9D: Δ9-Desaturase, Δ12D: Δ12-Desaturase, Δ15D: Δ15-Desaturase, E: Elongase

Fig. 9

Principal scheme of the biosynthesis and derivation of different amino acid and nitrogen compounds in rapeseed. Arrows indicate, whether concentrations of metabolites have increased or decreased under drought stress in three or more of the 8 observed winter oilseed rape genotypes