This review focuses mainly on eudicot seeds, and on the interactions between abscisic acid (ABA), gibberellins (GA), ethylene, brassinosteroids (BR), auxin and cytokinins in regulating the interconnected molecular processes that control dormancy release and germination. Signal transduction pathways, mediated by environmental and hormonal signals, regulate gene expression in seeds. Seed dormancy release and germination of species with coat dormancy is determined by the balance of forces between the growth potential of the embryo and the constraint exerted by the covering layers, e.g. testa and endosperm. Recent progress in the field of seed biology has been greatly aided by molecular approaches utilizing mutant and transgenic seeds ofArabidopsis thalianaand theSolanaceaemodel systems, tomato and tobacco, which are altered in hormone biology. ABA is a positive regulator of dormancy induction and most likely also maintenance, while it is a negative regulator of germination. GA releases dormancy, promotes germination and counteracts ABA effects. Ethylene and BR promote seed germination and also counteract ABA effects. We present an integrated view of the molecular genetics, physiology and biochemistry used to unravel how hormones control seed dormancy release and germination.

The regulation of water uptake of germinating tobacco (Nicotiana tabacum) seeds was studied spatially and temporally by in vivo 1 H-nuclear magnetic resonance (NMR) microimaging and 1 H-magic angle spinning NMR spectroscopy. These nondestructive state-of-the-art methods showed that water distribution in the water uptake phases II and III is inhomogeneous. The micropylar seed end is the major entry point of water. The micropylar endosperm and the radicle show the highest hydration. Germination of tobacco follows a distinct pattern of events: rupture of the testa is followed by rupture of the endosperm. Abscisic acid (ABA) specifically inhibits endosperm rupture and phase III water uptake, but does not alter the spatial and temporal pattern of phase I and II water uptake. Testa rupture was associated with an increase in water uptake due to initial embryo elongation, which was not inhibited by ABA. Overexpression of b-1,3-glucanase in the seed-covering layers of transgenic tobacco seeds did not alter the moisture sorption isotherms or the spatial pattern of water uptake during imbibition, but partially reverted the ABA inhibition of phase III water uptake and of endosperm rupture. In vivo 13 C-magic angle spinning NMR spectroscopy showed that seed oil mobilization is not inhibited by ABA. ABA therefore does not inhibit germination by preventing oil mobilization or by decreasing the water-holding capacity of the micropylar endosperm and the radicle. Our results support the proposal that different seed tissues and organs hydrate at different extents and that the micropylar endosperm region of tobacco acts as a water reservoir for the embryo.

The enzyme ␤-1,3-glucanase (␤Glu) was found to be strongly induced by ultraviolet (UV-B; 280-320 nm) radiation in primary leaves of French bean (Phaseolus vulgaris). This was demonstrated on the level of gene transcription, protein synthesis, and enzyme activity and was due to the expression of bean class I ␤Glu (␤Glu I). In contrast to other proteins of the family of pathogenesis-related proteins, the induction of ␤Glu I by UV correlated with the formation of photoreversible DNA damage, i.e. pyrimidine dimer formation. In conditions that allowed photorepair of this damage, ␤Glu I induction was blocked. Therefore, UV-induced DNA damage seems to constitute a primary signal in the pathway leading to the induction of the ␤Glu I gene(s). The induction was a local response because in partly irradiated leaves ␤Glu I was selectively found in leaf parts exposed to UV. Although short wavelength UV ( Ͻ 295 nm) was most efficient in ␤Glu I induction, longer wavelength UV ( Ͼ 295 nm) as present in natural radiation was still effective. In contrast to UV induction of ␤Glu I, the induction of flavonoids in bean leaves was optimally triggered by much more moderate fluences from the UV wavelength range no longer effective in ␤Glu I induction. UV induction of the flavonoid pathway shows no correlation with DNA damage and thus should be mediated via a different signal transduction pathway.Plants require sunlight for photosynthesis and thus are constantly also exposed to potentially damaging UV radiation that is present in sunlight. UV-C ( Ͻ 280 nm) is the band with the highest energy and most efficient in damaging DNA and proteins, but only wavelengths in the UV-B (280-320 nm) range greater than 290 nm reach the earth's surface; shorter wavelengths are completely absorbed by the stratospheric ozone layer. UV-C and UV-B cause DNA damage; this radiation induces the formation of pyrimidine dimers of which cyclobutane pyrimidine dimers (CPDs) constitute the major class (Taylor et al., 1997). Plants possess quite effective protection mechanisms against UV-induced DNA-damage. So, photolyase is an enzyme capable to repair CPDs after activation by violet light (photoreactivation; Strid et al., 1994;Taylor et al., 1997). Another important defense mechanism against UV is the production of UV-absorbing flavonoids and phenylpropanoid compounds in leaf epidermal cells in response to UV irradiation (Li et al., 1993; Beggs and Wellmann, 1994). Flavonoids are also induced by a range of other stimuli; including pathogen attack (Harborne and Williams, 2000).Pathogenesis-related (PR) proteins are implicated in plant defense and accumulate in response to pathogen attack or to treatment with other elicitors (Leubner-Metzger and Meins, 1999). Endo-␤-1,3-glucanases (␤Glu; EC 3.2.1.39) are assigned to the PR proteins, where they constitute the PR-2 family. ␤Glu are abundant proteins, widely distributed in seed plant species. Besides plant defense, they have been implicated in several physiological and developmental processes (Doblin et al., 2001;Leubner...

Endosperm rupture is the main germination-limiting process in members of the Asteraceae (e.g. lettuce (Lactuca sativa L.) ) and Solanaceae (e.g. tomato (Lycopersicon esculentum Mill.) and tobacco (Nicotiana tabacum L.). About four decades ago a 'hatching enzyme' was proposed to cause endosperm weakening (i.e. a decline in the mechanical resistance of the micropylar endosperm), which is likely to be essential for seeds to complete germination. Although, there are established model systems among Asteraceae and Solanaceae for endosperm weakening, its molecular mechanism(s) still remain(s) a mystery. No single 'hatching enzyme' or universal molecular mechanism has been demonstrated explicitly. For the time being, the provisional conclusion is that endosperm weakening is likely to be achieved by the collaborative or successive action of several distinct molecular mechanisms. The knowledge gained from these established model systems will be compared and discussed. However, consideration of their severe experimental limitations shows that there is an urgent need for novel model systems. Such an optimal system has been recently found within the Brassicaceae. In this emerging model system for endosperm weakening, a complete study of the process is possible on each experimental level, from the direct measurement of the weakening by 'puncture force' to molecular investigations (e.g. proteome and genome transcriptome analyses).