Physical interactions between ABA response loci of Arabidopsis.

Genetic and physiological studies have shown that the Arabidopsis thaliana abscisic acid-insensitive (ABI) loci interact to regulate seed-specific and/or ABA-inducible gene expression. We have used the yeast two-hybrid assay to determine whether any of these genetic interactions reflect direct physical interactions. By this criterion, only ABI3 and ABI5 physically interact with each other, and ABI5 can form homodimers. The B1 domain of ABI3 is essential for this interaction; this is the first specific function ascribed to this domain of the ABI3/VP1 family. The ABI5 domains required for interaction with ABI3 include two conserved charged domains in the amino-terminal half of the protein. An additional conserved charged domain appears to have intrinsic transcription activation function in this assay. Yeast one-hybrid assays with a lacZ reporter gene under control of the late embryogenesis-abundant AtEm6 promoter show that only ABI5 binds directly to this promoter fragment.

Regulation and function of the Arabidopsis ABA-insensitive4 gene in seed and abscisic acid response signaling networks.

We have characterized developmental, environmental, and genetic regulation of abscisic acid-insensitive (ABI)4 gene expression in Arabidopsis. Although expressed most strongly in seeds, ABI4 transcripts are also present at low levels in vegetative tissue; vegetative expression is not induced by abscisic acid (ABA) or stress treatments. Comparison of transcript levels in mature seeds of ABA-insensitive, ABA-hypersensitive, ABA-deficient, or heterochronic mutants indicates that ABI4 expression is altered in only two of the backgrounds, the ABA-insensitive mutants abi1-1 and abi3-1. To determine whether ABI4 is necessary and/or sufficient for ABA response, we assayed the effects of loss of ABI4 function and ectopic ABI4 expression on growth and gene expression. We examined genetic interactions among three ABA response loci, ABI3, ABI4, and ABI5, by comparing phenotypes of mutants, ectopic expression lines, mutants carrying an ectopically expressed transgene, and the corresponding wild-type lines. Our results indicate some cross-regulation of expression among ABI3, ABI4, and ABI5 and suggest that they function in a combinatorial network, rather than a regulatory hierarchy, controlling seed development and ABA response.

Abscisic acid inhibition of radicle emergence but not seedling growth is suppressed by sugars.

Low concentrations of sugars altered the sensitivity of seed germination to inhibition by exogenous abscisic acid (ABA). Germination of wild-type and ABA-insensitive (abi) Arabidopsis seeds was assayed on media containing ABA and a variety of sugars and sugar alcohols. The inhibitory effects of ABA were strongly repressed in the presence of 15 to 90 mM glucose (Glc), sucrose, or fructose, but not by comparable concentrations of sorbitol or mannitol. Several features of the response to Glc are inconsistent with a purely nutritional effect: The optimal sugar concentration is low and differs between the wild type and the abi mutants. Furthermore, Glc suppression of ABA inhibition is light dependent and limited to the process of radicle emergence.

The Arabidopsis abscisic acid response gene ABI5 encodes a basic leucine zipper transcription factor.

The Arabidopsis abscisic acid (ABA)-insensitive abi5 mutants have pleiotropic defects in ABA response, including decreased sensitivity to ABA inhibition of germination and altered expression of some ABA-regulated genes. We isolated the ABI5 gene by using a positional cloning approach and found that it encodes a member of the basic leucine zipper transcription factor family. The previously characterized abi5-1 allele encodes a protein that lacks the DNA binding and dimerization domains required for ABI5 function. Analyses of ABI5 expression provide evidence for ABA regulation, cross-regulation by other ABI genes, and possibly autoregulation. Comparison of seed and ABA-inducible vegetative gene expression in wild-type and abi5-1 plants indicates that ABI5 regulates a subset of late embryogenesis-abundant genes during both developmental stages.

The Arabidopsis abscisic acid response locus ABI4 encodes an APETALA 2 domain protein.

Arabidopsis abscisic acid (ABA)-insensitive abi4 mutants have pleiotropic defects in seed development, including decreased sensitivity to ABA inhibition of germination and altered seed-specific gene expression. This phenotype is consistent with a role for ABI4 in regulating seed responses to ABA and/or seed-specific signals. We isolated the ABI4 gene by positional cloning and confirmed its identity by complementation analysis. The predicted protein product shows homology to a plant-specific family of transcriptional regulators characterized by a conserved DNA binding domain, the APETALA 2 domain. The single mutant allele identified has a single base pair deletion, resulting in a frameshift that should disrupt the C-terminal half of the protein but leave the presumed DNA binding domain intact. Expression analyses showed that despite the seed-specific nature of the mutant phenotype, ABI4 expression is not seed specific.

Synthesis of small heat-shock proteins is part of the developmental program of late seed maturation.

Small heat-shock proteins (sHSPs) accumulate in plants in response to high-temperature stress. Specific sHSPs, the cytosolic class I and class II proteins, are also expressed in the absence of stress in maturing seeds of several species, and a role for these proteins in desiccation tolerance, dormancy, or germination has been hypothesized. We demonstrate that class I sHSPs are expressed during Arabidopsis seed development in a pattern similar to that previously observed in other species: they are first detected during mid-maturation, are most abundant in dry seeds, and decline rapidly during germination. Although the class I sHSP family in Arabidopsis appears to consist of four genes, expression of a single gene, Athsp 17.4, accounts for the majority of sHSPs in maturing seeds. sHSP levels were also examined in seeds of several Arabidopsis mutants with reduced sensitivity to abscisic acid inhibition, including aba1, abi1, and abi2, abi3-1, abi3-6, abi4, and abi5-1. The abi3-1 mutant has 10-fold reduced levels of sHSPs; sHSPs are undetectable in the abi3-6 mutant. All other mutants were indistinguishable from wild type. These results suggest that sHSP expression in seeds is regulated by the ABI3 response pathway and wild-type levels of sHSPs are not sufficient for seed dormancy and not necessary for desiccation tolerance. However, roles in either process cannot be ruled out. In total the data indicate that the expression of sHSPs in seeds is part of the normal developmental program of late seed maturation and the presence of sHSPs has adaptive significance for plant reproduction.

Maternal Effects Govern Variable Dominance of Two Abscisic Acid Response Mutations in Arabidopsis thaliana.

Three abscisic acid (ABA)-controlled responses (seed dormancy, inhibition of germination by applied ABA, and stomatal closure) were compared in wild-type versus homo- and heterozygotes of two Arabidopsis thaliana ABA-insensitive mutants, abi1 and abi2. We found that sensitivity of seeds to applied ABA is partially maternally controlled but that seed dormancy is determined by the embryonic genotype. The effects of the abi1 and abi2 mutations on ABA sensitivity of seed germination ranged from recessive to nearly fully dominant, depending on the parental source of the mutant allele. This maternal effect disappeared during vegetative growth. Stomatal regulation in heterozygotes showed substantial variability, but the average water loss was intermediate between that of homozygous mutants and wild type.

Abscisic acid-insensitive mutations provide evidence for stage-specific signal pathways regulating expression of an Arabidopsis late embryogenesis-abundant (lea) gene.

An Arabidopsis homolog of the abscisic acid (ABA)-inducible cotton D19 and wheat Em genes was cloned and its expression assayed at two developmental stages in wild-type, ABA-deficient (aba) and three ABA-insensitive (abi) lines of Arabidopsis thaliana. Expression of this gene was reduced slightly in seeds of aba mutants and approximately ten-fold in abi3 mutants, but seed expression was not decreased in either abi1 or abi2 monogenic mutants. In contrast, the abi1 and abi2 mutants showed a very slight reduction of ABA inducibility in 8-day-old plants, while the responses of aba and abi3 mutants were comparable to that of wild type. Although previous studies have shown that none of the abi mutations show completely stage-specific effects, the results reported here indicate that the importance of each of the ABI loci in regulating this single gene is stage-dependent. Furthermore, the fact that none of the abi mutations show more than minor effects on exogenous ABA inducibility of the Arabidopsis D19/Em homolog in young plants suggests that an additional ABA signalling pathway may be operating during vegetative growth.

Three Classes of Abscisic Acid (ABA)-Insensitive Mutations of Arabidopsis Define Genes that Control Overlapping Subsets of ABA Responses.

Wild type and three abscisic acid (ABA)-insensitive mutants of Arabidopsis (ABI1, ABI2, and ABI3) were compared for their ability to respond to ABA for a variety of ABA-inducible responses throughout the life cycle of the plants. The responses tested included effects on seedling growth, proline accumulation in seedlings, ABA-regulated protein synthesis in plantlets, and seed storage protein and lipid synthesis and accumulation. The abi1 and abi2 mutants showed reduced sensitivity to ABA for inhibition of seedling growth, induction of proline accumulation, and alterations in protein synthesis patterns during vegetative growth, but had wild type levels of storage reserves. In contrast, the abi3 mutant had wild type sensitivity for induction of proline accumulation and was only slightly less responsive to ABA with respect to effects on seedling growth and changes in patterns of protein synthesis. The major effects of this mutation were on seed development. Seeds of the abi3 mutant had two-thirds of the wild type level of storage protein and one-third the wild type level of eicosenoic acid, the major fatty acid component of storage lipids in wild type seeds. These results show that none of the abi mutants is insensitive for all ABA-inducible responses and that the abi3 effects are not seed-specific. Comparison of the degree of ABA sensitivity of monogenic mutant lines with that of digenic mutant lines carrying pairwise combinations of the abi mutations suggests that ABA responses in mature seeds are controlled by at least two parallel pathways.

Rapeseed embryo development in culture on high osmoticum is similar to that in seeds.

The development of Brassica napus L. cv Tower embryos of different ages cultured in vitro with and without high osmoticum (0.48 and 0.69 molar sorbitol) was compared with normal development in situ to investigate the role of a drying environment in embryo maturation. Sensitivity to osmoticum was assayed in terms of its ability to mimic normal development, i.e. to both suppress germination and maintain 12 S storage protein (cruciferin) synthesis at levels comparable to those seen in the developing seed. The osmotic conditions used block germination of predesiccation stage embryos but were not sufficient to prevent desiccation stage embryos from taking up water and germinating. At all stages tested, the osmotically treated embryos had approximately normal levels of cruciferin mRNA. Measurements of endogenous abscisic acid (ABA) levels by radioimmunoassay indicated that the osmotic effects on germination and gene expression were not mediated by elevated embryonic ABA. Comparison of the kinetics of osmotic and ABA effects on gene expression showed that the osmotic effect is more rapid. These results are consistent with the hypothesis that ABA acts by inhibiting water uptake, which mechanically prevents germination and affects gene expression in some unknown manner.

Role of ABA in Maturation of Rapeseed Embryos.

Development of Brassica napus L. cv Tower embryos of different ages cultured in vitro with and without abscisic acid (ABA) was compared with normal development in situ to investigate the role of ABA in embryo maturation. Endogenous ABA levels were measured by radioimmunoassay, and sensitivity to ABA was assayed in terms of its ability to suppress precocious germination and stimulate accumulation of storage protein and storage protein mRNA. During development in situ, the levels of endogenous ABA and 12S storage protein mRNA both reach their peaks just before the embryos begin to desiccate. The ABA levels during this phase of development also correlate with the time required in culture before germination is evident. Following these peaks, increasing concentrations of exogenous ABA are required to both suppress germination and continue storage protein accumulation in vitro. Thus, both endogenous ABA and ABA sensitivity decline during maturation. The concentrations of exogenous ABA required to suppress germination at these later stages result in abnormally high levels of endogenous ABA and appear to be toxic. These results are consistent with the hypothesis that in maturing rapeseeds, low water content rather than ABA prevents germination during the later stages of development.

Nucleotide sequence of a cDNA clone of Brassica napus 12S storage protein shows homology with legumin from Pisum sativum.

The most abundant protein in seeds of Brassica napus (L.) is cruciferin, a legumin-like 12S storage protein. By in vitro translation of embryo RNA, and pulse-chase labelling of cultured embryos with (14)C-leucine, we have shown that the 30 kd α polypeptides and 20 kd ß polypeptides of cruciferin are synthesized as a family of 50 kd precursors which are cleaved post-translationally. One member of the cruciferin family was cloned from embryo cDNA and sequenced. The nucleotide sequence of the cruciferin cDNA clone, pC1, contains one long open reading frame, which originates in a hydrophobic signal peptide region. Therefore, the complete sequence of the cruciferin mRNA was obtained by primer extension of the cDNA. The predicted precursor polypeptide is 488 amino acids long, including the 22 amino acids of the putative signal sequence. The amino acid composition of cruciferin protein is very similar to the predicted composition of the precursor. Comparison with an amino acid sequence of legumin from peas, deduced from the nucleotide sequence of a genomic clone, shows that the α polypeptide precedes the ß polypeptide on the precursor. Cruciferin and legumin share 40% homology in the regions which can be aligned. However, cruciferin contains a 38 amino acid region high in glutamine and glycine in the middle of the α subunit, which is absent in legumin. Legumin has a highly charged region, 57 amino acids long, at the carboxyl-end of the α subunit, which is not found in cruciferin. Both of these regions appear to have originated by reiteration of sequences. re]19850513 ac]19850715.

Precociously germinating rapeseed embryos retain characteristics of embryogeny.

We compared the germination of Brassica napus L. embryos at three stages of development-mid-cotyledon, maturation and mature dry-to determine at which stage they acquired the capacity for normal germination and seedling development. Embryos were removed from the seed and cultured on hormone-free medium, allowing them to germinate. The transition from embryogeny to germination was monitored both morphologically and biochemically, using synthesis of 12 S storage protein as a marker of embryogeny. The mature embryos (dry seeds) set the standard for normal seedling development: radicle emergence, hypocotyl extension and cotyledon expansion occurred within 2 d and true leaves were formed within a week of germination. Rocket immunoelectrophoresis indicated that the storage proteins in seedlings from mature dry embryos were completely degraded within a week. In contrast, the midcotyledon-stage embryos appeared to germinate abnormally, retaining many embryonic characteristics. Although the roots emerged, the hypocotyls did not elongate and secondary cotyledons instead of leaves were formed at the shoot apex. Also, the seedlings continued to synthesize and accumulate storage proteins. The maturation-stage embryos did develop into normal-looking seedlings, but complete degradation of storage proteins required several weeks, presumably reflecting continued synthesis and turnover. We conclude that embryogenic and germination-specific processes can occur concurrently and that the capacity to develop as normal seedlings is acquired gradually during the maturation process.