Freshly harvested seeds of Arabidopsis thaliana, Columbia (Col) accession were dormant when imbibed at 25°C in the dark. Their dormancy was alleviated by continuous light during imbibition or by 5 weeks of storage at 20°C (after-ripening). We investigated the possible role of reactive oxygen species (ROS) in the regulation of Col seed dormancy. After 24 h of imbibition at 25°C, non-dormant seeds produced more ROS than dormant seeds, and their catalase activity was lower. In situ ROS localization revealed that germination was associated with an accumulation of superoxide and hydrogen peroxide in the radicle. ROS production was temporally and spatially regulated: ROS were first localized within the cytoplasm upon imbibition of non-dormant seeds, then in the nucleus and finally in the cell wall, which suggests that ROS play different roles during germination. Imbibition of dormant and non-dormant seeds in the presence of ROS scavengers or donors, which inhibited or stimulated germination, respectively, confirmed the role of ROS in germination. Freshly harvested seeds of the mutants defective in catalase (cat2-1) and vitamin E (vte1-1) did not display dormancy; however, seeds of the NADPH oxidase mutants (rbohD) were deeply dormant. Expression of a set of genes related to dormancy upon imbibition in the cat2-1 and vet1-1 seeds revealed that their non-dormant phenotype was probably not related to ABA or gibberellin metabolism, but suggested that ROS could trigger germination through gibberellin signaling activation.

Changes in catalase (CAT) activity and in CAT isoform pattern and expression were investigated in developing sunflower (Helianthus annuus L.) seeds during desiccation on the mother plant and after artificial drying on the flowerheads. Seeds regularly desiccated during their development on the mother plant and reached mass maturity at c. 42 d after flowering (DAF). Freshly harvested seeds did not germinate at any stage of development because they were dormant, but their dormancy was broken after 5–6 months of dry storage. Immature seeds were desiccation‐tolerant at 24 DAF since they were able to germinate fully after artificial drying on the flowerheads followed by dry storage. CAT activity increased in non‐dehydrated seeds during their development, reaching a maximum a little after seed mass maturity and after artificial drying in immature seeds. This stimulation of CAT activity by natural and artificial drying was related to changes in CAT isoform pattern. Of the four constitutive CAT subunits, that of 59 kDa was always present, but dehydration induced the synthesis of a 55 kDa subunit. This synthesis of the CAT 55 kDa subunit resulted from an activation of the CATA1 gene, suggesting that the regulation of catalase activity and synthesis by drying occurred at the transcriptional level. The increase in CAT activity induced by seed drying was associated with a decrease in hydrogen peroxide level and in lipid peroxidation. These results suggest that CAT plays a role during seed desiccation by preventing dehydration‐related oxidative damage and that H₂O₂ may play a role in the regulation of CAT gene expression and the transduction pathway of the dehydration signal.

Two mechanisms have been suggested as being responsible for dormancy in barley grain: (i) ABA in the embryo, and (ii) limitation of oxygen supply to the embryo by oxygen fixation as a result of the oxidation of phenolic compounds in the glumellae. The aim of the present work was to investigate whether hypoxia imposed by the glumellae interferes with ABA metabolism in the embryo, thus resulting in dormancy. In dormant and non-dormant grains incubated at 20 °C and in non-dormant grains incubated at 30 °C (i.e. when dormancy is not expressed), ABA content in the embryo decreased dramatically during the first 5 h of incubation before germination was detected. By contrast, germination of dormant grains was less than 2% within 48 h at 30 °C and embryo ABA content increased during the first hours of incubation and then remained 2–4 times higher than in embryos from grains in which dormancy was not expressed. Removal of the glumellae allowed germination of dormant grains at 30 °C and the embryos did not display the initial increase in ABA content. Incubation of de-hulled grains under 5% oxygen to mimic the effect of glumellae, restored the initial increase ABA in content and completely inhibited germination. Incubation of embryos isolated from dormant grains, in the presence of a wide range of ABA concentrations and under various oxygen tensions, revealed that hypoxia increased embryo sensitivity to ABA by 2-fold. This effect was more pronounced at 30 °C than at 20 °C. Furthermore, when embryos from dormant grains were incubated at 30 °C in the presence of 10 μM ABA, their endogenous ABA content remained constant after 48 h of incubation under air, while it increased dramatically in embryos incubated under hypoxia, indicating that the apparent increase in embryo ABA responsiveness induced by hypoxia was, in part, mediated by an inability of the embryo to inactivate ABA. Taken together these results suggest that hypoxia, either imposed artificially or by the glumellae, increases embryo sensitivity to ABA and interferes with ABA metabolism.

We have investigated the stomatal and phototropic responses to blue light of a number of single and double mutants at various loci that encode proteins involved in blue-light responses in Arabidopsis. The stomatal responses of light-grown mutant plants (cry1, cry2, nph1, nph3, nph4, cry1cry2, andnph1cry1) did not differ significantly from those of their wild-type counterparts. Second positive phototropic responses of etiolated mutant seedlings, cry1, cry2, cry1cry2, andnpq1-2, were also similar to those of their wild-type counterparts. Although npq1 and single and double cry1cry2 mutants showed somewhat reduced amplitude for first positive phototropism, threshold, peak, and saturation fluence values for first positive phototropic responses of etiolated seedlings did not differ from those of wild-type seedlings. Similar to the cry1cry2 double mutants and tonpq1-2, a phyAphyB mutant showed reduced curvature but no change in the position or shape of the fluence-response curve. By contrast, the phototropism mutantnph1-5 failed to show phototropic curvature under any of the irradiation conditions used in the present study. We conclude that the chromoproteins cry1, cry2, nph1, and the blue-light photoreceptor for the stomatal response are genetically separable. Moreover, these photoreceptors appear to activate separate signal transduction pathways.