Oilseed crops are major sources of oils for human nutrition, and an increasing proportion is also being utilized for industrial purposes. Recent advances in our understanding of the basic biochemistry of seed oil biosynthesis, coupled with identification of genes for oilseed modification, have set the stage for the genetic engineering of oilseed crops that produce ‘designer’ plant seed oils tailored for specific applications. In this review we provide an overview of seed oil biosynthesis and highlight the enzymatic steps that have already been targeted for genetic manipulation, with the end goal of producing seed oils containing desired amounts of fatty acid components. Furthermore, we describe the identification of genes from various wild plant species that are capable of producing structurally diverse fatty acids, and how these advances open the door to the production of entirely novel oils in conventional oilseed crops. Transgenic oilseeds producing high amounts of these novel fatty acids represent renewable sources of raw materials that may compete with, and eventually replace, some petrochemicals that are derived from non-renewable crude oil.

The unique feature of the seed of tetraploid cotton (Gossypium hirsutum and Gossypium barbadense) is that about 30% of the seed coat epidermal cells develop into cellulose-enriched fibres, while the embryos synthesize oils and proteins. Hence, both the maternal and filial tissues of the cotton seed are of significant economic value. After initiation from the ovule epidermis at or just before anthesis, the single-celled fibres elongate to 2.5–6.0 cm long in the tetraploid species before they switch to intensive secondary cell wall cellulose synthesis. Thus, apart from its agronomic importance, the cotton fibre represents a model single-cell system to study the control of cell differentiation and elongation, carbon partitioning to cellulose synthesis and also the interaction between maternal (fibre) and embryonic tissues in seeds. Over the past decade or so, significant effort has been made to understand the cellular and molecular basis of cotton fibre development and oil biosynthesis in the embryo. Metabolic engineering of the oil biosynthetic pathway in cotton seed has successfully produced healthier and stable oils. A number of candidate genes and cellular processes that potentially regulate various aspects of fibre development have been identified. Further elucidation of the in vivo functions of those candidate genes could significantly deepen our understanding of fibre development and offer potential for improvement of fibre quality through genetic engineering or marker-assisted breeding approaches.

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 of Arabidopsis thaliana and the Solanaceae model 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.

Although jasmonates (JAs) are involved in germination and seedling development, the regulatory mechanism of JAs, and their relation with endogenous level modifications in these processes, is not well understood. We report here the detection of 12-oxo-phytodienoic acid (OPDA), jasmonic acid (JA), 11-hydroxyjasmonate (11-OH-JA), 12-hydroxyjasmonate (12-OH-JA) and methyljasmonate (JAME) in unimbibed seeds and seedlings of tomato Lycopersicon esculentum Mill cv. Moneymaker (wild type) and tss1, tss2, tos1 mutants. The main compounds in wild-type and tss1, tss2, tos1 seeds were the hydroxylate-JAs; 12-OH-JA was the major component in dry seeds of the wild type and in tss2 and tos1. The amounts of these derivatives were higher in seeds than in seedlings. Changes in JAs during wild-type and tss1 imbibition were analysed in seeds and the imbibition water. In wild-type imbibed seeds, 11-OH-JA content was higher than in tss1. 12-OH-JA showed a different tendency with respect to 11-OH-JA, with high levels in the wild type at early imbibition. In tss1, levels of 12-OH-JA rose from 24 to 48 h of imbibition. At 72 h of imbibition, when radicles had emerged, the amounts of both hydroxylates in wild-type and tss1 seeds were minimal. An important release of the hydroxylate forms was observed in the imbibition water. 11-OH-JA decreased in the imbibition water of wild-type seeds at 48 h. On the contrary, a high and sustained liberation of this compound was observed in tss1 after 24 h. 12-OH-JA increased in wild-type as well in tss1 until 24 h. Thereafter, a substantial reduction in the content of this compound was registered. NaCl-treated wild-type seedlings increased their 12-OH-JA, but tss1 seedlings increased their JA in response to salt treatment. In tss2 seedlings, NaCl caused a slight decrease in 11-OH-JA and JAME, whereas tos1 seedlings showed a dramatic OPDA and 12-OH-JA decrease in response to salt treatment. Under salt stress the mutant seedlings showed different patterns of JAs according to their differential hypersensitivity to abiotic stress. The JA-hydroxylate forms found, and the differential accumulation of JAs during germination, imbibition and seedling development, as well as their response to NaCl stress, provide new evidence about the control of many developmental processes by JA.

MicroRNAs (miRNAs) play critical roles in the development of animals and plants. Characterizing the stage- and tissue-specific expression of miRNAs that potentially regulate target transcription factor expression is becoming more important for understanding the regulatory mechanisms of critical events during plant development. A simple method for purifying small RNAs from seeds is described, as well as an efficient non-radioactive labelling system for making miRNA probes. In Arabidopsis thaliana seed extracts, low molecular-weight (LMW) RNAs (e.g. 5S rRNA, tRNA and miRNA) were separated from high molecular-weight (HMW) nucleic acids (e.g. 28S and 18S rRNA, mRNA and genomic DNA) by fractionation using isopropanol. HMW RNAs precipitated in 20% isopropanol, while most LMW RNAs remained in the supernatant. The purified LMW RNAs were used successfully for RNA gel blotting to detect miRNAs expressed in Arabidopsis and tomato (Lycopersicon esculentum) seeds. To increase the detection sensitivity of the microRNA probes, additional digoxigenin-labelled uridine triphosphates (UTPs) were incorporated into the miRNA probes by designing template oligo DNAs with three extra adenines (A) at each end of their sequence. These DNA oligomers were used to make double-stranded DNA templates for miRNA probe synthesis. This probe (termed AAAPLUS) exhibited stronger signals than normal probes. A technique was also developed to quickly screen expressed miRNAs in seeds using a miniblot system, which enabled simultaneous examination with multiple miRNA probes. This method provides a simple alternative to microRNA microarrays to identify the major miRNAs expressed in seeds.

Crop seeds accumulate soluble carbohydrates as part of their maturation process. In legume seeds, the major soluble carbohydrates are sucrose and its galactosides raffinose, stachyose and verbascose. In buckwheat (Fagopyrum esculentum Moench) seeds, the major soluble carbohydrates are sucrose and galactosides of d-chiro-inositol, named fagopyritols. This study was conducted to determine changes in soluble carbohydrate accumulation in embryos of buckwheat seeds after feeding solutions containing the free cyclitols D-chiro-inositol, myo-inositol and d-pinitol to stem–leaf–seed explants. Feeding D-chiro-inositol to explants resulted in a fourfold to fivefold increase in the accumulation of free d-chiro-inositol, fagopyritol A1 and fagopyritol B1 in embryos of mature seeds, but resulted in 30% less embryo dry weight compared to the control treatment without cyclitols. Feeding myo-inositol to buckwheat explants increased d-chiro-inositol in leaves and increased accumulation of fagopyritol A1 and fagopyritol B1 fivefold in embryos, fagopyritol A2 and fagopyritol B2 fourfold; fagopyritol A3 and fagopyritol B3 were also detected, with no reduction in accumulated embryo dry weight. Feeding d-pinitol to buckwheat explants resulted in accumulation of free d-pinitol in mature embryos, but not galactopinitols. D-Pinitol, galactopinitol A and galactopinitol B were not detected in embryos from explants fed solutions without d-pinitol. Feeding d-pinitol also resulted in reduced D-chiro-inositol accumulation by buckwheat seeds. The results indicate that myo-inositol may be the precursor to d-chiro-inositol synthesis, and fagopyritols accumulated in response to d-chiro-inositol availability in the embryo. We suggest that increasing myo-inositol in buckwheat maternal tissues may be an effective means to enhance the accumulation of D-chiro-inositol and fagopyritols in seeds, compounds that may be beneficial for the treatment of non-insulin-dependent diabetes mellitus.

Fire ephemerals are short-lived plants that primarily germinate after fire. Fresh and laboratory-stored seeds are difficult to germinate ex situ, even in response to fire-related cues such as heat and smoke. Seeds of eight Australian fire ephemeral species were buried in unburnt and recently burnt sites of natural bushland during autumn. Seeds were exhumed after 6 and 12 months and incubated in water and smoke water, either with or without a heat treatment at 70°C for 1 h. Generally, germination did not increase after 6 months of burial, but after 12 months of burial germination was enhanced in seven of the eight species. Actinotus leucocephalus produced higher germination following 12 months of burial without any further treatment, and smoke water and heat further improved germination. The four Gyrostemonaceae species, Codonocarpus cotinifolius, Gyrostemon racemiger, Gyrostemon ramulosus and Tersonia cyathiflora, only germinated in the presence of smoke water, and their germination was enhanced by burial. Burial improved germination in response to a heat treatment in Grevillea scapigera and Alyogyne huegelii seeds, but did not enhance Alyogyne hakeifolia germination. During concurrent dry laboratory storage of seeds at 15°C, only Actinotus leucocephalus produced increased germination in response to smoke water and heat over time. In summary, soil burial can alter the dormancy status of a number of Australian fire ephemeral seeds, rendering them more responsive to germination cues such as smoke water and heat. The requirement for a period of burial before seeds become responsive to smoke and/or heat would ensure that seeds persist in the soil until a subsequent fire, when there is an increase in nutrients available for growth and reduced competition from other plants.

This study addressed the difficulty of germinating drupelets (hereafter seeds) in the multi-seeded stony dispersal units (drupes) of Leptecophylla tameiameiae (Ericaceae). Embryos in fresh seeds were 77% the length of the endosperm, and seeds inside the intact drupes imbibed water. We monitored germination at 15/6, 20/10 and 25/15°C for 162 weeks, after which each drupe was cut open and ungerminated seeds counted. Drupes contained 1–6 seeds, and the total number of seeds in all treatments and controls was 1977, with 20, 29, 25, 18, 7 and <1% of them occurring in one-, two-, three-, four-, five- and six-seeded drupes, respectively. The percentage of seeds germinating in one-, two-, three-, four-, five- and six-seeded drupes was 74, 66, 65, 72, 56 and 0, respectively. Neither warm nor cold stratification for 6 or 12 weeks significantly increased germination percentages, compared to controls incubated continuously at 25/15°C for 162 weeks, where 72% of the seeds in the drupes germinated. At 25/15°C, 24–49 weeks were required for 20% of the seeds to germinate. Warm followed by cold stratification did not promote germination, and there was no widening of the temperature range for germination. Like seeds of other species known to have deep physiological dormancy (PD), those of L. tameiameiae required extended periods of time (16 to ≥162 weeks) to come out of dormancy and germinate, gibberellic acid (GA3) did not promote germination and excised embryos failed to grow. Thus, we conclude that seeds of L. tameiameiae have deep PD. However, unlike seeds of other species with deep PD, those of L. tameiameiae required an extensive period of warm rather than of cold stratification to come out of dormancy. It is suggested that a subtype a (seeds require a long period of cold stratification to come out of dormancy) and a subtype b (seeds require a long period of exposure to warm stratification to come out of dormancy) of deep PD be recognized in the Nikolaeva formula system for classifying seed dormancy.

Studies were conducted to determine if small embryos (i.e. low embryo length:seed length ratio) in mature dwarf seeds (0.2–2 mm) are underdeveloped. In this case, they would grow (inside the seed) prior to germination, and seeds would have morphological or morphophysiological dormancy. Prior to radicle emergence, embryo length in seeds of Drosera anglica (Droseraceae), Campanula americana, Lobelia appendiculata, L. spicata (Campanulaceae) and Sabatia angularis (Gentianaceae) increased 0, 103, 182, 83 and 57%, respectively. Since embryo growth did not occur in seeds of D. anglica prior to germination, embryos, although small, are fully developed; seeds have only physiological dormancy. The underdeveloped embryo in seeds of C. americana has little or no physiological dormancy; thus, seeds have morphological dormancy. On the other hand, underdeveloped embryos in seeds of L. appendiculata, L. spicata and S. angularis are physiologically dormant, and seeds have morphophysiological dormancy. Therefore, since small embryos in dwarf seeds may or may not be underdeveloped, assignment of seeds to a dormancy class requires that studies be done to determine if embryos grow inside the seed before germination can occur. Such information is important in understanding the evolutionary relationship of the different kinds of seed dormancy.

Seeds that store proteins in protein storage vacuoles are attractive bioreactors for producing and storing large amounts of pharmaceutical proteins. However, foreign proteins expressed in transgenic plants are subjected to the delivery and modification processes present within plant cells. Here, it is demonstrated that unique membrane sequences deliver a yellow fluorescent protein (YFP) to the seed protein storage vacuoles in transgenic tobacco (Nicotiana tabacum L.) plants, where the YFP is then separated from its membrane anchors. This precise targeting and separation is required for the successful delivery of useful proteins to seed protein storage vacuoles for their stable accumulation in transgenic crops.

In the paper ‘NO News’, Preston et al. (2004) make a number of erroneous assumptions regarding nitrogen oxide chemistry. These authors also present some very significant misinterpretations of previous research into the effects of various nitrogen oxides on germination of post-fire followers. Methodological differences between the study by Preston et al. (2004) and previous work are also problematic, such as using NO-donors in solution versus the use of direct application of various nitrogen oxides in the gaseous phase. A closer review of these studies, with the proper understanding of nitrogen oxide chemistry, and interpretations of the available literature, would lead to the conclusion that, contrary to the authors' assertions, the Preston et al. (2004) study supports, rather than refutes, earlier findings by Keeley and Fotheringham (1997, 1998a, b, 2000).

In their rebuttal, Fotheringham and Keeley (2005) (F&K, hereafter) assert that misinterpretations of previous research, errors in the presentation of the chemistry of nitrogen oxides and devious presentation of experimental results led to the conclusion of Preston et al. (2004). [These conclusions refute those of Keeley and Fotheringham's publication in Science (Keeley and Fotheringham, 1997).] We disagree and argue that the experimental evidence is consistent with the hypothesis that the ecologically relevant germination signals for the two post-fire annuals, Emmenanthe penduliflora and Nicotiana attenuata, are the specific pyrolysis products of cellulose rather than chemical scarification by nitrogen oxides (Keeley and Fotheringham, 1997).