A Sleep Like Death: Identification of Genes Related to Seed Longevity in Medicago truncatula and Arabidopsis.

Atmaturation, seeds generally becomedormant and desiccation tolerant; after dormancy breaks, the seeds remain quiescent for an indefinite period and will germinate only when providedwith adequatemoisture. Some seeds remain in this quiescent state for thousands of years, only to germinate under the right conditions, while other seeds fail to germinate after extendedstorage.Howdosomeseedssurvive for so long? This question has crucial importance for agriculture and for conservation of genetic resources. Environmental conditions strongly affect seed longevity, as high temperature, high humidity, and high oxygen can decrease the longevity of seeds. Many innate factors also affect seed longevity; forexample,seedsaccumulate factors thatprotect against damage from desiccation, suchasheatshockproteins, sugars, and lateembryogenesisabundantproteins.Seeds alsoproduce factors thathelpmaintaingenome integrity and mitigate the effects of oxidative stress (reviewed inWaterworthetal., 2015).To examine the establishment of longevity during maturation, Righetti et al. (2015) characterized 104Medicago truncatula transcriptomes collected under five different conditions tomeasure transcript levels of all genes expressed in during seed maturation. They constructed a coexpression network, MatNet (see figure, panelA),whichhas2912nodes,73,6290edges, and includesmany seed-specific genesand transcriptional regulators (see figure, panel B). Based on their comprehensive profiling of seed physiology during maturation under different conditions, they then used a trait-based measure of significance to determine the relationship of the expressed genes to longevity anddesiccationtolerance.Byidentifyinggenes that showed both strong correlation to the trait and strong coexpression, the authors defined key modules affecting longevity and desiccation tolerance (see figure, panel C). To facilitate functional validation of the longevity module and examine its conservation, the authors next comparedM. truncatula seed maturation toArabidopsis thaliana.Out of 130Arabidopsis genes homologous to members of the M. truncatula longevity module, 113 showed connections in the Arabidopsis interaction database, indicating conservation of 87% of the module nodes between both species. The longevity module nodes showed an overrepresentation of genes involved in defense responses, and the authors examined T-DNAknockoutmutantsofseveralsuchgenes. For example, thewrky3 and nf-x1-like1 (nfxl1) mutants, which affect two defense-related transcription factors, showed altered transcript levels at several of the longevitymodule nodes, as well as decreased seed longevity. The longevity module also showed overrepresentation of geneswith auxin response factor binding sites in their promoters, suggesting an important role for auxin in longevity. The authors tested the role of the central regulator ABSCISIC ACID INSENSITIVE3, which affects longevity and desiccation tolerance in both species. Examination of the transcriptomes ofM. truncatula abi3mutant seeds and ABI3overexpressing hairy root cultures indicated that nodes in the longevity module showed ABI3-dependent and -independent regulation. However, expression of WRKY3 and NFX11 did not change in the abi3mutants, indicating that they function independently of ABI3. Unlike the fairy tales, falling into a “sleep like death” happens as a normal part of seed development. However, failing to wake from this sleep has serious implications for conservation and agriculture. Identification of a conserved longevity module that includes many defense-related genes indicates evolutionary interplay between defense responses and longevity. Moreover, manipulation of this module might give farmers vigorous, long-lasting seeds— no Prince Charming required.