Fluctuation of Arabidopsis seed dormancy with relative humidity and temperature during dry storage

Abstract

The changes in germination potential of freshly harvested seeds of Arabidopsis thaliana stored in various combinations of temperature and relative humidity were investigated over 63 weeks of storage. Seeds of the wild type Col-0 and of two mutants displaying low and high levels of dormancy, cat2-1 and mtr4-1, respectively, were stored at harvest in 24 different environments including a combination of eight relative humidities, from 1 to 85%, and four temperatures (10, 15, 20, and 25 °C). These mutations did not influence behaviour of seeds during storage. Primary dormant seeds did not germinate in darkness at 25 °C but acquired the potential to germinate at this temperature within 7 weeks when stored in relative humidities close to 50% across all temperatures. Sorption isotherms and Arrhenius plots demonstrated that the seed moisture content of 0.06 g H2O/g dry weight was a critical value below which dormancy release was associated with reactions of negative activation energy and above which dormancy release increased with temperature. Longer storage times when relative humidity did not exceed 75-85% led to decreased germination at 25 °C, corresponding to the induction of secondary dormancy. Dormancy release and induction of secondary dormancy in the dry state were associated with induction or repression of key genes related to abscisic acid and gibberellins biosynthesis and signalling pathways. In high relative humidity, prolonged storage of seeds induced ageing and progressive loss of viability, but this was not related to the initial level of dormancy.

Keywords: After-ripening; Arabidopsis thaliana; ageing; germination; relative humidity; temperature..

Fig. 1.

Seed germination percentages at 25 °C in darkness of wild type Col-0 (A, B, C, D), mtr4-1 (E, F, G, H), and cat2-1 (I, J, K, L) after different durations of storage at 10 °C (A, E, I), 15 °C (B, F, J), 20 °C (C, G, K), and 25 °C (D, H, L) under RHs of 1% (black circles), 33% (diamonds), 45% (black squares), 56% (triangles), 75% (white circles), and 85% (white squares). Mean ± SD of three biological replicates for each measurement.

Fig. 2.

Water sorption isotherms measured at 10, 15, 20, and 25 °C of seeds of Col-0 (A), mtr4-1 (B), and cat2-1 (C). Arrows within graphs indicate limits of regions of water binding. Mean ± SD of 10 replicates of 10 seeds.

Fig. 3.

Germination after 7 d at 25 °C of Col-0 (A), mtr4-1 (C), and cat2-1 (E) seeds stored for 7 weeks at 10 °C (circles), 15 °C (diamonds), 20 °C (triangles), and 25 °C (squares) at various RHs (1, 33, 45, 56, 75, and 85%) which produced the indicated MC. Arrows in (A) indicate the optimal MC for dormancy release at each temperature. Arrhenius plots of seed dormancy release, where ln(germination), determined after 7 d at 25 °C, is plotted against the inverse of the temperature [Temperature^–1 (1000/K)] for Col-0 (B), mtr4-1 (D), and cat2-1 (F). Seed MC during after-ripening (which lasted for 7 weeks) and R ² values of the linear regression lines are indicated within the graph. (A colour version of this figure is available at JXB online.)

Fig. 4.

Germination after 7 d at 15 °C of Col-0, mtr4-1, and cat2-1 seeds stored for 63 weeks at 10 °C (A), 15 °C (B), 20 °C (C), and 25 °C (D) under different RHs. Mean ± SD of three biological replicates for each measurement.

Fig. 5.

Transcript abundance of NCED3, NCED6, NCED9, ABI5, DOG1, CYP707A2, Ga3ox1, Ga20ox4, and Ga2ox2 in Col-0 seeds, dry (white bars) or imbibed for 24h at 25 °C (black bars). Analysis was performed on dormant seeds after harvest (D1), non-dormant seeds after-ripened at 56% RH and 20 °C for 7 weeks (ND), and secondary dormant seeds stored for 63 weeks at 56% RH and 20 °C (D2). The relative expression was calculated from qRT-PCR data with two references genes, Ubiquitin 5 and AT4G12590, and expressed in arbitrary units with a value of 100 attributed to the dry dormant seeds after harvest. Means ± SD of three biological replicates are shown. a, b and c indicate homogeneous groups in a corresponding class (Anova test and Newman–Keuls tests, P=0.05).