Seed Dormancy and Seed Longevitywageningenseedlab.nl/thesis/tpnguyen/Thesis Thu-Phuong...expression of seed storage proteins (Gutierrez et al., 2007). Seeds of abi3 , lec1 , and fus3

Seed Dormancy and Seed Longevityfrom genetic variation to gene identification

Thu-Phuong Nguyen

ISBN: 978-90-393-6094-1

Printing: GVO drukkers & vormgevers B.V. (www.gvo.nl)

Layout and Cover design: agilecolor design studio/atelier (www.agilecolor.com)

Seed Dormancy and Seed Longevityfrom genetic variation to gene identification

Kiemrust en bewaarbaarheid van zaden:van genetische variatie naar gen identificatie

(met een samenvatting in het Nederlands)

Proefschrift

ter verkrijging van de graad van doctor aan de Universiteit Utrecht op gezag van de rector magnificus, prof. dr. G. J. van der Zwaan, ingevolge het besluit van het college voor promoties in het openbaar te verdedigen op woensdag 26 februari 2014 des

middags te 2.30 uur

door

Thu-Phuong Nguyengeboren op 27 november 1983 te Hanoi, Vietnam

Promotors:Prof. dr. J.C.M. Smeekens

Prof. dr. ir. H.J. Bouwmeester

Co-promotor:Dr. ing. L. Bentsink

This thesis was accomplished with financial support from the Dutch Technology Foundation STW, which is part of the Netherlands Organisation for Scientific

Research (NWO) and partly funded by the Ministry of Economic Affairs (project number STW 10328)

Table of Contents

Chapter 1General introduction

p.7

Chapter 2Natural variation for seed longevity and seed dormancy are negatively correlated in Arabidopsis thaliana

p.21

Chapter 3The identification and validation of factors involved in seed longevity, a genetics and proteomics approach

p.41

Chapter 4The genetic analyses and fine-mapping of two seed QTLs; Delay Of Germination2 and Germination Ability After Storage1

p.69

Chapter 5Physiological and genetic characterizationof novel seed dormancy mutants

p.87

Chapter 6General discussion

p.105

References p.115

Samenvatting p.133

Acknowledgements p.137

Curriculum Vitae p.140

General Introduction

1Chapter

general introductionChapter 1

8

Abstract

Higher plants produce seeds as a reproductive strategy to assure survival in the ecological system. A seed serves as a dispersal unit to a new location and nourishes the embryo for seedling establishment. Seeds are able to identify favorable conditions for germination by using different mechanisms based on the environmental information as experienced by the mother plant. The most prominent mechanism is seed dormancy, which allows seeds to postpone germination to conditions that support seedling formation and plant growth. After dispersal, seeds can remain viable in the soil seed bank for very long periods and are able to germinate under favorable conditions, a strategy named seed longevity. Seed dormancy and seed longevity are adaptive traits for which genetic variation is present in nature and identifying the mechanisms involved is important to understand how nature selects for these traits. Both traits are of tremendous agronomical importance and are key traits for crop improvement. For example pre-harvest sprouting is most detrimental and must be suppressed in crops such as barley and sweet pepper. Moreover, controlling seed longevity is important for biodiversity conservation especially for seed plants. In this introduction, an overview of the various aspects related to seed development, seed dormancy and seed longevity is presented for the model plant Arabidopsis thaliana.


9

Arabidopsis as a model plant

Arabidopsis thaliana is a member of Brassicaceae family, and has become a model for plant biology since Arabidopsis possesses crucial advantages to address fundamental questions. Arabidopsis is a small self-pollinator, that completes its life cycle in approximately six to eight weeks. Arabidopsis has a relatively small genome of ca. 130 Mb divided over five chromosomes and has been sequenced and annotated (Weigel and Mott, 2009). Gene functions can be assessed in publicly available T-DNA knock-out mutants (http://arabidopsis.org). Furthermore, introducing genes and gene silencing are easily obtained by standardized transformation protocols (Weigel and Glazebrook, 2006).

Arabidopsis has a world-wide distribution, from the native continental Eurasia and North Africa to the rest of the world, especially in the Northern hemisphere (Hoffmann, 2002). This broad geographic distribution harbors substantial and quantitative phenotypic variation that reflects the genetic variation needed for adaptation to specific environmental conditions (Alonso-Blanco and Koornneef, 2000). This allows the generation of advanced genetic material to identify genetic factors underlying quantitative traits using quantitative trait loci (QTL) mapping. The prerequisite of QTL mapping is a segregating population in combination with a genetic map preferably based on molecular markers. The segregating population is phenotyped for the trait of interest followed by association mapping between genetic markers and phenotypes using advanced statistical methods. This results in an estimation of the QTL position and its effect on trait variation (Doerge, 2002). The preferred type of mapping population is a recombinant inbred line (RIL) population due to homozygousity and genetic stability that allows simultaneous analyses on multiple replicates per genotype. The phenotypic and genetic characterization of an individual QTL is best accomplished by the mendelization of a QTL (Alonso-Blanco and Koornneef, 2000), which preferably makes use of near isogenic lines (NILs). NILs carry an introgression of a genomic fragment from another accession, spanning a few cM around the QTL of interest. NILs allow further fine-mapping and map-based cloning of the locus. Induced genetic variation by mutagenesis is widely used in forward genetics. Mutagenesis is mostly performed by ethyl methane sulphonate (EMS) treatment or, to a lesser extent, by fast neutron radiation. Recently the identification of mutated genes has been accelerated by mapping by sequencing, the so called SHOREmap method (Schneeberger et al., 2009; James et al., 2013).

The ease of handling and the advanced tools and technology makes Arabidopsis a powerful ‘trendsetter’ in plant biology (Page and Grossniklaus, 2002).


10

Seed development and maturation

Seed developmentArabidopsis seeds are produced in a silique, that normally contains approximately 50 seeds. A fully developed Arabidopsis seed contains a zygotic embryo surrounded by a single cell endosperm layer. Both embryo and endosperm are formed after the fertilization of the embryo sac, and are covered by integuments that later become the seed coat (testa). Seed development in Arabidopsis takes about 20 days depending on the growing conditions. Seed development can be divided into two major phases: embryogenesis and seed maturation. Embryogenesis starts when cell division takes place following fertilization from a single-cell zygote until the heart stage embryo structure is reached (Mayer et al., 1991). Next, the embryo grows and fills the seed sac (Goldberg et al., 1994), followed by cell division arrest of the embryo (Raz et al., 2001). At this stage the seed contains the full size embryo and enters the maturation phase where food reserves accumulate and seed characteristics such as seed dormancy, seed longevity, and seed desiccation tolerance are induced. Also in the late maturation phase, the breakdown of chlorophyll is observed. At the same time, flavonoid pigments, accumulate in the testa giving the mature Arabidopsis seeds a brown color.

Seed maturationSeed maturation is genetically controlled by four major regulators, Abscisic Acid Insensitive3 (ABI3), Leafy Cotyledon1 (LEC1), LEC2, and Fusca3 (FUS3) (Raz et al., 2001). These four factors interact in a network to control various aspects of seed maturation. LEC1 and LEC2 positively regulate ABI3 and FUS3; ABI3 and FUS3 positively regulate themselves and each other, and form feedback loops essential for their sustained and uniform expression in the embryo (Kroj et al., 2003; Kagaya et al., 2005; To et al., 2006). The abi3, lec1, lec2, and fus3 mutants are severely affected in seed maturation and share several phenotypes, for example reduced seed dormancy (Raz et al., 2001) and reduced expression of seed storage proteins (Gutierrez et al., 2007). Seeds of abi3, lec1, and fus3 are intolerant to desiccation and storage (Ooms et al., 1993; Clerkx et al., 2004a; Tiedemann et al., 2008; Sugliani et al., 2009).

Seed maturation coincides with the accumulation of food reserves that are mobilized upon seed germination. Lipids in the form of triacylglycerols accumulate in cytosolic oil bodies that can occupy up to 30% of the mature seed volume (Browse and Somerville, 1994). Seed storage proteins serve as nitrogen source and are abundantly present as 12S globulin and 2S albumin and are similar in size and subunit composition as the cruciferin and napin seed storage proteins of Brassica, respectively (Heath et al., 1986). The gene family encoding the 12S proteins contains four members (Pang et al., 1988) while the 2S proteins encoding family has five members (Krebbers et al., 1988; Van der Klei et al., 1993). Starch, glucose, and fructose are also storage compounds, however,


11

very low amounts remain in the dry seeds (Focks and Benning, 1998). Sucrose and its galactose conjugates, the raffinose series oligosaccharides, accumulate at the end of seed maturation (Ooms et al., 1993). Phytate is a major form of organic phosphorus and is stored during seed development (Quick et al., 1997).

Seed desiccationSeed dormancy, longevity, and desiccation tolerance are established during seed maturation. These traits appear linked and possibly share common mechanisms. However, trait specific genetic factors exist. The induction of desiccation is, physiologically, the most visible trait since it is indicated by a reduction in water content to approximately 7% in dry mature seeds. This feature enables the mature seed to be stored in dry conditions and resume metabolic activity upon hydration. Several factors are involved in the acquisition of desiccation tolerance. The phytohormone abscisic acid (ABA) has an essential role, however, ABA alone was not sufficient (Farnsworth, 2000). ABA and stresses can trigger the accumulation of late embryo abundant (LEA) and dehydrin (Responsive to ABA; RAB) proteins (Hoekstra et al., 2001). These proteins were shown to be involved in the induction and re-establishment of desiccation tolerance (Maia et al., 2011; Verdier et al., 2013). The small heat shock protein 17.6 (sHSP17.6) in the cytosol appears to respond to specific developmental signals associated with the acquisition of desiccation tolerance (Wehmeyer and Vierling, 2000), and correlated positively with seed performance (Bettey and Finch-Savage, 1998).

Desiccation results in reduced cellular volumes and causes the compaction of cytoplasmic components. Such compaction increases molecular interactions and results in protein denaturation and membrane fusion (Hoekstra et al., 2001). Moreover, desiccation induces membrane damage and thereby cytoplasmic leakage, that can be measured by an increased conductivity upon hydration of desiccated tissues (Leprince and Hoekstra, 1998). Desiccation tolerance involves the interplay of several mechanisms simultaneously (Hoekstra et al., 2001). During moderate desiccation, protection of proteins and membranes by water replacement is involved. Water molecules are replaced by sugars at the hydrogen bonding sites to preserve the native structure of proteins and the spacing between the phospholipids, in which LEA and sHSP17.6 can act as molecular chaperones (Hoekstra et al., 2001). Further dehydration establishes a biological glass, which resembles a solid material but retains the physical properties of a liquid. Oligosaccharides, such as raffinose and stachyose, are more effective in forming glasses than sucrose and monosaccharides that have more plasticizing effects (Hoekstra et al., 2001). The solid-like glassy matrix in dry seeds is supported by the absence of monosaccharides, however, Ooms et al. (1993) suggested that the ratio between oligo and mono-saccharides is more important for desiccation tolerance than the absolute abundance. LEA proteins play a role in stabilizing dense glasses thus stabilizing cellular membranes. In such a glassy matrix, the high viscosity decreases the molecular mobility


12

and prevents diffusion within the cytoplasm. As a result, deleterious reactions and changes in structure and chemical compositions are slowed down during aging (Buitink et al., 2000).

Seed dormancy

Seed dormancy and germinationSeed dormancy is defined as a temporary failure of an intact, viable seed to complete germination under conditions that favor germination (Bewley, 1997). Seed dormancy assessment is based on seed germination, which is the result of the balance between the degree of embryo dormancy and the embryo growth potential to overcome dormancy imposed by structures surrounding the embryo, i.e. the testa and the endosperm. Seed germination incorporates events that commence with the uptake of water by a dry seed and terminates with the elongation of the embryonic axis, which is visible as radicle protrusion. Seed dormancy is a complex trait to analyze, since it is affected by various factors affecting either dormancy or germination. These factors operate during seed maturation and seed germination, and in maternal testa and zygotic embryo plus endosperm tissues. Maternal effects are inherited from the mother plant, and include maternal tissues surrounding the embryo (testa), but also factors that are delivered to the seed by the mother plant. Analyses on F1 seeds derived from reciprocal crosses, where parental genotypes are used as both male and female parent, can imply maternal inheritance. Although reciprocal differences might also be caused by the cytoplasmic inheritance of the trait, genetic segregation in the subsequent generation (F2 seeds) allows the distinction between cytoplasmic and maternal inheritance (Léon-Kloosterziel et al., 1994).

The induction and release of seed dormancyTwo well-known plant hormones that regulate seed dormancy are ABA and gibberellins (GA). A crucial role of ABA in inducing and maintaining seed dormancy and inhibiting seed germination has been extensively reviewed (Kucera et al., 2005; Finch-Savage and Leubner-Metzger, 2006; Holdsworth et al., 2008; Graeber et al., 2012). GA counteracts the inhibitory ABA effects by releasing dormancy and promoting germination. The dormant state is characterized by an intrinsic balance of ABA and GA biosynthesis and catabolism, instead of the absolute amount of these two hormones (Finch-Savage and Leubner-Metzger, 2006).

In seeds, ABA levels peak at the middle of development (Karssen et al., 1983). During this period, ABA deficiency or ABA over-accumulation is associated with either absence or enhancement of seed dormancy (Finkelstein et al., 2002; Nambara and Marion-Poll, 2003; Kushiro et al., 2004). ABA produced by embryo and endosperm can


13

impose dormancy, but not maternal ABA or exogenously applied ABA; such maternal and applied ABA do affect seedling establishment (Kucera et al., 2005). De-novo ABA biosynthesis of imbibed dormant seeds has been interpreted as a mechanism for dormancy maintenance (Ali-Rachedi et al., 2004). Dormancy release and germination are characterized by a shift to a low ABA:GA ratio resulting from increased GA biosynthesis and ABA degradation (Ali-Rachedi et al., 2004; Cadman et al., 2006). However, dormancy release in the embryo is not regulated by GA. GA is proposed to stimulate the elongation of embryo cells that can overcome coat restrictions and induce endosperm weakening. In addition, a decreased ABA and an increased GA sensitivity are also involved in the transition from dormant to non-dormant state (Finch-Savage and Leubner-Metzger, 2006).

Other hormones that affect seed dormancy release and seed germination are also reviewed by Kucera et al. (2005). Ethylene counteracts ABA effects and promotes seed germination through the promotion of radial cell expansion in the embryonic hypocotyl, increased seed respiration and water potential. Brassinosteroids (BR) act in parallel with GA to promote cell elongation and germination and are antagonistic to the inhibitory ABA action. BR can stimulate ethylene production and the application of ethylene can rescue ga1, a GA deficient mutant. Cytokinin and ethylene are linked because cytokinin resistant1 (ckr1) is allelic to the ethylene insensitive2 (ein2) mutant. Strigolactones and karrikins play a role in seed dormancy and germination. Karrikin Insensitive1 (KAI1) is allelic to More Axillary Branches2 (MAX2), the kai1/max2 mutant has an increased seed dormancy (Nelson et al., 2011). The above shows the complexity of hormone interactions in the regulation of seed dormancy and germination.

Other factors such as light, cold treatment, GA and nitrate (KNO3) present during seed imbibition can promote germination and alleviate seed dormancy (Derkx and Karssen, 1993b). Seeds that have released dormancy can enter dormancy again (secondary dormancy) when exposed for some time to unfavorable germination conditions, e.g. seed imbibition at relatively high temperature in darkness (Cone and Spruit, 1983; Derkx and Karssen, 1993a). Transcriptome analyses have shown that the mechanism underlying secondary dormancy resembles that of primary dormancy (Cadman et al., 2006).

Seed dormancy measurementSeed dormancy can be overcome by seed dry storage, regarded as after-ripening (AR). During AR, seeds gain germination ability, thus fully AR seeds are able to germinate. Seed dormancy is measured as germination percentage during AR, or expressed by days of seed dry storage required for 50% germination (DSDS50).

Genetic variation of seed dormancy mutantsSeveral genes underlying seed dormancy have been identified by mutants that exhibit


14

altered seed dormancy levels. Hormone related mutants were isolated based on screens for germination in the presence of hormones or hormone synthesis inhibitors. ABA related mutants are ABA deficient (aba) and ABA insensitive (abi) mutants, that display a markedly reduced seed dormancy. The absence of ABA-induced dormancy allows seeds to germinate without GA and mutants that germinate in the presence of the GA biosynthesis inhibitors paclobutrazol or tetcyclacis, resulted in the isolation of the ABA biosynthesis mutants (aba mutants) (Léon-Kloosterziel et al., 1996a). Defective ABA signaling also leads to changes in germination behavior, thus a screen for ABA signaling mutants, abi, is based on germination on ABA concentrations that are able to inhibit germination of wild type (Koornneef et al., 1984). In contrast to ABA related mutants, GA deficient mutants are unable to germinate without exogenous GA, thereby exhibiting enhanced seed dormancy (Koornneef and Van der Veen, 1980). The GA signal transduction mutant, gai1 (GA-insensitive1), has a reduced sensitivity to GA but does not have a strongly reduced seed germination (Koornneef et al., 1985). The ethylene receptor and signaling mutants, ethylene resistant1 (etr1) and ein2, result in a poor germination and a deeper dormancy but in contrast, constitutive triple response1 (ctr1) seeds germinate slightly faster (Bleecker et al., 1988; Leubner-Metzger et al., 1998; Beaudoin et al., 2000).

Moreover, seed germination is determined by the embryo, testa and endosperm that surround the embryo (Bewley, 1997). Therefore, also mutants with an affected seed coat showed reduced seed dormancy levels. Examples of these are the transparent testa (tt) and transparent testa glabra (ttg) mutants that are defective in flavonoid pigmentation and display seed colors ranging from yellow to pale brown (Koornneef, 1981, 1990; Debeaujon et al., 2000), and mutants altered in testa structure such as aberrant testa shape (ats) and apetala2 (ap2) (Jofuku et al., 1994; Léon-Kloosterziel et al., 1994). An exception is the glabra2 (gl2) mutant, that has a deformed testa surface structure (Bowman and Koornneef, 1994) and displays a slightly increased seed dormancy (Debeaujon et al., 2000).

In addition, a screen for mutants that display reduced seed dormancy led to the identification of the reduced dormancy (rdo) mutants, rdo1 to rdo4 (in the Landsberg erecta (Ler) genetic background; (Léon-Kloosterziel et al., 1996b; Peeters et al., 2002), and the delay of germination1 (dog1) mutant in the dormant NILDOG1-Cvi that carries the DOG1 Cape Verde Island (Cvi) allele (Bentsink et al., 2006). The cloning of RDO4, later named Histone Monoubiquitination1 (HUB1), suggests a role of chromatin remodeling in seed dormancy (Liu et al., 2007). RDO2 is allelic to the transcription elongation factor TFIIS, that interacts with RNA polymerase II associated factor 1 complex (PAF1C). Mutations in all the factors related to this PAF1C complex such as Vernalization Independence4 (VIP4), VIP5, Early Flowering7 (ELF7), ELF8 and Arabidopsis Trithorax-related7 (ATXR7) result in reduced seed dormancy (Liu et al., 2011).


15

Natural genetic variation for seed dormancy Additional regulators that control seed dormancy and germination were identified using natural variation. Eleven QTLs underlying seed dormancy variation in six RIL populations were reported, including the Delay Of Germination (DOG) series, (Alonso-Blanco et al., 2003; Bentsink et al., 2010). Of these, only DOG1, a major seed dormancy QTL, was cloned (Bentsink et al., 2006). Putative orthologous DOG1 genes are present in other Brassicaceae species, such as Lepidium sativum and Brassica rapa (Graeber et al., 2010). The DOG1 promotor regions in both Arabidopsis and B. rapa contain a RY repeat, required for ABI3/VIP1-mediated gene expression (Graeber et al., 2010). Possibly, DOG1 is regulated by a highly conserved dormancy factor (Graeber et al., 2010), however the molecular mechanism is yet unrevealed. Seed dormancy QTLs have also been identified in rice (Oryza sativa) (Lin et al., 1998; Gu et al., 2006). Seed Dormancy4 (SDR4) was cloned as a dormancy QTL in rice, but its putative homologues in Arabidopsis were not associated with seed dormancy (Sugimoto et al., 2010).

“Omics” analyses for seed dormancy and germinationMolecular and biochemical processes contributing to the control of seed dormancy and germination were studied by transcriptomics and proteomics studies. This revealed information about the timing and tissue specific gene expression (Gallardo et al., 2001; Masubelele et al., 2005; Nakabayashi et al., 2005; Dekkers et al., 2013), effects of environments (Cadman et al., 2006; Finch-Savage et al., 2007; Carrera et al., 2008; Arc et al., 2012), and functions of regulators (Gallardo et al., 2002; Rajjou et al., 2004; Cao et al., 2006; Carrera et al., 2007). It has been shown that dormant and AR seeds have distinct gene expression programs. Highly expressed genes in dormant (primary and secondary) states are mostly stress-related genes, thus ABA, stress and dormancy responses display significantly overlapping transcriptomes (Cadman et al., 2006). Carrera et al. (2008) demonstrated that AR acts as a developmental pathway distinguishable from dormancy in imbibed seeds, and that ABA is not a major regulator of AR. GA is required to overcome the constraints of the testa and about half of the GA-regulated genes are dependent on DELLA; thus DELLA independent pathways are also involved in the regulation of germination (Cao et al., 2006). Regulatory mechanisms for seed germination include epigenetic factors. Cytosine methylation was observed in large silent retrotransposon clusters in centromeric regions and also in non-centromeric silent gene clusters, and in cis-acting elements, RY motif and ABA-responsive elements (ABREs), that determine the pattern of stored mRNAs (Nakabayashi et al., 2005). Most of the germination-associated changes occur within the first 6 h of imbibition (Nakabayashi et al., 2005; Dekkers et al., 2013).

Proteomics studies have identified abundant changes of many proteins in both dry seed during storage and following imbibition, but also demonstrated the importance of post-translational modifications (PTMs) for the expression of the proteome (Gallardo et


16

al., 2001; Gallardo et al., 2002). In imbibed dormant seeds, enzymes involved in several energetically costly processes associated to germination are repressed, and dormancy maintenance is an ABA-dependent recapitulation of the late maturation program (Arc et al., 2012). The essential role of translation in germination was shown by Rajjou et al. (2004) as cycloheximide, a translation inhibitor, blocked germination. Chibani et al. (2006) showed that the lack of germination of dormant seeds was not due to the lack of de novo protein synthesis because the levels of [35S] methionine incorporation during imbibition were similar in dormant and non-dormant states. In agreement with transcriptome analyses, proteomics also reveals that ABA treatment in germination act in a separate pathway to that of dormancy (Chibani et al., 2006; Carrera et al., 2008).

Seed longevity

Seed longevity and germinationSeeds gradually lose germination ability when stored in dry laboratory conditions. Consequently, low germination percentages are obtained. As for seed dormancy, there is no direct assay to measure seed longevity, and the most used one is through seed germination measurements, hence also making seed longevity difficult to study. Seed longevity is defined as the germination ability of seeds after long-term storage. Studies that have been conducted to analyze seed longevity (storability) mainly investigated the mechanisms underlying seed deterioration, aging and seed vigor loss. The germination percentage of long-term stored seeds is a result of a combination of growth machinery protection, maintenance, and repair.

The role of antioxidant systems in seed dry storageReactive oxygen species (ROS) production is the most detrimental aging reaction during seed dry storage. Both stressed and unstressed seed cells generate ROS in various cell compartments, typically in peroxisomes, mitochondria, and cytosol. The accumulation of ROS leads to mitochondrial dysfunction, enzyme inactivation, membrane perturbation, and oxidation of lipids, proteins, and genetic material (DNA and RNA) (Moller et al., 2007). The ability of seeds to cope with such stresses is determined by the efficiency of the protecting mechanisms to avoid, scavenge, and neutralize reactive molecules. Oxidative stresses leading to an unstable redox homeostasis can occur due to an imbalance in pro-oxidant and antioxidant levels. Dry mature seeds are equipped with antioxidant systems that consist of enzymes and non-enzymatic antioxidants. Superoxide radicals are converted to hydrogen peroxide (H2O2) by superoxide dismutase (SOD) (Grene, 2002; Bailly et al., 2008). H2O2 is then neutralized by catalase (CAT) and ascorbate peroxididase (APX). CAT turns H2O2 directly into water and oxygen while APX catalyses the reaction between ascorbic acid (vitamin C) and H2O2 to form dehydroascorbate


17

and water (Blokhina et al., 2003). Removal of H2O2 by APX requires the involvement of glutathione (GSH) as a part of the ascorbate-glutathione cycle (Nocter and Foyer, 1998). Metallothioneins (MTs) play a role in the scavenging of ROS. Zhou et al. (2012) showed that NnMT2a, NnMT2b and NnMT3 were highly expressed in developing and germinating sacred lotus seeds (Nelumbo nucifera), which exhibit an exceptional longevity of 1300 years (Shen-Miller, 2002). Overexpression of NnMT2a and NnMT3 in Arabidopsis significantly enhances seed germination vigor after aging treatment and under abiotic stresses (Zhou et al., 2012). In tobacco, simultaneously over-expressing genes encoding Cu/Zn-superoxide dismutase (CuZnSOD) and APX in plastids improves seed longevity and germination under various environmental stress conditions (Lee et al., 2010).Vitamin E (tocopherol) acts against phospholipid radicals and prevents lipid peroxidation during seed storage, germination and seedling establishment (Grene, 2002; Sattler et al., 2004). Indeed, the vitamin E deficient vte1 and vte2 mutants exhibited a significantly reduced seed longevity (Sattler et al., 2004). The importance of radical scavenging compounds is supported by the fact that recalcitrant seeds need a higher constant level of vitamin C than orthodox seeds to protect themselves from ROS (Tommasi et al., 1999). Seeds are equipped with antioxidant systems that function redundantly, thus a single mutant might not confer the phenotype. The vitamin-C deficent1-1 mutant is affected in ROS scavenging, and the glutathione deficient cadmium sensitive2-1 mutant has a similar seed longevity phenotype as wild type (Clerkx et al., 2004a). Free radical-counteracting processes and detoxification mechanisms are closely related to control of the pro-oxidant/antioxidant balance both during seed storage and germination. When the pro-oxidant and antioxidant scavenging systems are saturated by ROS oxidation, detoxification mechanisms might be affected and ROS control is lost, resulting in dead seeds.

Deterioration-related damagesDNA lesions are induced not only during seed desiccation but also during seed aging and seed germination (Osborne et al., 1981; Osborne et al., 1984; Bray and West, 2005). ROS induces a variety of lesions in DNA (Morales-Ruiz et al., 2003), in which a prominent DNA lesion is 7,8-dihydro-8-oxoguanine (8-oxo-G), which can form base pairs with adenine instead of cytosine during DNA replication, leading to GC/TA transversions. AtOGG1 is a DNA glycosylase/apurinic/apyrimidinic (AP) lyase that is involved in base excision repair for eliminating 8-oxo-G from DNA. AtOGG1 was strongly up-regulated during seed desiccation and imbibition, over-expression of AtOGG1 in Arabidopsis enhances seed longevity and abiotic stress tolerance (Chen et al., 2012). ROS can also result in the production of single and double-strand DNA breaks, that are re-joined by the action of DNA ligases. Waterworth et al. (2010) showed that AtLIG6 and AtLIG4 are major determinants of Arabidopsis seed quality and longevity because seeds of the lig6 mutant and the lig6 lig4 double mutant were hypersensitive to artificial aging.


18

Therefore, maintenance of a functional DNA repair complex appears to be essentials for seed viability.

The formation of abnormal amino acid residues is a major source of spontaneous age-related protein damage in cells. Protein synthesis was proven to be essential for seed germination since cycloheximide completely blocks seed germination (Rajjou et al., 2004). Protein translation ability was correlated with seed aging (Rajjou et al., 2008). The de novo synthesis of proteins from stored mRNA can allow the renewal of non-functional proteins that were affected during aging. In terms of energy, protein repair is less costly than synthesizing new proteins. Protein L-isoaspartyl methyltransferase (PIMT), that repairs age-damaged aspartyl and asparaginyl residues in protein, improves seed longevity and vigor in Arabidopsis (Oge et al., 2008; Verma et al., 2013). PIMT1 is in charge of repairing abnormal cytosolic proteins while PIMT2 repairs nuclear proteins. Protein oxidation occurs also at methionine residues, that are oxidized to methionine sulfoxide. This oxidation is reversed by methionine sulfoxide reductases (MSRs). Chatelain et al. (2013) demonstrated that seed longevity is strongly linked to the abundance and enzymatic capacity of MSR in Medicago (Medicago truncatula) and Arabidopsis. These studies indicate an important role of maintaining functional protein repair systems in seed longevity.

Seed longevity measurementIn the laboratory, natural aging occurs when seeds are stored at uncontrolled ambient conditions for extended periods of time. These stored seeds can be withdrawn at certain intervals to test their germination ability (seed longevity). The reduction of seed germination is an indicator of seed vigor loss and aging. Under such laboratory conditions, orthodox seeds can remain viable for many years, particularly Arabidopsis seeds can still perform relatively well after five years of storage. To overcome the waiting time due to slow natural aging, artificial aging methods have been developed and applied to study seed longevity in various species. The rate of seed aging depends on the seed moisture content, temperature, and initial seed quality (Walters, 1998; Walters et al., 2005), thus often high relative humidity and high temperature are used in accelerated aging (AA) and controlled deterioration tests (CDT) (Tesnier et al., 2002). The difference between these two methods is that seeds are treated in high humidity and temperature at the same time in the AA while seeds are equilibrated in humidity prior the temperature application in CDT (ISTA, 2012). The effect of oxygen during storage is discussed above, and recently Groot et al. (2012) reported a longevity method applying elevated partial pressure of oxygen (EPPO) to seeds. Also after artificial aging, seed viability is evaluated by a germination assay.

Genetic variation of seed longevity mutantsMost of the mutants that exhibit seed longevity phenotypes were not isolated in screens


19

for the trait itself, thus seed longevity is a pleotropic effect of genes that were initially identified to regulate other traits. Seed coat mutants, as reported by Debeaujon et al. (2000), have significant reduced seed longevity. The seed coat is suggested to protect the embryo against mechanical damages, acting as a physical barrier for water and gases, particularly oxygen, thereby slowing down ROS formation. The seed dormancy mutants dog1 and rdo4 also have a reduced seed longevity (Bentsink et al., 2006; Liu et al., 2007). In the case of rdo4, the involvement of histone H2B monoubiquitination and chromatin remodeling was shown to be involved in seed longevity (Liu et al., 2007). The frostbite1 (fro1) mutant, isolated for its reduced response to low temperature, constitutively accumulates ROS (Lee et al., 2002). As a result, it showed a reduced seed germination after artificial aging, which supports the negative effect of oxidative stress on seed longevity (Clerkx et al., 2004a). Most of plant hormones appear to not directly affect seed longevity, except for ABA. The abi3 mutant is severely affected in seed longevity, likely due to the fact that ABI3 is a key regulator in seed maturation and consequently results in an improper acquisition of seed longevity in the first place (Sugliani et al., 2009). The role of ABA is also indicated by the aba1-5 mutant, that is more sensitive to aging than wild type seeds (Clerkx et al., 2004a). The involvement of other hormones such as GA, ethylene, jasmonic acid was not established because seed longevity of gai, ga1-3, ethylene resistant1 (etr1) and jasmonic acid resistant1 (jar1) mutants did not differ from that of wild type (Clerkx et al., 2004a). A screen for better seed longevity in the abi3-5 mutant led to the isolation of the suppressor of abi3-5 mutants (sua1 to sua4). Sugliani et al. (2010) showed that sua1 suppressed abi3-5 in a allele specific manner and acted as a splicing factor that influences seed maturation by controlling the alternative splicing of ABI3.

Natural genetic variation for seed longevitySeed longevity is a quantitative trait for which variation is present among naturally occurring accessions. QTLs for seed longevity have been identified following both natural aging and artificial aging in the Arabidopsis RIL populations Ler/Cvi (Bentsink et al., 2000) and Ler/Shakdara (Sha) (Clerkx et al., 2004b). The colocation of major QTLs detected in two assays indicates similar effects obtained by natural and artificial aging. Natural modifiers for better seed longevity in abi3-5 and lec1 mutant background were present in Seis am Schlern and Sha Arabidopsis accessions (Sugliani et al., 2009).

“Omics” analyses for seed longevitySo far, transcriptome analyses to study seed longevity have not been reported. However, recently Verdier et al. (2013) dissected late maturation events related to the acquisition of desiccation tolerance and longevity in Medicago seeds using transcriptomic, and metabolic profiling and a regulatory network-based approach. The network revealed


20

distinct co-expression modules related to the acquisition of desiccation tolerance, longevity, and pod abscission, in which the seed longevity module was enriched for genes involved in RNA processing and translation. Seed longevity genes were highly connected to two MtAP2/EREBP genes and two MtbZIP transcription factors. Metabolome profiles indicate that gulose and stachyose levels were increased and correlated with longevity during seed maturation.

The seed longevity proteome has been investigated in dry and imbibed Arabidopsis seeds, after natural and artificial aging (Rajjou et al., 2008). As a result of aging, the seed proteome is highly oxidized as revealed by increased levels of protein carbonylation. This work proposes the importance of the detoxification of the toxic compound cyanin by β-mercaptopyruvate sulfurtransferase. The translational capacity and mobilization of seed storage reserve efficiency are essential mechanisms for seed longevity. In agreement with that, proteome analyses in maize (Zea mays) seeds upon aging showed that proteins involved in metabolism and energy (glycolysis, tricarboxylic acid cycle, the electron transport chain and oxidative phosphorylation) were the largest down-regulated protein groups (Wu et al., 2011; Xin et al., 2011).

Scope of the thesisIn this thesis, I studied seed dormancy and seed longevity in Arabidopsis exploiting genetic variation to identify the underlying genes. The study starts with the genetic analyses for seed longevity after natural aging using natural variation exhibited in six RIL populations. The objective is to reveal genetic regulators for seed longevity and their possible association with seed dormancy that had earlier been identified in the same populations (Chapter 2). The identification of genes involved in seed longevity was continued with proteomic analyses on dry seeds at two physiological states, AR and 4-year old seeds, of four genetic lines (Chapter 3). This chapter aims to investigate seed longevity pathways that are shared by all genotypes, and those that are specific for the studied genotypes. To understand the negative correlation between seed dormancy and seed longevity, that is identified in Chapter 2, I performed a genetic characterization and fine-mapping of two collocating seed loci; the second strongest QTL for seed dormancy DOG2 and the strongest QTL for seed longevity Germination Ability After Storage1 (GAAS1) in the Ler/Cvi population (Chapter 4). Chapter 5 describes the genetic variation for seed dormancy induced by mutagenesis in NILDOG6. This chapter aims at the isolation and characterization of novel seed dormancy modifier genes. In Chapter 6, the findings presented in this thesis are discussed.


Natural variation for seed longevity and seed dormancy

are negatively correlated in Arabidopsis thaliana

Thu-Phuong Nguyen 1,2, Paul Keizer 3,4, Fred van Eeuwijk 3,4, Sjef Smeekens 1,3 and Leónie Bentsink 1,2

1 Wageningen Seed Laboratory, Laboratory of Plant Physiology, Wageningen University, 6708 PB Wageningen, The Netherlands; 2 Department of Molecular Plant Physiology, Utrecht University, 3584 CH Utrecht, The Netherlands; 3 Centre for BioSystems Genomics, 6700 AB Wageningen, The Netherlands; 4 Biometris–Applied Statistics, Wageningen University and Research Centre, 6708 PB Wageningen, The Netherlands

Published in Plant Physiology (2012) 160: 2083–2092

2Chapter

Chapter 2

22

Abstract

Dormancy is a state of metabolic arrest that facilitates survival of organisms during environmental conditions incompatible with their regular course of life. Many organisms have deep dormant stages to promote an extended life span (increased longevity). In contrast, plants have seed dormancy and seed longevity described as two traits. Seed dormancy is defined as a temporary failure of a viable seed to germinate in conditions that favor germination, whereas seed longevity is defined as seed viability after dry storage (storability). In plants, the association of seed longevity with seed dormancy has not been studied in detail. This is surprising given the ecological, agronomical and economic importance of seed longevity. We studied seed longevity to reveal its genetic regulators and its association with seed dormancy in Arabidopsis thaliana. Integrated quantitative trait locus (QTL) analyses for seed longevity, in six recombinant inbred line (RIL) populations, revealed five loci: Germination Ability After Storage1 (GAAS1) to GAAS5). GAAS loci colocated with seed dormancy loci, Delay Of Germination (DOG), earlier identified in the same six RIL populations. Both GAAS loci and their colocation with DOG loci were validated by near isogenic lines. A negative correlation was observed, deep seed dormancy correlating with low seed longevity and vice versa. Detailed analysis on the collocating GAAS5 and DOG1 QTL revealed that the DOG1-Cape Verde Islands allele both reduces seed longevity and increases seed dormancy. To our knowledge, this study is the first to report a negative correlation between seed longevity and seed dormancy.

QTLs for seed longevity after natural agingChapter 2

23

Introduction

Dormancy describes a state of apparent metabolic arrest during which normal progression of life activities and development are dramatically reduced or brought to a halt. Dormancy facilitates the survival of organisms during environmental conditions that cannot support the regular course of life. Many organisms have dormant stages, which in different species have different names, such as dauer stage in Drosophila ssp., diapause in water flea and fish embryos, akinetes in cyanobacteria, spores in yeast (Saccharomyces cerevisiae) and dormancy in plant seeds and flower buds. Organisms can enter the dormant state due to environmental cues such as a lack of water by undergoing desiccation, low temperature or through developmentally programmed arrest as occurs, for example,. in yeast spores and plant seeds. In most organisms dormancy has been related to an extension of their life span (increasing longevity) (Lubzens et al., 2010).

In contrast to most other organisms described above, in plant seeds dormancy and longevity have been described as two separate traits. Seed dormancy is defined as a temporal failure of a seed to germinate in conditions that favor germination (Bewley, 1997). Seed dormancy can be overcome by environmental cues (i.e. seed dry storage [after-ripening] and cold stratification). Seed dormancy has been studied extensively (Finch-Savage and Leubner-Metzger, 2006; Holdsworth et al., 2008) and recently, a quantitative trait locus (QTL) analysis in combination with transcriptome analyses in Arabidopsis (Arabidopsis thaliana) has revealed that natural variation for seed dormancy is controlled by independent genetic and molecular pathways (Bentsink et al., 2010).

Seed longevity is defined as seed viability after seed dry storage (storability) and, therefore, describes the total seed life span (Rajjou and Debeaujon, 2008). This storability period includes both the dormant and non-dormant states. During seed storage, seeds deteriorate, lose vigor, and, as a result, become more sensitive to stresses during germination, and ultimately die. The rate of this aging depends on the seed moisture content, temperature, and initial seed quality (Walters, 1998; Walters et al., 2005). Seed longevity is a quantitative trait for which variation is present among naturally occurring accessions. QTLs for seed longevity have been identified after natural aging in Arabidopsis (Bentsink et al., 2000; Clerkx et al., 2004b), lettuce (Lactuca sativa) (Schwember and Bradford, 2010) and rice (Oryza sativa) (Sasaki et al., 2005), and after artificial aging imposed by a controlled deterioration test (CDT) in Arabidopsis (Bentsink et al., 2000; Clerkx et al., 2004b), rice (Miura et al., 2002), and wheat (Landjeva et al., 2009). The ability of the CDT to predict seed longevity was shown by the colocalization of major QTLs after artificial and natural aging in two different studies using the Arabidopsis recombinant inbred line (RIL) populations Landsberg erecta (Ler)/Cape Verde Islands

Chapter 2

24

(Cvi) (Bentsink et al., 2000) and Ler/Shakdara (Sha) (Clerkx et al., 2004b). Besides these genetic analyses, also proteome studies in Arabidopsis have shown that similar molecular events occur during natural and artificial (CDT) aging (Rajjou et al., 2008). In contrast with this, Schwember and Bradford (2010) did not find overlap between seed longevity QTLs under conventional and CD storage conditions in lettuce.

The genetic basis of seed longevity is unclear. However, there are several groups of mutants that have altered seed longevity. The majority of mutants with known effects on seed longevity are the seed developmental mutants. Mutations in the key regulators of seed maturation lead to rapid loss of viability upon storage, as has been shown for leafy cotyledon1 (lec1) and abscisic acid intensitive3 (abi3) mutants (Ooms et al., 1993; Clerkx et al., 2004a; Sugliani et al., 2009). Another group of mutants with a seed longevity phenotype consists of the testa mutants. The seed coat or testa acts as a structural barrier to protect the embryo and seed reserves from biotic and abiotic stresses. The testa-defective mutants, including the transparent testa (tt) mutants and the aberrant testa shape (ats) (Debeaujon et al., 2000), display considerably reduced seed longevity. Moreover, mutations in protection and repair systems that prevent seed vigor loss lead to decreased seed longevity. Arabidopsis mutants affected in vitamin E (lipophilic antioxidant) biosynthesis, vte1 and vte2, exhibited significantly reduced seed longevity (Sattler et al., 2004). Waterworth et al. (2010) showed that DNA Ligase6 and 4, which are essential to maintain genome integrity in plants, are major determinants of Arabidopsis seed quality and longevity. The atlig6 mutant and atlig6 atlig4 double mutant are most sensitive to controlled seed aging.

Proteins and enzymes are also described as factors that may determine seed longevity. Heat stress transcription factor (HSF) over-accumulating seeds of transgenic Arabidopsis display enhanced accumulation of heat stress protein and improved tolerance to aging (Prieto-Dapena et al., 2006). Protein repair appears to play a key role in long-term survival of seeds in the dry state. PIMT (protein L-isoaspartyl methyltransferase), which limits and repairs age-damaged aspartyl and asparaginyl residues in proteins, has been associated with greater seed longevity because it highly accumulated in sacred lotus (Nelumbo nucifera) seed, one of the world’s longest-living seeds (1300 years) (Shen-Miller, 2002). Over-expression of PIMT1 in Arabidopsis enhanced both seed longevity and germination vigor, whereas reduced PIMT1 expression led to increased sensitivity to aging treatments and loss of seed vigor under stressful germination conditions (Oge et al., 2008). However, PIMT exhibited a decreased activity in naturally aged barley (Hordeum vulgare) seeds (Mudgett et al., 1997). In addition, enzymes playing roles in the detoxification of reactive oxygen species, such as glutathione peroxidase and glutathione reductase (Bailly et al., 1996), and toxic cyanide compounds such as ß-mercaptopyruvate sulfurtransferase (Rajjou et al., 2008), are important to prolong seed longevity.


25

Given the important precondition of many organisms to become dormant before exposure to and survival of long-term (desiccation) stress, as well as the ecological, agronomical, and economic importance of seed longevity, it is surprising that the association of seed longevity and seed dormancy has not been studied in much detail. The current idea is that seed dormancy and seed longevity are positively correlated. This hypothesis is based mainly on the performance of the earlier mentioned lec1, abi3 (Ooms et al., 1993; Clerkx et al., 2004a; Sugliani et al., 2009), tt, and ats mutants (Debeaujon et al., 2000) but also on the loss-of-function mutant in the DOG1 gene (Bentsink et al., 2006), and the green seed mutant (enhancer of abi3-1) (Clerkx et al., 2003). All these mutants have a reduced dormancy level that correlates with reduced seed longevity.

Here, we study natural variation for seed longevity in order to reveal its genetic regulators and their possible association with seed dormancy. We have performed integrated QTL analyses for seed longevity, measured as germination ability after storage at ambient conditions, in six RIL populations. These populations were derived from crosses between the Arabidopsis standard laboratory accession Ler and the accessions Cvi, Antwerp (An-1), St. Maria do Feira (Fei-0), Kashmir (Kas-2), Kondara (Kond), and Sha, which were previously used for seed dormancy analyses (Bentsink et al., 2010). The major seed longevity QTLs colocated with the earlier identified seed dormancy QTLs. QTLs and colocation have been validated by near isogenic lines (NILs). The results are discussed in the context of the current knowledge on seed dormancy and seed longevity.

Results

Seed longevity of the parental accessions and their RIL populationsSeed longevity was measured as germination ability after seed dry storage at ambient conditions, which is referred to as “natural aging” in this work. We have used seeds of seven accessions (Ler, An-1, Cvi, Fei-0, Kas-2, Kond and Sha) and six RIL populations that were constructed from crosses between Ler and the other accessions for this study. The RILs were grown between 2002 to 2005 (Table S1). After harvest, seed dormancy behavior was measured until germination reached 100% (Bentsink et al., 2010). In early 2010 (after 4 to 7 years of dry storage), the same RILs were assessed for seed longevity (germination ability after aging) using the Germinator tool developed by Joosen et al. (2010). As a result of aging, the maximal germination percentage (Gmax) decreased. Overall, the six populations exhibited variation for Gmax (ranging from 0% to 100%) (Fig. 1A). Transgression beyond values of both parents was observed in all populations, indicating that both parental lines carried alleles-increasing and -decreasing seed

Chapter 2

26

longevity (Fig. 1A). The Ler/Kas-2 population harvested in 2002 was the most aged population; approximately one-third of the RILs germinated to less than 50%, whereas the majority of the RILs in Ler/Fei-0 (harvested in 2004) and Ler/Kond (harvested in 2003) populations had a Gmax higher than 75%.

Figure 1. Frequency distributions of seed longevity presented by four germination parameters in six RIL populations. The different RIL populations are indicated at the top. The x axis contain the trait values for Gmax (%; A), AUC (B), t10totS (h; C), and t50Gmax (h; D). Arrowheads depict the value of parental lines (black arrowheads for Ler and gray arrowheads for other accessions).

The Ler parent, which was grown together with each population had a Gmax ranging from 54% to 100% (Table S1). The Ler parent grown with the Ler/Cvi population harvested in 2005 had a lower germination ability (Gmax of 80%) than the longer stored Ler parents grown in earlier years (2003 and 2004). This might be the consequence of different growing environments, since it is known that environmental conditions during seed maturation strongly affect seed quality (Contreras et al., 2008).

Seed germination after storage is not only analyzed by Gmax but also by other germination parameters, such as germination rates (time to reach 10% germination of the total number of seeds [t10totS] and time to reach 50% germination of the total number of germinated seeds [t50Gmax]) and area under the curve (AUC, a parameter that describes the germination curve based on the germination rate and the Gmax), which were also measured by the Germinator tool. There was a lot of variation for these three additional parameters in the six populations, t10totS ranged from 30h to 120h,


27

t50Gmax from 30h to 150h and AUC from 0 to 85 (Fig. 1B-1D). For these germination parameters we do not have the initial values, since the Germinator tool which allows scoring of large populations was not developed at the time the seeds were harvested. Moreover, as a result of aging, a reduction of germination behavior becomes first apparent in a lower germination rate (10totS and t50Gmax), followed by a decrease of Gmax, which are both reflected by a reduction in AUC. For this reason, these parameters might still contain valuable information for the longevity analyses. Therefore, we will focus our study mainly on QTL found for Gmax but will compare these also with QTLs identified for AUC and use these together to identify colocation with seed dormancy.

Integrated QTL analyses for seed longevity in six RIL populationsIn order to identify loci controlling seed longevity, mixed-model QTL analyses were performed as described by Bentsink et al. (2010). The six RIL populations were explored simultaneously, and the different allele effects were examined for each QTL and every population. In total, five QTLs for seed longevity measured as Gmax were identified. We named these QTL Germination Ability After Storage1 (GAAS1) to GAAS5, (Fig. 2). These loci showed strong additive effects accounting for an average of 23.3% of the total explained variance (Table I); no epistatic interaction among loci was detected. The mapping results revealed three major loci, GAAS1, GAAS2, and GAAS5, which were also detected for the other parameters (Table I and S3). The genome-wide significances of the major QTLs, calculated on a -10Log(P) scale using a final QTL model after backward selection from the composite interval mapping model, were 15.6, 20.5, and 7.2, respectively, while those of the minor QTLs were below 7.2.

The major locus GAAS2 explained 1.9 to 15.5% of the phenotypic variation in the individual populations and had a significant effect in almost all populations (Table I, Fig. 2). The Ler allele of this locus decreased seed longevity (lower Gmax) in all cases. GAAS1, the second strongest QTL, accounting for 0.6 to 12.5% of the effect in the individual populations, only showed a significant effect in the Ler/Cvi population; also for this QTL the Ler allele decreased seed longevity. GAAS5, explaining 0.4 to 5.4% of phenotypic variation in the single populations, has a significant effect in two of the six populations (Ler/An-1and Ler/Cvi); for this QTL the Ler allele increased seed longevity.

Seven additional minor QTLs identified for the other parameters (AUC, t10totS, and, t50Gmax) are named GAAS6 to GAAS12 (Table S2).

Confirmation of the GAAS lociTo characterize the GAAS loci, NILs carrying single genomic fragments of different accessions into the Ler genetic background were used (Table II). The germination behavior of these lines was analyzed after five years of seed dry storage. We could confirm four of the five GAAS (Gmax) loci. For the two strongest QTLs, GAAS1 and GAAS2, we had only one NIL each available (NILGAAS1-Cvi and NILGAAS2-An-1, respectively). The seeds of these lines performed significantly better after storage when compared

Chapter 2

28

Gmax

Supp

ort

inte

rval

QTL

effe

cts

Rang

e of

ex

pl. v

aria

nce

by lo

cus (

%)

QTL

nam

eCh

r.Po

sitio

n-1

0Log

(P)

Ler/

An-1

Ler/

Cvi

Ler/

Fei-0

Ler/

Kas-

2Le

r/Ko

ndLe

r/Sh

a

GAAS

11

29.8

15.6

27.6

-29.

8-0

.027

-0.1

010.

011

0.00

7-0

.005

-0.0

260.

6 –

12.5

GAAS

23

5.4

20.5

2.6-

7.8

-0.0

61-0

.044

-0.0

59-0

.095

-0.0

17-0

.043

1.9

– 15

.5

GAAS

33

67.9

4.1

65.6

-72.

90.

019

0.02

80.

027

0.03

10.

009

0.01

11.

0 –

3.5

GAAS

44

6.2

4.0

3.3-

75.4

-0.0

02-0

.025

0.00

90.

024

0.03

1-0

.028

0.2

– 3.

9

GAAS

55

93.3

7.2

89.5

-102

.70.

045

0.04

10.

032

0.01

6-0

.004

0.02

80.

4 –

5.4

Mea

n0.

830.

810.

910.

610.

900.

84

Expl

aine

d va

rian

ce b

y m

ain-

effe

ct Q

TL (%

)22

.349

.222

.418

.47.

919

.8

Aver

age

of e

xpla

ined

var

ianc

e by

mai

n-ef

fect

QTL

(%)

23.3

Tabl

e I.

QTLs

for G

max

in si

x po

pula

tions

, as o

btai

ned

by in

tegr

ated

ana

lyse

s com

pris

ing

com

posi

te in

terv

al m

appi

ng a

nd b

ackw

ard

sele

ctio

nTh

e ge

nom

e-w

ide

thre

shol

d in

com

posi

te in

terv

al m

appi

ng w

as 2

.74

on th

e −1

0Log

(P) s

cale

, with

P th

e P

valu

e. Q

TL n

ame,

chr

omos

ome,

pos

ition

, sig

nific

ance

ex

pres

sed

on a

−10

Log(

P) s

cale

, and

a ±

1.5

dro

poff

inte

rval

on

the

−10L

og(P

) sca

le (s

uppo

rt in

terv

al) a

re p

rese

nted

. P v

alue

s ar

e ta

ken

from

the

final

mul

ti-QT

L m

odel

afte

r ba

ckw

ard

sele

ctio

n. D

ropo

ff in

terv

als

are

asse

ssed

on

the

com

posi

te in

terv

al m

appi

ng p

rofil

e. Q

TL a

llele

sub

stitu

tion

effe

cts

(Gm

ax) a

re g

iven

in th

e ri

ght p

art o

f the

tabl

e. A

neg

ativ

e va

lue

indi

cate

s tha

t Ler

is d

ecre

asin

g th

e Gm

ax, w

here

as a

pos

itive

val

ue in

dica

tes t

hat L

er in

crea

ses t

he G

max

. Sig

nific

ant e

ffect

s ar

e in

dica

ted

in b

old;

thes

e ar

e th

e ef

fect

s of t

he Q

TL in

dica

ted

in F

igur

e 1.

In th

e bo

ttom

par

t of t

he ta

ble,

for e

ach

popu

latio

n’s m

ean

Gmax

, the

exp

lain

ed v

aria

nce,

an

d th

e av

erag

e of

exp

lain

ed v

aria

nce

by m

ain-

effe

ct Q

TLs

are

pres

ente

d in

per

cent

ages

. The

last

col

umn

show

s th

e ra

nge

of e

xpla

ined

effe

ct o

f eve

ry lo

cus

in

perc

enta

ges.


29

Figure 2. Integrated QTL analyses for seed longevity expressed as Gmax in six RIL populationsGenome-wide profiling of simple interval mapping with genome-wide threshold of 2.74 on the -10Log(P) scale (A). Composite interval mapping with fixed cofactor revealed five QTL loci (GAAS1 to GAAS5) (B). Cofac-tor positions are depicted by black vertical bars. The confidence interval of each locus is presented by the blue columns, the darker the color, the more significant the QTL. QTL effects for every single population (C). Orange indicates that the Ler allele increases seed longevity (Gmax), and cyan indicates that the Ler allele decreases seed longevity. The intensity of the color corresponds to the size of the QTL effect: the higher the intensity, the stronger the effect. Black vertical bars indicate significance of the QTL effect in that population. Confirmation of the major seed longevity loci (GAAS1, GAAS2, GAAS3, and GAAS5) (D). The genotypes of NIL-GAAS1-Cvi, NILGAAS2-An-1, NILGAAS3-Kas-2, and NILGAAS5-Sha as well as Ler are schematically presented. Orange indicates the Ler alleles, and cyan indicates the alleles of the other accessions; missing marker data are indicated in light gray. Seed longevity (Gmax) values of the four NILs and Ler after natural aging for 5 years have been indicated at the right. The Gmax of the NILs are significantly different from that of Ler (P < 0.05).

with their genetic background Ler (Table II, Fig. 2D). GAAS3 was validated by NILGAAS3-Kas-2, which showed a significant reduction in Gmax when compared with Ler. For the GAAS5 locus, we had five NILs available (NILGAAS5-Cvi, NILGAAS5-Fei-0, NILGAAS5-Kas-2, NILGAAS5-Kond, and NILGAAS5-Sha), of which only NILGAAS5-Fei-0 and NILGAAS5-Sha showed a significant reduction for Gmax in comparison with Ler. Non-

Chapter 2

30

Tabl

e II

. Con

firm

atio

n of

GAA

S lo

ci b

y N

ILs

The

germ

inat

ion

beha

vior

of a

set o

f NIL

s gro

wn

in 2

007

(QTL

locu

s, N

IL n

ame,

and

ori

gina

l nam

e [B

ents

ink

et a

l., 2

010]

) afte

r 5 y

ears

of n

atur

al a

ging

is p

rese

nted

by

Gm

ax, A

UC, t

10to

tS, t

50Gm

ax, a

nd p

erce

ntag

e of

nor

mal

see

dlin

gs. A

vera

ge v

alue

s an

d st

anda

rd e

rror

s (a

s in

dica

ted

for

each

NIL

) are

indi

cate

d. G

erm

inat

ion

beha

vior

s tha

t are

sign

ifica

ntly

diff

eren

t fro

m th

at o

f Ler

are

indi

cate

d by

ast

eris

ks (*

P <

0.0

5, **

P <

0.0

1).

QTL

Locu

sN

IL n

ame

Orig

inal

nam

eGm

ax (%

)AU

Ct1

0tot

S (h

)t5

0Gm

ax (h

)N

orm

al se

edlin

g (%

)

GAAS

1N

ILGA

AS1-

Cvi

NIL

DOG2

-Cvi

49.

1 ±

6.9*

* 2

2.7

± 4.

9*65

.4 ±

6.4

74.5

± 5

.421

.4 ±

4.9

*

GAAS

2N

ILGA

AS2-

An-1

NIL

DOG2

2-An

-1 4

2.4

± 4.

4**

18.4

± 2

.1**

62.7

± 2

.773

.9 ±

1.3

16.1

± 2

.6*

GAAS

3N

ILGA

AS3-

Kas-

2N

ILDO

G6-K

as-2

13.

0 ±

3.2*

4.

1 ±

1.0*

10

7.9

± 22

.793

.4 ±

7.7

4.3

± 1

.1

GAAS

5N

ILGA

AS5-

Cvi

NIL

DOG1

-Cvi

14.6

± 3

.3

4.3

± 1.

2*

84.2

± 3

.095

.3 ±

4.2

2

.2 ±

0.5

**

GAAS

5N

ILGA

AS5-

Fei-0

NIL

DOG1

-Fei

-0

4.4

± 2

.3**

1

.4 ±

0.9

**80

.5 ±

0.0

100.

9 ±

10.3

0

.6 ±

0.4

**

GAAS

5N

ILGA

AS5-

Kas-

2N

ILDO

G1-K

as-2

22.5

± 3

.4 8

.1 ±

1.6

88.4

± 8

.684

.8 ±

3.6

6.1

± 1

.6

GAAS

5N

ILGA

AS5-

Kond

NIL

DOG1

-Kon

d24

.3 ±

4.1

11.0

± 2

.268

.9 ±

4.3

74.4

± 3

.1 8

.3 ±

2.3

GAAS

5N

ILGA

AS5-

Sha

NIL

DOG1

-Sha

8

.6 ±

1.6

**

2.3

± 0

.5**

12

5.1

± 10

.1**

97.6

± 9

.4

2.3

± 0

.6**

GAAS

7N

ILGA

AS7-

Fei-0

NIL

DOG2

0-Fe

i-034

.3 ±

5.5

15.7

± 3

.068

.6 ±

6.5

74.0

± 4

.514

.9 ±

3.1

Le

r

24.2

± 3

.6 9

.6 ±

1.9

81.8

± 6

.880

.7 ±

3.8

8.9

± 1

.7


31

germinating seeds were stained with 2,3,5 triphenyl tetrazolium chloride to determine whether they were dead (Moore, 1985). We concluded that the seeds were deteriorated, since the majority of the seeds did not stain (Fig. S1).

Comparison between seed longevity and seed dormancyIn order to explore the relationship between seed longevity and seed dormancy, we investigated the overlap between QTLs that have been mapped for these two traits. Integrated QTL analyses for both seed longevity and seed dormancy (Delay of Germination [DOG]) have been performed in the same populations (Bentsink et al., 2010), which allow a neat comparison between those two traits. Four of the five GAAS loci identified for Gmax overlapped with DOG loci (GAAS2/DOG22, GAAS3/DOG6, and GAAS5/DOG1) or mapped in very close vicinity (GAAS1/DOG2) (Fig. 3A). All five GAAS Gmax QTLs were also identified using AUC as longevity parameter, which can be explained by the high correlation between these two parameters (R2 = 0.75-0.90; Table S3). AUC showed the highest variation (Fig. 1B), which provided more statistical power in the QTL analyses and led to the identification of four additional QTLs (Table S3). Seven of the nine GAAS AUC QTLs colocated with seed dormancy QTLs (Fig. 3A). The three additional genomic regions that showed colocation are GAAS7/DOG20, GAAS10/DOG5, and GAAS11/DOG4 (Fig. 3A). Unexpectedly, a negative relationship between seed dormancy and seed longevity was observed. Deep dormancy correlated with low storability and shallow dormancy with high storability, shown by the direction of the arrow heads in Figure 3A. Since we had NILs available for most of the QTLs, we were able to analyze this correlation in more detail (Fig. 3B and 3C). The NILs were grown in two independent experiments in the greenhouse (in 2006 and 2007) and analyzed for their seed dormancy (directly after harvest) and seed longevity behavior (after five years of storage; Table II and S4). The negative correlation between seed longevity (Gmax) and seed dormancy (days of seed dry storage required to reach 50% germination [DSDS50]) that was identified by the QTL analyses was proven to be very significant, given the very high correlation coefficients (R2) of 0.66 (P = 0.03) and 0.89 (P < 0.01), respectively, for the two experiments (Fig. 3B and 3C).

Next we investigated the correlation between seed longevity and seed dormancy in the RILs, however, no correlation was found (R2 between 0.01 and 0.12 for the six populations). This lack of correlation might be explained by the fact that total genetic variation was not fully explained by the identified QTLs (Table I) and by the low variation for seed longevity in RIL populations (Fig. 1). Furthermore, we have ended up with a random combination of seed longevity and dormancy loci in the RILs due to the broken linkage (which is the nature of this type of populations). This results in a random combination of phenotypes since some of these loci have a stronger longevity effect and others a stronger dormancy effect. To remove the above-mentioned noise,

Chapter 2

32

we have performed correlation analyses on RILs that have been selected for either seed longevity increasing or decreasing alleles at the position of strongest four QTLs (GAAS1, GAAS2, GAAS3, and GAAS5) in two of the populations that show the largest variation for seed longevity (Ler/Kas-2) or the maximum possibility of QTL colocation within a population (Ler/Cvi) (Table S5 and S6). These analyses showed again the negative relationship between seed longevity and dormancy (Fig. S4) with R2 of 0.58 for Ler/Kas-2 (P < 0.01) and 0.78 for Ler/Cvi (P < 0.01).

Figure 3. Colocation and correlation between seed longevity (GAAS) and seed dormancy (DOG) QTLsIntegrated composite interval mapping profiles of seed longevity (AUC; top) and seed dormancy (DSDS5; bottom) performed in the six RIL populations (A). The confidence interval of each locus is presented by the grey columns: the darker the color, the more significant the QTL. Cofactor positions are depicted by black arrows: arrows pointing up indicate that the Ler allele is increasing the trait value, and when arrows point down, Ler decreases the trait value. Correlation of seeds longevity (Germination percentage) and seed dormancy (DSDS50) in NILs (B and C). Experiment harvested in 2006 including NILDOG1/GAAS5-Sha, NILDOG22/GAAS2-An-1, NILDOG2/GAAS1-Cvi, NILDOG6/GAAS3-Fei-0, NILDOG6/GAAS3-Kas-2, NILDOG20/GAAS7-Fei-0, and Ler (B). Experiment harvested in 2007 including NILDOG20/GAAS7-Fei-0, NILDOG1/GAAS5-Sha, NILDOG1/GAAS5-Kond, NILDOG1/GAAS5-Kas-2, NILDOG22/GAAS2-An-1, NILDOG2/GAAS1-Cvi, NILDOG6/GAAS3-Kas-2, and Ler (C).

Seed longevity and dormancy are regulated by one gene at the position of GAAS5/DOG1The colocation between seed longevity and seed dormancy QTLs can be caused by a single gene or by separate linked genes. The genetic nature of the colocation can only be studied when the genes underlying the QTLs are identified. So far, the only identified QTL is the dormancy locus DOG1 (collocating with GAAS5). We used transgenic lines that carry the DOG1 Cvi allele in the Ler genetic background to investigate whether these lines, in addition to their increased dormancy level also showed a seed longevity phenotype. The Gmax of seeds (Ler, NILGAAS5/DOG1-Cvi, and two independent


33

transformants) that had been naturally aged for 7 years was still above 70% and did not reveal clear differences between any of the genotypes (Fig. S3). In order to accelerate the aging, we have stored these seeds at 75% relative humidity (RH) for 59 days and analyzed the Gmax at several time points during this storage. After 59 days of storage in 75% RH, all seeds of all genotypes were completely deteriorated. However, the two transformants are significantly less storable than the Ler control, which shows that DOG1 does not only control seed dormancy but also seed longevity (Fig. 4 and S3). This result supports the negative correlation between seed longevity and seed dormancy.

Figure 4. Seed dormancy and longevity phenotypes of Ler, NILDOG1/GAAS5-Cvi and two independent transformantsSeed dormancy measured as DSDS50 and longevity after 7 years of storage and 45 days in 75% relative humidity (Germination percentage) is shown for Ler, NILDOG1/GAAS5-Cvi, and two independent transformants (SR3-1 and SR3-2), which contain the DOG1 Cvi allele in the Ler genetic background.

Discussion

In order to study natural variation for seed longevity, we used seed batches that had been stored for 4 to 7 years in ambient conditions. Seed longevity had been analyzed in Arabidopsis populations previously, using artificial (CDT) (Bentsink et al., 2000; Clerkx et al., 2004b; Joosen et al., 2012) and natural aging (Bentsink et al., 2000). Bentsink et al. (2000) showed that naturally aged seeds of the Ler/Cvi RIL population that had been stored for 4 years led to the detection of one QTL on chromosome 1 (GAAS1 region). This locus was also the major QTL in our current experiment for the Ler/Cvi population. However, with this long-term aged seed, we were able to identify two additional QTLs (GAAS2 and GAAS5) in this Ler/Cvi genetic background. GAAS1 and GAAS2 or collocating QTL were also detected after artificial aging in the Ler/Cvi, Ler/Sha, and Bayreuth/Sha (not for GAAS2) RIL populations (Bentsink et al., 2000; Clerkx et al., 2004b; Joosen et al., 2012). GAAS5 appears to be specific for natural aging and does not show a QTL after controlled deterioration in the Ler/Cvi and Ler/Sha populations, which indicates that the controlled deterioration test does not completely mimic natural aging. Overall, we were able to identify many more loci than in earlier work (Bentsink et al., 2000; Clerkx et al., 2004b; Joosen et al., 2012), probably due to the much longer time that the seeds had been stored, and because of the integrated approach on the multiple populations that was taken.

Chapter 2

34

The parents of every population were grown together with the RIL populations. Ler differed the most from Fei-0 and Kas-2 accessions for seed longevity (Table S2). In both cases, Ler seeds were better storable, which might be explained by the fact that the Ler allele contributes to better storability for almost all GAAS loci except for GAAS2 (Fig. 2). The seed longevity of Ler was not very different from those of the accessions An-1, Cvi, Kond, and Sha, but transgressions beyond these parents were identified for those populations (Fig. 1), which resulted in the identification of seed longevity QTLs for which alleles of both parents contributed to better storability (Fig. 2).

GAAS2 and GAAS5 had significant allelic effects in more than one population (Table I and S2), which may indicate the importance of these seed longevity loci under natural selection. For GAAS1 and GAAS2, the Ler allele decreased seed longevity while the GAAS5 Ler allele increased seed longevity. Epistatic interactions between the seed longevity loci were not identified, which indicated that natural variation for seed longevity in these populations is determined by additive loci. However, part of the differences might result from genotype-by-environment interactions due to differences in growing environments of each population.

Candidate genesSeveral GAAS genomic regions identified here contain genes previously associated with seed longevity. Most obvious is the colocation of GAAS2 with the vitamin E locus. Sattler et al. (2004) have shown that vitamin E (tocopherol) is essential for seed longevity in an artificial aging assay, as it prevents lipid oxidation during seed germination. Genetic analysis of seed vitamin E levels in the Cvi/Ler and Columbia/Ler populations exhibited a common QTL on the top of chromosome 3 (GAAS2 region), namely QVE7 and QVE8, respectively (Sattler et al., 2004).

The DNA ligase AtLIG4 coincides within the confidence interval of GAAS6 in the middle of chromosome 1. Repair of DNA damage in seeds to maintain genome integrity is one of the mechanisms to prevent seed deterioration (Waterworth et al., 2010). DNA damage is associated with single and double strand breaks, which can be re-joined by DNA ligase. Two DNA ligase genes, AtLIG4 and AtLIG6, were shown to be involved in DNA repair upon seed imbibition in Arabidopsis. Mutations in these genes lead to a decreased seed viability and seed germination vigor after a CDT.

PIMT genes might be the underlying loci for GAAS3 and 5. PIMT1 and PIMT2, that are located on the lower arms of chromosome 3 and 5, respectively, play a role in the repair of age-related protein damage in Arabidopsis thaliana (Oge et al., 2008). The accumulation of the PIMT1 enzyme enhances resistance to seed vigor loss induced by CDT.


35

Colocation between seed longevity QTLs and seed dormancy QTLsSeveral QTL studies have been performed for seed longevity (Bentsink et al., 2000; Miura et al., 2002; Clerkx et al., 2004b; Sasaki et al., 2005; Landjeva et al., 2009; Schwember and Bradford, 2010) and seed dormancy (Bentsink et al., 2010), however none of these discuss the relationship (colocation and/or correlation) of both traits. Therefore, we report, to our knowledge for the first time, a negative correlation between seed longevity and seed dormancy QTLs. Lower storability levels correlated with higher seed dormancy levels, and conversely, better storability with lower seed dormancy. This finding is unexpected since current seed literature only describes correlations of low longevity with low dormancy and high longevity with high dormancy. However, these correlations were mainly based on mutants such as lec1, abi3, tt, ats, dog1, and the green seed mutant. We assume that these contrasting observations are based on the nature of the mutations. The induced mutants all have a defective seed maturation and consequently did not become dormant and desiccation tolerant, as these features are acquired during seed maturation. Very likely, these mutants represent artifacts that do not survive in nature. Moreover, none of the earlier mentioned seed longevity mutants (i.e. Atlig4, Atlig6, pimt1, and pimt2) have been investigated for their seed dormancy behavior. Such an analysis would reveal whether the same processes could affect seed longevity and seed dormancy, and thereby also provide insight in the underlying mechanisms. We expect that the natural variants that we used in our study display the ecologically relevant germination behavior. The GAAS5/DOG1-Fei-0 allele (Fig. S2) is special in that it has lower longevity and a lower seed dormancy when compared with Ler. The Fei-0 DOG1 allele has an opposite allelic effect as compared with the other alleles for this QTL (Cvi, Kas-2, Kond, and Sha), indicating that this allele is even weaker than the Ler allele, which is not a null allele (Bentsink et al., 2006; 2010). The Fei-0 allele of DOG1 might therefore result in a non-functional DOG1 gene, since it has a similar phenotype (lack of dormancy and low storability) to the dog1 mutant (Bentsink et al., 2006).

The colocation between seed longevity and seed dormancy QTLs can be caused by a single gene or by separate linked genes, and this remains to be investigated. Detailed analyses on GAAS5/DOG1 showed that DOG1 is controlling both seed longevity and seed dormancy. DOG1 has been cloned, and the Ler transformant containing the DOG1 allele of Cvi shows complementation of the seed dormancy phenotype of NILDOG1 (Bentsink et al., 2006) and also reduction in seed longevity (Fig. 4). This result shows that GAAS5 is actually the same locus as DOG1 and that the DOG1-Cvi allele leads to both higher seed dormancy and lower seed longevity. The molecular mechanism by which DOG1 controls seed dormancy is still unclear, however, recently it has been propose that DOG1 protein abundance in freshly harvested seeds acts as a timer for seed dormancy release (Nakabayashi et al., 2012). How the DOG1 protein affects seed storability remains to be investigated.

Chapter 2

36

The novel observation of a negative correlation between dormancy and longevity strongly suggests that seeds are able to extend their life span either by dormancy (and dormancy cycling) or by an active longevity mechanism. Selection for the different mechanisms could be based on the natural environments in which the seeds are dispersed, dry environments resulting in active longevity mechanisms and humid environments resulting in dormancy cycling during which aging damage may be prevented or repaired. The presence of loci that either improve longevity or increase seed dormancy within one accession will allow adaptive plasticity, resulting in the expression of the optimal phenotype over a range of environments (i.e. dry to humid) (Simons, 2011).

Conclusion

We performed integrated QTL analyses on natural variation that exists for seed longevity after natural aging. We used six RIL populations that were stored between 4 and 7 years at ambient conditions. The major QTLs could be confirmed by NILs that also had been stored for 5 years. Seed longevity and dormancy data revealed a negative correlation, which contrasts with the common notion in seed biology research. High storability correlated with shallow seed dormancy and low storability with high levels of seed dormancy.

Material and methods

Plant and seed materialsRILsThe six RIL populations derived from crosses between the Arabidopsis standard laboratory accession Ler and the accessions An-1, Cvi, Fei-0, Kas-2, Kond, and Sha were used. These populations have been analyzed previously for seed dormancy as described by Bentsink et al. (2010). The RILs were grown and harvested between 2003 and 2005 as described by Bentsink et al. (2010). Seeds of every RIL were stored in 6 x 13 cm cellophane flat bags at room temperature without humidity control until seed longevity was analyzed (Table I).

NILsThe NILs we used in this work were originally developed by the introgression of the identified dormancy QTL regions into Ler genetic background (Table II and S3) (Alonso-Blanco et al., 2003; Bentsink et al., 2010). In 2006, seven NILs, NILGAAS5-Sha, NILGAAS2-An-1, NILGAAS1-Cvi, NILGAAS3-Cvi, NILGAAS3-Fei-0, NILGAAS3-Kas-2, and NILGAAS5-Fei-0 were grown and harvested together with Ler. In 2007, nine NILs, NILGAAS1-Cvi,


37

NILGAAS2-An-1, NILGAAS3-Kas-2, NILGAAS5-Cvi, NILGAAS5-Fei-0, NILGAAS5-Kas, NILGAAS5-Kond, NILGAAS5-Sha, and Ler were grown and harvested. Growing and harvesting methods are as described by Bentsink et al. (2010) and storage method is as described for the RILs.

The DOG1 transformantsThe two transformants, SR3-1 and SR3-2, containing DOG1 Cvi allele into the Ler genetic background were obtained as described by Bentsink et al. (2006). These lines and their control lines (Ler and NILDOG1-Cvi) were grown as in the study of Bentsink et al. (2006) and stored as described for the RILs.

Seed longevity measurementNatural agingSeed longevity was evaluated as germination ability after several years of storage in natural condition. We have used the same seed batches that are described by (Bentsink et al., 2010). The germination percentage after dormancy release was 100% for all lines, as was described by these authors. Germination assays after aging were performed according to Joosen at al. (2010) over a period of 7 days. Briefly, six samples, 50 to 200 seeds per each, were sown on two layers of blue germination papers equilibrated with 43 mL of demi water in plastic tray (15 x 21 cm). Trays were piled and wrapped in a closed and transparent plastic bag. Germination was incubated in 22oC incubator, under continuous light (30 W m-2).

Seed longevity was confirmed by viability test with 2,3,5 triphenyl tetrazolium chloride according to the International Seed Testing Association (Moore, 1985). After 7 days of germination assay, non-germinated seeds were taken out for staining. Seeds were punched gently by sharp forceps to make staining solution easily penetrate the embryo, then placed on filter paper (Sartorius filter discs 3hw) soaked with 1% 2,3,5 triphenyl tetrazolium chloride, sealed, and incubated at 28oC for 2 days. Seeds that are viable stain red, and seeds that are dead do not stain.

Artificial aging In order to accelerate the aging, we have stored Ler, NILDOG1-Cvi, and the two independent transformants (SR3-1 and SR3-2) above a saturated NaCl solution in a closed desiccator (relative humidity of 75%) for 59 days. At 0, 13, 31, 38, 45, 52, and 59 days we have performed germination assays as described above. Two-way ANOVA analyses (P < 0.05) were performed in order to identify the differences in Gmax between the lines.

Seed dormancy measurementGermination assays during after-ripening were performed on the same seed lots as described for the seed longevity measurement. The data and methods used are presented in Bentsink et. al. (2010). Germination data were fitted to logistic curves by nonlinear

Chapter 2

38

regression analysis to determine DSDS50 as described in Bentsink et. al. (2010).

Germination parametersImages of germination assays were taken twice a day over a 7-day period. Automatic scoring and curve fitting were analyzed by the Germinator package (Joosen et al., 2010). Four parameters, Gmax, AUC, t10totS, and t50Gmax, representing the germination ability were extracted. Gmax is the final germination percentage at the end of germination assay. The AUC parameter was measured at 120 h after sowing. AUC describes the germination curve, by combining germination rate and Gmax. The germinator curve-fitting script, a part of the Germinator package, enables to calculate averages and to perform statistic Student’s t-test. At the end of the germination assay, the percentage of normal seedlings was also scored.

Integrated QTL analysesIntegrated QTL analyses of the six RIL populations were carried out for every germination parameter defined above. The analysis method was performed as described by Bentsink et al. (2010).

Acknowledgements

We thank Corrie Hanhart and Maarten Koornneef (Laboratory of Genetics, Wageningen University, The Netherlands) for storing the seeds of the RIL populations over all these years. We also thank Henk Hilhorst (Laboratory of plant physiology, Wageningen University, The Netherlands), Bas Dekkers (Department of Molecular Plant Physiology, Utrecht University, The Netherlands) and two anonymous reviewers for useful comments and critical reading of the manuscript.

Supplemental data

Supplemental information can be downloaded from http://www.wageningenseedlab.nl/thesis/tpnguyen/chapter2

Figure S1. Viability test for seeds after natural aging with 2,3,5 triphenyl tetrazolium chloride

Figure S2. Seed germination behavior of Ler, dog1 mutant and NILDOG1/GAAS5-Fei-0 during seed dry storage

Figure S3. Seed life span of Ler, NILDOG1/GAAS5-Cvi and two independent transformants (SR3-1 and SR3-2)

Figure S4. Correlation analyses between seed longevity and seed dormancy of the RILs

Table S1. Details of the six RIL populations


39

Table S2. QTLs for AUC, t10totS and t50Gmax in six populations

Table S3. Correlation coefficient of germination parameters: Gmax, AUC, t10totS and t50Gmax in six populations

Table S4. Seed longevity and dormancy phenotypes of NILs of two independent experiments

Supplemental information can be downloaded from http://www.plantphysiol.org/content/160/4/2083/suppl/DC1

Table S5. Genotypes of the RILs selected for seed longevity increasing and decreasing alleles for the Ler/Kas-2 population

Table S6. Genotypes of the RILs selected for seed longevity increasing and decreasing alleles for the Ler/Cvi population

The identification and validation of factors involved in seed longevity, a genetics and proteomics approach

Thu-Phuong Nguyen 1,2, Gwendal Cueff 3,4, Loïc Rajjou 3,4 and Leónie Bentsink 1,2

1 Wageningen Seed Laboratory, Laboratory of Plant Physiology, Wageningen University, 6708 PB Wageningen, The Netherlands; 2 Department of Molecular Plant Physiology, Utrecht University, 3584 CH Utrecht, The Netherlands; 3 INRA, Jean-Pierre Bourgin Institute, Laboratory of Excellence “Saclay Plant Sciences”, RD10, F-78002 Versailles Cedex, France; 4 AgroParisTech, Chair of Plant Physiology, 16 rue Claude Bernard, F-75231 Paris Cedex 05, France

3Chapter

Proteins involved in seed longevityChapter 3

42

Abstract

Seed germination ability during the seed life span is described as a bell-shaped curve, that consists of an after-ripening (AR) and aging (deterioration) phase. Seeds release dormancy and gain a full germination ability when after-ripened, thereafter, seeds lose germination vigor due to aging when stored for extended time-periods. Several mechanisms have been proposed to account for seed longevity, but only a limited number of genetic factors have been reported. Germination Ability After Storage (GAAS) loci were identified as quantitative trait loci that control seed longevity after natural aging in Arabidopsis thaliana. The effects of the strongest three loci, GAAS1, GAAS2, and GAAS5 were validated by near isogenic lines (NILs). NILGAAS1 and NILGAAS2 seeds are more storable while NILGAAS5 seeds are less storable compared to that of Ler. This study combines a proteomics and genetics approach to identify molecular mechanisms as well as genetic factors that are involved in seed aging by performing proteome profiling on seeds of the four genotypes at two physiological states (AR and aged). During aging, dry seed proteomes are markedly changed in a genotype specific manner, implying that different mechanisms are involved in seed longevity conferred by the three NILs. The importance of seed reserves (seed storage proteins, SSPs), and antioxidant systems, notably vitamin E, as well as the protection and maintenance of the translational machinery and energy pathways were revealed. Reverse genetics using T-DNA knock-out mutants reinforce proteomic results suggesting that SSPs (cruciferins and napins), the small ribosomal subunit RPS12C, and the NADP-dependent malic enzyme1 are involved in seed longevity.


43

Introduction

Ecologically, seeds represent a critical stage in the survival of higher plants. Seeds are also important for biodiversity conservation especially for plants producing orthodox seeds, allowing desiccation tolerance, viability and germination ability after long-term dry storage. The germination ability of seeds changes over the seed life span in dry storage conditions, which is displayed as a bell-shaped curve (Fig. 1A). The first phase after seed harvest is reflected by a period in which seeds gradually lose dormancy, during this so called after-ripening (AR) period, seeds gain full germination ability. Thereafter, dry seeds slowly deteriorate and lose vigor, which ultimately results in germination failure. It is of both ecological and agronomical relevance to understand the mechanisms governing seed vigor loss during aging in order to improve seed longevity (life span).

Seed longevity is affected by storage conditions including temperature and humidity (or seed moisture content). It has been shown that both low temperature and low seed moisture content prolong the seed life span during storage (Walters, 1998; Walters et al., 2005). Besides long-term storage (natural aging), the controlled deterioration test (CDT, artificial aging), in which seeds are stored in high temperature and relative humidity, can be used to assess seed longevity. This test is widely accepted since it allows to considerably accelerate experiments and is presumed to mimic natural aging (Tesnier et al., 2002; Rajjou et al., 2008).

Seed longevity is strongly determined by genetic components. Quantitative trait loci (QTLs) for seed longevity after both natural aging and CDT were identified in Arabidopsis (Arabidopsis thaliana) (Bentsink et al., 2000; Clerkx et al., 2004b; Nguyen et al., 2012; Chapter 2), lettuce (Lactuca sativa) (Schwember and Bradford, 2010), rice (Oryza sativa) (Miura et al., 2002; Sasaki et al., 2005) and wheat (Triticum aestivum) (Landjeva et al., 2009). In addition several Arabidopsis mutants and over-expressor lines show altered seed longevity phenotypes. Mutations in seed maturation and dormancy genes, such as leafy cotyledon1 (lec1) and abscisic acid intensitive3 (abi3) lead to dramatic losses of seed viability (Ooms et al., 1993; Clerkx et al., 2004a; Sugliani et al., 2009). Debeaujon et al. (2000) showed that testa-defective mutants including the transparent testa (tt) and the aberrant testa shape (ats) have reduced seed longevity levels. The non-dormant dog1 (delay of germination1) mutant (Bentsink et al., 2006) and DOG1-Cvi transformed into Ler (Landsberg erecta) (Nguyen et al., 2012; Chapter 2) also showed reduced seed longevity phenotypes. Vitamin E, an antioxidant preventing non-enzymatic lipid oxidation, is proven to promote seed longevity since mutants in the vitamin E synthesis genes (vte1 and vte2), show a decreased seed longevity (Sattler et al., 2004). DNA Ligase4 and 6 are necessary to maintain genome integrity, as was revealed by the high sensitivity of the lig6 and the lig6 lig4 double mutant to seed aging (Waterworth et al., 2010). Tobacco (Nicotiana tabacum) seeds, over-expressing HaHSFA9 (a heat stress


44

transcription factor isolated from sunflower Helianthus annuus), accumulate elevated heat stress protein levels, and are more tolerant to CDT-induced aging (Prieto-Dapena et al., 2006). Over-expression of Protein-L-Isoaspartate Methyltransferase1 (PIMT1) enhances seed longevity and germination vigor in Arabidopsis (Oge et al., 2008). Besides PIMT1, that repairs age induced damage to aspartyl and asparaginyl residues, methionine sulfoxide reductases (MSRs) also repair damaged proteins at methionine residues. Recently, Chaletain et al. (2013) reported that MSR protein abundance and enzymatic capacity are strongly linked to seed longevity in both Medicago (Medicago truncatula) and Arabidopsis.

Proteomics approaches have been proposed as a powerful tool for determining the biological roles and functions of individual proteins and identifying the molecular mechanisms of seed germination, vigor and viability during aging, in which the role of stored mRNA and translation potential are essential. Rajjou et al. (2004; 2007) reported that de novo mRNA synthesis is not required for seed germination, and that Arabidopsis seeds are able to germinate in the presence of the transcription inhibitor α-amanitin. Moreover, the quality of the stored mRNAs is important for seed vigor. These studies also showed that translation is required for seed germination since cycloheximide, a translation inhibitor, blocks seed germination. The translational potential of dry seeds is reduced during aging, thus the maintenance and repair of the a functional translational machinery is important for seed longevity (Rajjou et al., 2008). Post-translational modification (PTM) regulation in dry seeds are reviewed to play central role in seed dormancy release, metabolism resumption, and aging processes (Arc et al., 2011). These authors also demonstrated that the elevated accumulation of oxidation (carbonylation) in dry seed is associated with aging, and might induce loss-of-function protein and enzyme properties. Therefore, detoxification of reactive oxygen species (ROS) that result in oxidation stress and maintenance of redox homeostasis are crucial for seed vigor (Rajjou et al., 2008). The detoxification of toxic compounds, e.g. cyanine, that accumulate during seed storage and germination, by β-mercaptopyruvate sulfurtransferase (MST) showed to be related to seed longevity since the abundance of this enzyme was reduced during seed aging (Rajjou et al., 2008).

The above-mentioned studies revealed that seed longevity is a complex trait, which requires a better understanding of the underlying molecular mechanisms. In our previous study we identified twelve GAAS (Germination Ability After Storage) loci controlling seed longevity after natural aging in Arabidopsis (Nguyen et al., 2012; Chapter 2). In order to investigate seed longevity mechanisms of the GAAS loci, here, proteome analyses were performed on dry seeds at two physiological states, after-ripened (AR) and 4 year-old (aged) seeds of four genetic lines, whose seed longevity behavior after natural aging was validated. The genotypes used are the background line Ler and three near isogenic lines (NILs) that contain introgression fragments of Cape


45

Verde Islands (Cvi), Antwerp (An-1) and Shakdara (Sha) accessions, respectively, at the position of the earlier identified seed longevity QTL, NILGAAS1-Cvi, NILGAAS2-An-1, and NILGAAS5-Sha. Proteome profiles were compared within the genotypes for the two physiological states (aged vs. AR) and between the genotypes (NILs vs. Ler) for the AR state, all in a pair-wise manner (Fig. 1). Differentially accumulated protein spots found in the mentioned comparisons were excised from the 2D gels and submitted to mass spectrometry for identification. A subset of candidate proteins were selected to confirm their role in seed longevity by phenotypic analyses on T-DNA knock-out lines. Our study shows the power of combining genetics and proteomics to identify and validate factors that are important for seed longevity.

Figure 1. Schematic presentation of the performed experimentA bell-shaped curve showing the germination ability (germination %) along seed dry storage over the two physiological states; after-ripened (AR) and aged (A). Total soluble proteomes were isolated from dry seeds and separated by 2D PAGE. Comparisons were made between the mentioned physiological states within the genotypes (indicated by the solid arrows); and between genotypes Ler and NILs (NILGAAS1, NILGAAS2and NILGAAS5) at the AR state (indicated by the dashed arrows) (B).

Results and discussion

Effect of natural aging on seed germination abilityThe effects of seed dry storage on the germination ability of AR and 4-year old (aged) seeds of four genotypes were investigated. These genotypes are Ler, NILGAAS1, NILGAAS2, and NILGAAS5. Upon storage all genotypes showed a significant reduction in germination percentage, however with a different level of sensitivity to aging (Fig. 2). NILGAAS1 and NILGAAS2 had a better seed longevity (89.3% and 79.6%, respectively) compared with Ler (60.7%), whereas NILGAAS5 was less storable (37.7%). These result confirmed the seed longevity phenotypes described by Nguyen et al. (2012; Chapter 2).


46

Figure 2. Seed germination ability after seed dry storageSeed germination (%) of Ler, NILGAAS1, NILGAAS2, and NILGAAS5 were analyzed for after-ripened (open bars) and naturally aged seeds after 4 years of storage (filled bars). Averages of four biological replicates and standard errors are presented. The asterisks indicate significant differences between aged NILs and there genetic background Ler (P < 0.05).

Seed dry storage affects the proteome in a genotype specific mannerProteomics profiling for AR and aged seeds have been performed to identify mechanisms associated with the loss of germination ability in dry seeds during storage. Total soluble protein extracts of all the samples were separated using two-dimension polyacrylamide gel electrophoresis (2D PAGE). Thus seven pair-wise comparisons were made. The four pair-wise comparisons between the two physiological states (aged vs. AR) for each genotype could reveal proteins that are affected by aging (57 for Ler, 89 for NILGAAS1, 109 for NILGAAS2, and 126 for NILGAAS5). The three NILs mentioned above contain introgression fragments at three genome positions and express diverse levels of seed longevity. The variation in seed longevity could result from proteome variation already present in the AR seeds, which can be revealed by the pair-wise comparisons between the genotypes and Ler. We identified 51, 16 and 11 genotype specific proteins spots of which 8, 2 and 2 spots (Table I) were overlapping with the aged vs. AR comparison for respectively NILGAAS1, NILGAAS2 and NILGAAS5. A total of 309 differentially expressed protein spots were detected in the seven pair-wise comparisons (Fig. 1B) despite the fact that we studied the proteome of metabolically quiescent dry seeds. These proteins can either reflect causes or consequences of aging, therefore protein identification was performed to identify the factors involved, by both reference maps and sequencing. Approximately 70% of the total number of differentially expressed genes could be identified. Principle component analysis (PCA) on the 309 spots separated the samples into two groups which represent the two physiological states (AR and aged seeds; Fig. 3). The time component (storage) accounts for 24% of the variation. The second component, that explains 11% of variation, was the genotype. NILGAAS1 was the most distinct genotype, the location of its aged seeds in the PCA might reflect its better longevity since it was more closely located to the non-aged (AR) samples on the


47

Tabl

e I.

Geno

type

spec

ific p

rote

ins t

hat w

ere

iden

tifie

d in

bot

h ph

ysio

logi

cal s

tate

and

gen

otyp

e co

mpa

riso

ns fo

r NIL

GAAS

1, N

ILGA

AS2

and

NIL

GAAS

5 (F

ig. 9

)Th

e ta

ble

disp

lays

pro

tein

spo

ts b

ased

on

the

seve

n co

mpa

riso

ns. S

pot I

D, th

e ge

ne c

orre

spon

ding

to th

e pr

otei

n un

derly

ing

the

spot

, mol

ecul

ar w

eigh

t (M

W in

kD

) and

the

theo

retic

al (T

h) a

nd e

xper

imen

tal (

Exp)

isoe

lect

ric p

oint

(pI)

are

pre

sent

ed, r

espe

ctiv

ely;

n.a

mea

ns n

ot a

vaila

ble.

Fur

ther

mor

e th

e re

lativ

e ab

unda

nce

(fol

d ch

ange

) of t

he s

pots

in b

oth

type

s of

com

pari

sons

(Fig

. 1; p

hysi

olog

ical

sta

te a

nd g

enot

ype)

is in

dica

ted.

Pos

itive

fold

cha

nges

indi

cate

hig

her

abun

danc

es,

and

nega

tive

low

er a

bund

ance

s. Sp

ots

in b

old

exhi

bit s

eed

long

evity

up

or s

eed

long

evity

dow

n pr

otei

n pr

ofile

. Fol

d ch

ange

s in

bol

d in

dica

te th

e is

sta

tistic

ally

si

gnifi

cant

cha

nges

. NG1

sta

nds

for N

ILGA

AS1,

NG2

for N

ILGA

AS2

and

NG5

for N

ILGA

AS5.

Spo

ts th

at w

ere

iden

tifie

d ba

sed

on c

ompa

riso

n to

the

refe

renc

e pr

otei

n m

ap (h

ttp:

//w

ww

.seed

-pro

teom

e.co

m) a

re la

belle

d w

ith R .

Spot

IDGe

nePr

otei

n

MW

(kD)

pIRe

lativ

e ab

unda

nce

(fol

d ch

ange

)

ThEx

pTh

Exp

Phys

iolo

gica

l sta

te:

Aged

vs.

ARGe

noty

pe A

R:N

IL v

s. Le

r

Ler

NG1

NG2

NG5

NG1

NG2

NG5

NIL

GAAS

1ID

1279

n.a

n.a

n.a

22.8

2n.

a8.

68-1

.0 2

.5-1

.1 1

.5-2

.6 1

.3-1

.3eI

D02

55 R

AT1G

0389

0Cu

pin

fam

ily p

rote

in49

.67

29.3

35.

455.

63-1

.0 1

.9 1

.0 1

.0-2

.3-1

.0-1

.0ID

0426

n.a

n.a

n.a

65.5

1n.

a5.

41 1

.2 1

.8 1

.2 1

.4-1

.7-1

.4-1

.2ID

1104

AT1G

0388

0Cr

ucife

rin B

50.5

627

.38

7.0

5.61

1.1

1.5

1.0

1.1

1.9

1.1

1.0

AT5F

3559

020

S pro

teas

ome a

lpha

subu

nit

27.2

95.

66AT

1G03

890

Cupi

n fa

mily

pro

tein

49.6

75.

45eI

D013

8 R

AT4G

2852

0Cr

ucife

rin C

ww

58.2

425

.40

6.99

5.81

1.9

1.5

1.5

1.6

2.3

1.3

1.0

ID09

94AT

1G54

870

Oxid

ored

ucta

se fa

mily

pro

tein

36.7

632

.75

8.76

5.65

2.3

1.7

2.0

2.4

1.5

1.2

1.3

AT4G

2852

0Cr

ucife

rin C

58.2

46.

99ID

0715

AT1G

0388

0Cr

ucife

rin B

50.5

647

.74

7.00

6.02

-1.2

-1.6

-1.3

-1.6

1.6

-1.0

1.1

AT1G

7496

0Fa

tty ac

id b

iosy

nthe

sis 1

57.6

07.

93ID

0537

AT2G

1417

0Al

dehy

de d

ehyd

roge

nase

6B2

53.4

058

.77

5.79

5.68

-2.0

-3.1

-2.3

-2.0

1.5

1.3

1.1

AT5G

0867

0AT

P sy

ntha

se b

eta c

hain

159

.63

6.53

AT5G

0868

0AT

P sy

ntha

se b

eta c

hain

59.8

66.

45AT

5G08

690

ATP

synt

hase

bet

a cha

in 2

59.7

16.

60N

ILGA

AS2

ID04

58AT

3G20

050

T-co

mpl

ex p

rote

in 1

alph

a sub

unit

59.2

363

.81

6.22

5.87

-1.2

1.1

1.5

1.1

-1.4

-2.0

-1.2

ID09

55n.

an.

an.

a38

.44

n.a

7.29

1.4

-1.7

-2.1

-1.8

1.4

1.7

1.5

NIL

GAAS

5eI

D022

8AT

5G19

510

Elon

gatio

n fa

ctor

EF1

B24

.20

43.2

44.

173.

88-1

.6 1

.3 1

.2 1

.7-1

.3-1

.1-1

.6ID

1146

AT2G

3167

0Un

know

n pr

otei

n28

.86

25.3

66.

965.

10 1

.1 1

.3 1

.4 1

.8-1

.0-1

.3-1

.5


48

time component. This time component is also reflected in NILGAAS5 which is the least storable genotype and is located most left of all the genotypes (Fig. 3).

Figure 3. Principle component analysis (PCA) for proteome profiles of Ler, NILGAAS1, NILGAAS2, and NILGAAS5The PCA was performed on the differentially accumulated protein spots in the seven comparisons (n = 309).

The aged vs. AR physiological state comparisons in the four genotypes showed a total of 247 protein spots whose abundance significantly changed (P-value ≤ 0.05) upon aging (Fig. 4A). A large number of the 247 spots are genotype specific (15 for Ler, 41 for NILGAAS1, 46 for NILGAAS2, and 64 for NILGAAS5) (Fig. 4A). The three genotype comparisons led to the identification of 74 altered protein spots (Fig. 4B). Most of them were unique for the genotypes (47 for NILGAAS1, 12 for NILGAAS2, and 11 for NILGAAS5) (Fig. 4B). The four spots in common between NILGAAS1, and NILGAAS2 (Fig. 4B, Table S3) might play a role in seed longevity, because both genotypes are better storable than Ler. Generally, these results indicate that seed dry storage affects the proteome in a genotype specific manner, suggesting that different genetic pathways are involved in seed aging. However, one cannot exclude that the differences in the AR seed proteomes are caused by the nature of the introgression in the NILs and are therefore unrelated to seed longevity.

In the following sections we will first discuss proteins displaying an altered abundance in response to aging in more than one genotype since they likely reflect a general effect. After that we will reveal more details on genotype specific changes. Overall we will only deal with the identified proteins.


49

Figure 4. The number of protein spots that changed in abundanceComparison between the two physiological states, Aged versus after-ripened (AR), within the genotype for Ler, NILGAAS1, NILGAAS2, and NILGAAS5 (n = 247) (A). Genotype comparison at the AR state between NILs and Ler (n = 74) (B).

The role of seed storage proteins in agingA large portion of the identified spots were seed storage protein (SSP) 12S globulin fragments, a predominant type of seed storage protein referred to as cruciferins (Pang et al., 1988). Cruciferins in Arabidopsis are encoded by four paralogous genes AT5G44120 (CRUA), AT1G13880 (CRUB), AT4G28520 (CRUC) and AT1G03890 (CUPIN). Their abundance in Arabidopsis seeds and wide range of PTMs can result in marked changes of SSPs during aging (Job et al., 2005; Wan et al., 2007).

Muntz et al. (2001) suggested that SSPs contributed to seed germination vigor and support early seedling growth when mobilized upon germination. A set of single, double and triple knock-out lines for CRUA, CRUB and CRUC (Withana-Gamage et al., 2013) (Table S2) was analyzed in order to study the role of cruciferins in seed longevity. The cruciferin single mutants that lack one of the isoforms did not differ in seed longevity compared to wild type Col (Fig. 5). Only the double mutant lacking both CRUA and CRUC showed a reduced seed longevity. CRUB is very poorly transcribed and therefore the lowest abundant CRU isoform (Withana-Gamage et al., 2013), which explains why double mutants that have CRUB eliminated in addition to CRUA or CRUC did not have a defective seed longevity. The triple crua crub cruc mutant was very sensitive to artificial aging (Fig. 5). The role of cruciferins in seed longevity cannot completely be explained by the reduced protein levels since the crua cruc double mutant has wild type protein levels in both the endosperm and the embryo. It might be that the localization and the distribution of the proteins plays a role since the triple crua crub cruc mutant contains very small protein storage vesicles with very little protein therein (Withana-Gamage et al., 2013). SSPs are subjected to a wide range of PTMs (Job et al., 2005; Wan et al., 2007), suggesting that the effects of SSPs on seed longevity are regulated by PTMs. These results have underlined the importance of


50

cruciferins in seed longevity, but how cruciferins play a role in seed longevity mechanisms requires further investigation.

Figure 5. The effect of seed storage proteins (SSPs) on seed longevitySeed longevity presented as germination (%) of different SSP knock-out lines was measured after 10 days of artificial aging treatment. The lines include single (crua; aBC, crub; AbC and cruc; ABc), double (crubcruc; Abc, cruacruc; aBc; and cruacrub; abC) and triple (cruacrubcruc; abc) knock-out lines of cruciferins and a RNAi Napin line that is depleted in napins.

The effect of aging on translationProtein translation is essential for seed germination, since the presence of the translation inhibitor cycloheximide abolished seed germination (Rajjou et al., 2004). Furthermore, aged seeds were strongly affected in their translation capacity (Rajjou et al., 2008). These authors also demonstrated that many proteins involved in protein metabolism were highly carbonylated in deteriorated seeds. Consistent with previous studies, we observed that the elongation factor EF1B family protein levels were down accumulated in aged compared to AR seeds (Table II). Moreover, spot ID1345 containing the ribosomal protein 40S subunit RPS12C has a reduced abundance after storage in seeds of Ler, NILGAAS1 and NILGAAS2 (Fig. 6B, Table II). Protein metabolism plays a role in PTMs, such as protein folding, protein translocation, and protein conformation, thus impairment of protein metabolism related proteins might lead to malfunctional PTMs. However, the T-DNA knock-out line of RPS12C showed an increased seed longevity (Fig. 7). It is known that ribosomal proteins are encoded by multiple genes (Carroll et al., 2008) and there are three genes encoding for the RPS12 subunit. Therefore, the depletion of RPS12C might lead to the activation of the other two genes, and possibly a more efficient RPS12 production that overcomes the effect of aging. Seed longevity analyses for these other genes encoding RPS12, including double and triple mutants and the quantification of


51

RPS12 transcript and protein expression might support this hypothesis.

Figure 6. Presentation of proteins whose abundance is affected during agingSilver nitrate-stained 2D gel of total soluble proteins from dry seeds (A). The indicated areas on the gel (1 and 2) are enlarged in the B and C panels. Area 1 selected on the gel depicts the abundance of protein spot ID1345 that contains RPS12C and TPX1 proteins for the four genotypes (Ler, NILGAAS1, NILGAAS2, and NILGAAS5) at two physiological states (aged and after-ripened [AR]) (B). Area 2 shows the change in abundance of protein spot ID0667, corresponding to VTE1 protein for the four mentioned genotypes at two mentioned physiological states (C). Arrows indicate the position of the discussed proteins.

Energy metabolism related pathways in seed longevityEnergy metabolism from carbohydrates is important for germinating seeds, since it provides energy and intermediate metabolites. The energy related metabolic pathways are glycolysis, oxidative pentose phosphate pathway (OPP), fermentation, tricarboxylic acid cycle (TCA), glyoxylate cycle, and electron transport chain (Fig. 8). Several enzymes involved in these pathways were altered in the seed proteome upon storage.

Glycolytic pathwayThe UDP-1 glucose uridylyltransferase (UGP) catalyzes a reversible reaction where glucose-1-phosphate (Glu1P) is converted to uridine diphosphate glucose (UDPGlu). UGP proteins behaved differently in NILGAAS2 and NILGAAS5. UGP2 present in spot ID0621 was 1.5 times more abundant in aged NILGAAS2 while both UGP1 and UGP2 in spot ID0603 were 1.5 times less abundant in aged NILGAAS5 in comparison with AR seeds (Table II). This opposite behavior might reflect the difference in seed longevity since NILGAAS5 seeds were significantly less storable than those of NILGAAS2. Fait et al. (2006) showed that during seed germination, the levels of hexose phosphates were elevated, indicating the importance of hexose phosphate in germination. In order to reveal the role of UGPs in seed longevity, the knock-out mutants were analyzed. A


52

homozygous knock-out could only be obtained for UGP1, but this mutant did not show a seed longevity phenotype (Fig. 7), suggesting that UGP1 is not important for seed longevity or that UGP1 and UGP2 act redundantly.

Figure 7. Seed longevity of T-DNA knock-out mutants after artificial agingSeed longevity presented as germination (%) of T-DNA knock-out mutants of candidate genes was measured after 8 days of artificial aging treatment. Standard errors are calculated on four biological replicates. The asterisks indicate significant differences between mutants and Col wild type (P ≤ 0.05).

The abundance of two spots ID0832 and ID0997 corresponding to glyceraldehyde-3-P dehydrogenase C1 (GAPC1) and C2 (GAPC2) were higher in aged than in AR seeds (Table II). GAPCs catalyzes the conversion of glyceraldehyde 3-phosphate (G3P) to D-glycerate 1,3-bisphosphate (DPGA), that is further converted to 3-phosphoglycerate (3PGA). The fact that aged seeds contain an increased level of GAPCs suggests that natural aging could cause oxidative stresses to dry seeds, since Arabidopsis cells that undergo oxidative stress have substantially increased 3PGA levels (Baxter et al., 2007). Modification of the glycolytic pathway was also observed in dry seeds treated with CDT (Rajjou et al., 2008) where GAPCs protein levels were significantly increased compared with control seeds.

OPPOPP is the alternative pathway of carbohydrate oxidation and consists of an oxidative and non-oxidative phases. In the oxidative phase NADPH is generated when glucose 6-phosphate (Glu6P) is converted to ribulose 5-phosphate. In the non-oxidative phase ribulose 5-phosphate is used to generate fructose 6-phosphate (Fru6P) and G3P, and this process is controlled by transketolase and transaldolase. Identified spot eID0214 and eID0096 correspond to the transketolase genes AT2G45290 and AT3G60750, respectively. The abundance of these proteins was reduced in aged seeds of all


53

genotypes, except for eID0214 that did not change in NILGAAS1 (Table II). The reduction of transketolase could result in the loss of germination in aged seeds by a decline in NADPH pool needed for the maintenance of redox-homeostasis, and a reduction in ribulose 5-phosphate as a source for nucleotide synthesis upon germination.

Figure 8. Schematic representation of energy metabolism related pathways affected during seed storageGlycolysis, oxidative pentose phosphate (OPP), fermentation, tricarboxylic acid cycle (TCA), glyoxylate, and electron transport chain (ETC) pathways are shown. Identified enzymes are indicated in back box. UGP, UDP-1 glucose uridylytransferase; GAPC, Glyceraldehyde3-P dehydrogenase C; TKL, Transketolase; PDC, Pyruvate decarboxylase; MDH, Malate dehydrogenase; ME, NADP-dependent malic enzyme; ATPS, ATP synthase β chain; Fru6P, Fructose 6-phosphate; Glu6P, Glucose 6-phosphate; G3P, Glyceraldehyde 3-phosphate; PEP, Phosphoenolpyruvate; Rib5P, Ribulose 5-phosphate; OAA, Oxaloacetate.

FermentationPyruvate, the end product of the glycolysis and OPP, can go to the TCA cycle or to fermentation reactions. Pyruvate decarboxylase (PDC) catalyzes the first step of the fermentative pathway where pyruvate is converted into ethanol or lactate. We observed a decreased abundance of PDC2 (eID0275) in aged compared with AR seeds of the three NILs (Table II). PDC2 reduction likely inhibits the fermentation. The role of PDC2 in seed longevity was examined


54

Tabl

e II

. Diff

eren

tially

exp

ress

ed p

rote

ins

upon

agi

ng th

at a

re re

late

d to

tran

slat

ion,

ene

rgy

met

abol

ism

, rea

ctiv

atio

n of

cel

l act

ivity

, red

ox h

omeo

stas

is, a

nd A

BA

sign

allin

g Th

e ta

ble

disp

lays

pro

tein

spo

ts b

ased

on

the

seve

n co

mpa

riso

ns. S

pot I

D, th

e ge

ne c

orre

spon

ding

to th

e pr

otei

n un

derly

ing

the

spot

, mol

ecul

ar w

eigh

t (M

W in

kD

) and

the

theo

retic

al (T

h) a

nd e

xper

imen

tal (

Exp)

isoe

lect

ric p

oint

(pI)

are

pre

sent

ed, r

espe

ctiv

ely;

n.a

mea

ns n

ot a

vaila

ble.

Fur

ther

mor

e th

e re

lativ

e ab

unda

nce

(fol

d ch

ange

) of t

he sp

ots i

n bo

th ty

pes o

f com

pari

sons

(Fig

. 1; p

hysi

olog

ical

stat

e an

d ge

noty

pe) i

s ind

icat

ed. P

ositi

ve fo

ld ch

ange

s ind

icat

e hi

gher

abu

ndan

ces,

and

nega

tive

low

er a

bund

ance

s. Fo

ld ch

ange

s in

bold

indi

cate

the

stat

istic

ally

sign

ifica

nt ch

ange

s. N

G1 st

ands

for N

ILGA

AS1,

NG2

for N

ILGA

AS2,

and

NG5

for N

ILGA

AS5.

Sp

ots t

hat w

ere

iden

tifie

d ba

sed

on co

mpa

riso

n to

the

refe

renc

e pr

otei

n m

ap (h

ttp:

//w

ww

.seed

-pro

teom

e.co

m) a

re la

belle

d w

ith R .

Spot

IDGe

nePr

otei

n

MW

(kD)

pIRe

lativ

e ab

unda

nce

(fol

d ch

ange

)

ThEx

pTh

Exp

Phys

iolo

gica

l sta

te:

Aged

vs.

ARGe

noty

pe A

R:N

IL v

s. Le

r

Ler

NG1

NG2

NG5

NG1

NG2

NG5

Tran

slat

ion

ID13

45AT

2G32

060

Ribo

som

al p

rote

in 4

0S su

buni

t15

.33

16.2

55.

705.

07-1

.7-2

.5-1

.5 1

.1-1

.1-1

.3-1

.5

ID06

67 R

AT1G

5772

0El

onga

tion

fact

or E

F1Bγ

46.4

049

.95

5.40

5.54

-1.4

-1.1

-1.5

-1.5

1.2

1.4

1.2

eID0

228

RAT

5G19

510

Elon

gatio

n fa

ctor

EF1

B24

.20

43.2

44.

173.

88-1

.6 1

.3 1

.2 1

.7-1

.3-1

.1-1

.6

ID08

85AT

1G30

230

Elon

gatio

n fa

ctor

EF1

β28

.77

38.6

74.

363.

98-1

.3-1

.0-1

.5-1

.7-1

.2 1

.1 1

.2

ID12

26AT

1G07

930

Elon

gatio

n fa

ctor

EF1

α49

.50

25.1

19.

647.

81-1

.2-1

.6 1

.0-1

.1 1

.4 1

.1-2

.1

AT5G

6039

0El

onga

tion

fact

or E

F1α

49.5

09.

64

Ener

gy m

etab

olis

m

ID06

21AT

5G17

310

UGP-

1 gl

ucos

e urid

ylytra

nsfe

rase

251

.92

55.5

35.

805.

80 1

.3 1

.5 1

.5 1

.3-1

.1-1

.0-1

.1

ID06

03AT

3G03

250

UGP-

1 gl

ucos

e urid

ylytra

nsfe

rase

151

.74

56.9

45.

985.

71 1

.1-1

.1-1

.4-1

.5 1

.3 1

.4 1

.3

AT5G

1731

0UG

P-1

gluc

ose u

ridyly

trans

fera

se 2

51.9

25.

80

ID08

32AT

1G13

440

Glyc

eral

dehy

de-3

-P d

ehyd

roge

nase

C2

36.9

143

.15

7.22

6.05

1.9

1.7

1.7

1.3

-1.0

1.2

1.4

AT3G

0412

0Gl

ycer

alde

hyde

-3-P

deh

ydro

gena

se C

136

.91

7.15

ID09

97AT

1G13

440

Glyc

eral

dehy

de-3

-P d

ehyd

roge

nase

C2

36.9

133

.67

7.22

5.72

1.7

1.7

1.9

2.7

1.2

1.1

-1.1

AT3G

0412

0Gl

ycer

alde

hyde

-3-P

deh

ydro

gena

se C

136

.91

7.15

eID0

214

AT2G

4529

0Tr

ansk

etol

ase

79.9

281

.80

6.55

5.77

-2.0

-1.8

-2.3

-1.8

-1.0

-1.1

-1.1


55

Tabl

e II

cont

inue

eID0

096

AT3G

6075

0Tr

ansk

etol

ase

79.9

778

.47

6.32

5.45

-1.6

-1.9

-1.6

-1.9

1.2

1.1

1.2

eID0

275

AT5G

5496

0Py

ruva

te d

ecar

boxy

lase

265

.82

64.3

55.

845.

54-1

.7-1

.5-1

.7-2

.0 1

.2 1

.4 1

.2

ID11

80AT

1G04

410

Mal

ate d

ehyd

roge

nase

135

.57

25.2

06.

515.

64 1

.6-1

.0 1

.5 2

.0 1

.3-1

.2-1

.2

ID04

48AT

2G19

900

NADP

-dep

ende

nt m

alic

enz

yme

164

.28

n.a

6.73

n.a

-1.3

-1.7

-1.7

-1.7

1.1

-1.1

-1.0

ID05

26AT

5G08

670

ATP

synt

hase

β ch

ain

159

.63

59.3

46.

535.

35 1

.2 1

.3 1

.2 1

.6 1

.0 1

.0-1

.1

AT5G

0868

0AT

P sy

ntha

se β

chai

n59

.86

6.45

AT5G

0869

0AT

P sy

ntha

se β

chai

n 2

59.7

16.

60

ID05

35AT

5G08

680

ATP

synt

hase

β ch

ain

59.8

658

.73

6.45

5.49

-1.3

-1.5

1.2

1.3

1.0

-1.3

-1.4

ID05

38AT

5G08

670

ATP

synt

hase

β ch

ain

159

.63

58.2

26.

535.

51-1

.3-1

.3-1

.5-1

.4 1

.1 1

.3 1

.1

ID05

56AT

5G08

670

ATP

synt

hase

β ch

ain

159

.63

57.4

46.

535.

49 1

.0-1

.6 1

.1 1

.0 1

.9 1

.4 1

.4

Reac

tivat

ion

of ce

ll

ID06

93AT

2G36

880

Met

hion

ine

aden

osyl

tran

sfer

ase

342

.50

46.0

36.

095.

73-1

.1 1

.6-1

.3-1

.4 1

.3 1

.1 1

.2

ID06

94 R

AT4G

0185

0M

ethi

onin

e ad

enos

yltr

ansf

eras

e 2

43.2

547

.58

5.94

5.65

-1.1

-1.0

-1.6

-1.1

-1.0

1.3

1.3

Redo

x ho

meo

stas

is

ID13

45AT

1G65

980

Thio

redo

xin-

depe

nden

t per

oxid

ase

117

.43

16.2

55.

005.

06-1

.7-2

.5-1

.5 1

.1-1

.1-1

.3-1

.5

ID06

67AT

4G32

770

Vita

min

E d

efic

ient

154

.72

49.9

56.

265.

54-1

.4-1

.1-1

.5-1

.5 1

.2 1

.4 1

.2

ABA

sign

allin

g

eID0

200

AT1G

4389

0Re

spon

sive

to a

bsci

sic a

cid

1823

.53

18.5

35.

435.

79 1

.2 1

.4 1

.7 1

.7-1

.1-1

.4-1

.3


56

by reverse genetics. The pdc2-1 and pdc2-2 mutants did not perform statistically different from the Col wild type (Fig. 7), which might be a result of redundant functions of PDCs. The double mutant pdc1 pcd2 might reveal the effect of fermentation on seed longevity.

TCA and glyoxylate cycleWe did not observe significant changes during aging in TCA cycle enzymes except for aged NILGAAS5 seeds. The expression of malate dehydrogenase (MDH, ID1180) increased two fold and that of the NADP-dependent malic enzyme1 (NADP-ME1, ID0448) decreased 1.7 times (Table II). MDH catalyzes the reversible oxaloacetate to malate reaction, while NADP-ME1 catalyzes the reversible pyruvate to malate reaction. The role of the pyruvate-malate turnover in seed longevity was confirmed by analyzing the T-DNA knock-out line of NADP-ME1, which is significantly less storable than Col (Fig. 7).

Rajjou et al. (2006) demonstrated that protein abundance of isocitrate lyase, a key enzyme in the glyoxylate cylcle, was increased during seed germination, and decreased according to the time of CDT treatment. This association of isocitrate lyase and seed germination vigor suggests the fundamental role of the glyoxylate cycle for seed germination and seed vigor. However, we did not identify enzymes related to this pathway.

Electron transport chainATP is synthesized in the mitochondria by the ATP synthase β using reduced equivalents derived from the TCA cycle (Fig 8). The ATP synthase β subunit is encoded by a three member gene family (AT5G08670, AT5G08680 and AT5G08690). Several ATP synthase β chain spots changed in abundance during aging. The abundance of spot ID0526 including the three isoforms was higher in aged NILGAAS5; while spots ID0556 (AT5G08670) and ID0535 (AT5G08680) were reduced in aged NILGAAS1; and spot ID0538 (AT5G08670) was less abundant in aged NILGAAS2 (Table II). Fait et al. (2006) showed that during seed germination, the levels of most sugars variously decline while Fru6P, Glu6P, and TCA cycle intermediates dramatically increase, indicating the need for ATP synthesis for germination vigor. The affected glycolysis and TCA pathways due to aging and the increase of proteins for ATP synthesis might cause energy leakage, which explains the loss of viability in the aged seeds of NILGAAS5.

Reactivation of cellular activityWe have observed protein changes of the methionine adenosyltransferase (MAT) enzyme family in both NILGAAS1 and NILGAAS2 after aging. MAT takes part in S-adenosylmethionine (Ado-Met) biosynthesis, that is generally important for reactivating cellular activity in germinating seeds (Ravanel et al., 1998; Gallardo et al., 2002). MAT3 (ID0693) abundance increased by 1.6 times in aged compared to AR


57

seeds of NILGAAS1 (the most storable line), but MAT2 (ID0694) was more expressed (1.6 times) in AR than in aged seeds of NILGAAS2 (Table II). This result suggests possible genotype dependent pathways related to Ado-Met metabolism and therefore to seed longevity.

Redox homeostasis and antioxidants in seed longevitySeed storage and germination are coupled to extensive changes in the redox state of seed proteins and even in the dry state seed proteins are subjected to various types of PTMs which also include redox modifications (Arc et al., 2011; Rajjou et al., 2012). Thioredoxin has a vital role in redox processes, in which it transforms essential proteins from the oxidized to the reduced form, and thereby retrieves the molecular function of those proteins. The redox regulation might maintain the low metabolic rate in dry seeds and allow metabolism to resume during germination, since it can increase the solubility and remobilization of SSPs. Thioredoxin-dependent peroxidase 1 (TPX1) (ID1345) protein accumulation declined in aged compared to AR seeds of Ler, NILGAAS1, NILGAAS2 (Fig. 6B, Table II). TPX1 is a thioredoxin-dependent peroxidases type II B (Prx IIB), one of the six isoforms in this family, that has a wide range of redox buffering activities (Rouhier and Jacquot, 2005). Thus, TPX1 might be very important for retaining the redox balance during aging and germination. In order to investigate the role of TPX1 we have analyzed the tpx1 mutant for seed longevity. This revealed a similar seed longevity phenotype as that of Col (Fig. 7). The lack of a phenotype for tpx1 could be due to the nature of the T-DNA insertion (3’-UTR region of the gene) (Table S2) or due to redundancy of enzymatic antioxidant systems in seeds, thus missing one might not have an obvious effect.

Vitamin E is another antioxidant important for seed longevity since the vitamin E deficient1 (vte1) mutant seeds were more sensitive to artificial aging than those of wild type (Sattler et al., 2004). In our data set, the protein (ID0667) abundance of VTE1 was lower in aged seeds of NILGAAS2 and NILGAAS5 (Fig. 6C and Table II), in agreement with previous studies suggesting a role of VTE1 in seed longevity.

ABA signaling and seed longevityWe found a higher abundance of the responsive to ABA18 (RAB18) protein (eID0200) in aged compared to AR seeds of NILGAAS2 and NILGAAS5 (Table II). RAB18 was proposed to be involved in seed longevity and stress tolerance in cabbage (Brassica oleracea) (Soeda et al., 2005) and in Arabidopsis (Rajjou et al., 2008) due to its association with membrane stability and/or protein structure maintenance. Soeda et al. (2005) showed that the mRNA of the RAB18 cabbage homolog was increased during seed maturation and re-induced during a slow and warm drying treatment of primed seeds to improve their storability. Accumulation of RAB18 was very high in dry mature Arabidopsis


58

seeds and progressively disappeared in CDT treated seeds (Rajjou et al., 2008). This correlation was opposite to what we found, which might be due to the presence of different RAB18 isoforms. In addition, a recent study in Arabidopsis on the role of seed-specific dehydrins demonstrated that down-regulation of LEA14, XERO1 and RAB18 reduced seed survival in the dry state (Hundertmark et al., 2011). However, Wechsberg et al. (1994) did not recognize any correlation between protein accumulation of RAB group and seed longevity in Ranunculus sceleratus. These contradicting findings suggest that the alteration in RAB18 protein abundance might be a consequence of the treatment rather than the cause of the reduced seed survival. Therefore, the role of RAB18 and ABA signaling in seed aging is remains to be investigated.

Genotype specific changes in proteome profilesThe four genotypes, Ler and NILGAAS1, NILGAAS2 and NILGAAS5, could specify different genetic pathways involved in seed longevity, reflected by proteins that are significantly altered in a genotype specific manner. Genotype specific proteins include proteins differentially expressed in only one genotype derived from physiological comparisons (aged vs. AR seeds), and proteins identified in genotype comparisons (AR seeds of NILs vs. Ler) (Fig. 4). Overlapping spots of both comparison types are presented in Figure 9 and Table III (8 for NILGAAS1, 2 for NILGAAS2 and 2 for NILGAAS5). These spots exhibit two profiles of interest, which are longevity up (LU) and longevity down (LD). LU is displayed by proteins that are more abundant in AR seeds of the better storable lines and are reduced in their abundance upon aging (higher in AR than in aged seeds). LD are proteins that are down accumulated in AR seeds of the better storable lines and have increased abundance upon aging (lower in AR than in aged seeds). Below, we will discuss and hypothesize the possible seed longevity mechanisms in the three NILs based on genotype specific proteins.

Figure 9. The number of genotype specific protein spotsGenotype specific proteins include proteins differentially expressed in only one genotype derived from the physiological state comparisons (aged vs. after-ripened [AR] seeds), and proteins identified in genotype comparisons (AR seeds of NILs vs. Ler). Genotype specific spots for NILGAAS1 (A), NILGAAS2 (B), and NILGAAS5 (C) are presented.


59

Mechanism related to seed aging expressed in NILGAAS1NILGAAS1 is the most storable line of the four tested genotypes (Fig. 2). NILGAAS1 has 41 (Fig. 4A) unique protein spots differentially accumulated when comparing aged to AR seeds, and 47 when comparing AR seeds of NILGAAS1 to that of Ler (Fig. 4B). Eight specific spots were identified in both comparison types (Table I, Fig. 9A), of which two spots, ID0715 and ID1104, contain CRUB. NILGAAS1 carries the truncated CRUB Cvi allele (Hou et al., 2005), due to this truncation the α subunit of CRUB has a smaller molecular weight (MW) compared to that of the Ler allele present in two other NILs used, which might account for the identification of CRUB in NILGAAS1. There are two spots, ID0715 and ID0537, displaying a LU protein profile. Spot ID0715 also corresponds to the fatty acid biosynthesis1 (FAB1) protein. This is not necessarily an interesting change since Arabidopsis seeds consist for around 30% oil bodies. Spot ID0537 contains aldehyde dehydrogenases 6B2 (ALDH6B2) and three isoforms of the ATP synthase β chain. ALDH proteins are encoded by a superfamily containing 14 genes that belong to nine distinctive families. ALDH6B2 is a unique gene encoding a methylmalonyl semialdehyde dehydrogenase (MM-ALDH) protein family (Kirch et al., 2004). MM-ALDHs are putatively involved in the degradation of valine, leucine and isoleucine. Isoleucine homeostasis in plants is maintained by threonine and methionine metabolism (Joshi et al., 2010). Methionine is a substrate for the synthesis of Ado-Met, which was demonstrated to be important for seed germination and seedling growth (Ravanel et al., 1998; Gallardo et al., 2002). Thus, the degradation of isoleucine by MM-ALDH might disrupt the homeostasis of threonine and methionine metabolism, and indirectly affect seed germination after storage.

Dry seeds are well equipped to confront oxidative stress during storage due to a very low water content that reduces metabolic activities. Auto-oxidation leading to accumulation of reactive oxygen species (ROS), however, occurs in dry seed. Particularly in seeds, proteins are the major targets of ROS due to their abundance and high affinity to oxidative reactions (Davies, 2005). To control ROS-induced damages, seeds have detoxification mechanisms to scavenge ROS through antioxidants. Protein spot ID0765 corresponding to monodehydroascorbate reductase 1 (MDAR1), is a well-known antioxidant enzyme involved in the ascorbate-glutathione cycle to remove hydrogen peroxide (Bailly, 2004). MDAR1 was 2.6 fold more abundant in AR NILGAAS1 than in Ler (Table III), which support a role of MDAR in ROS scavenging. However, T-DNA knock-out mutant analysis for seed longevity showed that mdar1.1 and mdar1.2 did not differ in seed longevity compared to Col (Fig. 7). The lack of a phenotype might be due to redundancy since the MDAR gene family contains five members. Over-expression of AtMDAR1 in tobacco conferred enhanced tolerance to ozone, salt and osmotic stresses (Eltayeb et al., 2007) suggesting that over-expressed MDAR1 can prove the role of ascorbate-glutathione in seed aging.


60

Moreover, GDP-D-mannose 3’,5’-epimerase (GME), a key enzyme for ascorbate (vitamin C) synthesis in plants (Wolucka et al., 2001; Wolucka and Van Montagu, 2003), was found in two protein spots that probably represent different modified forms of the same protein. Spot ID0712 had an 1.9 fold higher protein abundance in AR NILGAAS1 than Ler (Table III). GME accumulation in the second spot (ID0734) was 1.7 fold more abundant in aged than in AR seeds of NILGAAS1 (Table III). Probably the isoform in spot ID0734 was a non-functional protein. The fact that we could not isolate homozygous T-DNA knock-outs for GME suggests this to be an essential protein.

Receptor for activated C kinase1 (RACK1) is a WD40-type protein consisting of isoform A, B and C. RACK1A is the predominant isoform in the family. RACK1A was more abundant in AR seeds of NILGAAS1 than that of Ler (Table III), which might result in a better seed longevity for this NIL. RACK1 is a multi-function protein, that plays a regulatory role in diverse signal transduction pathways, and its transcripts are present during seed germination (Guo et al., 2009). RACK1A T-DNA knock-out analysis was performed to study the RACK1A effect on seed germination after artificial aging, however, seed germination of rack1A did not significantly differ from that of wild type Col (Fig. 7). Again in this case, we might deal with the redundant function family gene.

Mechanism involved in seed aging expressed in NILGAAS2NILGAAS2, the second most storable genotype, has 46 unique protein spots differentially accumulated when comparing aged to AR seeds (Fig. 4A) and 12 when comparing AR seeds of NILGAAS2 to that of Ler (Fig. 4B). Two spots were identified in both comparisons (Table II, Fig. 9B). Spot ID0458 corresponds to the T-complex protein 1 alpha subunit (TCP1) and had a LD protein profile (Table II). The involvement of TCP1 in seed longevity has not been studied yet.

Two protein spots corresponding to responsive to dehydration 29B (RD29B) increased abundance in AR seeds. Spot ID0206 was 1.7 fold higher in AR than in aged seeds of NILGAAS2, and spot ID0196 was 1.6 fold less abundant in AR seeds of NILGAAS2 than in Ler (Table III). RD29B transcript is known to be up-regulated by methylglyoxal, that inhibit seed germination (Hoque et al., 2012). Methylglyoxal is a side product of several metabolic pathways such as the glycolysis, lipid peroxidation and oxidative degradation of proteins, and accumulates in plants under various stress conditions (Hossain et al., 2009). The increased level of RD29B could be an indication for a decreased seed longevity.

It was noted that de-novo synthesis of new transcripts is not required for germination since seeds are able to germinate until radicle protrusion under the presence of α-amanitin (a transcription inhibitor targeting RNA polymerase II), however subsequent seedling growth was blocked (Rajjou et al., 2004). The newly synthesized gene transcripts necessary for germination vigor might be of importance for aged seeds. The accumulation of RNA polymerase II protein (AT2G15430, ID0762) was reduced by


61

Tabl

e II

I. Ge

noty

pe sp

ecifi

c pro

tein

s tha

t wer

e id

entif

ied

in e

ither

phy

siol

ogic

al st

ate

or g

enot

ype

com

pari

son

for f

or N

ILGA

AS1,

NIL

GAAS

2 an

d N

ILGA

AS5

The

tabl

e di

spla

ys p

rote

in s

pots

bas

ed o

n th

e se

ven

com

pari

sons

. Spo

t ID,

the

gene

cor

resp

ondi

ng to

the

prot

ein

unde

rlyin

g th

e sp

ot, m

olec

ular

wei

ght (

MW

in

kD) a

nd th

e th

eore

tical

(Th)

and

exp

erim

enta

l (Ex

p) is

oele

ctri

c poi

nt (p

I) a

re p

rese

nted

, res

pect

ivel

y; n

.a m

eans

not

ava

ilabl

e. F

urth

erm

ore

the

rela

tive

abun

danc

e (f

old

chan

ge) o

f the

spot

s in

both

type

s of c

ompa

riso

ns (F

ig. 1

; phy

siol

ogic

al st

ate

and

geno

type

) is i

ndic

ated

. Pos

itive

fold

chan

ges i

ndic

ate

high

er a

bund

ance

s, an

d ne

gativ

e low

er ab

unda

nces

. Fol

d ch

ange

s in

bold

indi

cate

the i

s sta

tistic

ally

sign

ifica

nt ch

ange

s. N

G1 st

ands

for N

ILGA

AS1,

NG2

for N

ILGA

AS2

and

NG5

for N

ILGA

AS5.

Sp

ots t

hat w

ere

iden

tifie

d ba

sed

on co

mpa

riso

n to

the

refe

renc

e pr

otei

n m

ap (h

ttp:

//w

ww

.seed

-pro

teom

e.co

m) a

re la

belle

d w

ith R .

Spot

IDGe

nePr

otei

n

MW

(kD)

pIRe

lativ

e ab

unda

nce

(fol

d ch

ange

)

ThEx

pTh

Exp

Phys

iolo

gica

l sta

te:

Aged

vs.

ARGe

noty

pe A

R:N

IL v

s. Le

r

Ler

NG1

NG2

NG5

NG1

NG2

NG5

NIL

GAAS

1

ID07

65AT

3G52

880

Mon

odeh

ydro

asco

rbat

e re

duct

ase 1

50.1

643

.31

8.38

6.08

-1.3

-1.3

-1.4

1.3

2.6

-1.0

-1.6

ID07

12AT

5G28

840

GDP-

D-m

anno

se 3

’,5’-e

pim

eras

e42

.76

45.1

46.

155.

78 1

.2

1.2

-1.0

-1.0

1.9

-1.1

1.0

ID07

34AT

5G28

840

GDP-

D-m

anno

se 3

’,5’-e

pim

eras

e42

.76

44.9

96.

155.

74 1

.2 1

.7 1

.2 1

.5 1

.1-1

.1-1

.1

ID09

51AT

1G18

080

Rece

ptor

for a

ctiva

ted

C ki

nase

1A

35.7

538

.56

7.81

7.13

-1.1

-1.6

-1.3

-1.8

1.6

1.4

1.4

NIL

GAAS

2

ID02

06AT

5G52

300

Resp

onsiv

e to d

ehyd

ratio

n 29

B65

.97

94.2

14.

815.

05-1

.5-1

.2-1

.7-1

.5-1

.3-1

.2 1

.1

ID01

96 R

AT5G

5230

0Re

spon

sive t

o deh

ydra

tion

29B

65.9

794

.21

4.81

5.00

-1.4

-1.9

1.0

1.1

1.2

-1.6

-1.5

ID07

62AT

2G15

430

RNA

polym

eras

e II

35.4

642

.82

4.39

4.46

-1.3

-1.1

-1.6

-1.1

-1.4

-1.1

-1.3

NIL

GAAS

5

ID15

05AT

4G27

160

Napi

n AT

2S3

18.7

69.

857.

878.

69 1

.5 1

.1-1

.2 1

.7-1

.0 1

.2 1

.2

ID06

32AT

5G38

470

Radi

atio

n se

nsiti

ve 2

3D40

.07

50.0

04.

294.

60 1

.1-1

.1-1

.0-1

.1 1

.0-1

.1-2

.0

ID02

58AT

3G15

670

Late

embr

yo ab

unda

nt p

rote

in24

.19

26.5

79.

438.

95-1

.2-1

.0 1

.1 1

.1 1

.0-1

.1-1

.9

ID11

44AT

3G15

670

Late

embr

yo ab

unda

nt p

rote

in24

.19

29.7

39.

437.

79-1

.1-1

.2-1

.1-1

.5 1

.0 1

.0 1

.2

ID09

76AT

3G17

520

Late

embr

yo ab

unda

nt p

rote

in32

.56

36.1

15.

025.

54 1

.2 1

.2 1

.4 1

.6 1

.0-1

.1-1

.2

ID04

48AT

2G19

900

NADP

-dep

ende

nt m

alic

enzy

me 1

64.2

8n.

a6.

73n.

a.-1

.3-1

.7-1

.7-1

.7 1

.1-1

.1-1

.0


62

1.6 fold after aging of NILGAAS2 seeds (Table III), which may contribute to a reduced seed germination after storage. For RNA polymerase II we could not isolate homozygous T-DNA insertion lines, probable because of lethality of the homozygous mutant.

Seed longevity mechanisms in NILGAAS5NILGAAS5, the most sensitive genotype to aging, has 64 unique protein spots differentially accumulated when comparing aged to AR seeds (Fig. 4A), and 11 when comparing AR seeds of NILGAAS5 to that of Ler (Fig. 4B). Two specific spots eID0228 and ID1146 identified in both comparisons, however, did not present either a LU or LD profile (Table I, Fig. 9C). These appear not involved in aging. DOG1 causes reduced seed longevity in NILGAAS5 (Nguyen et al., 2012; Chapter 2), but we did not identify the DOG1 protein as being differentially expressed in this study, probably due to its too low abundance.

During storage, seeds are subjected to DNA damage and genome instability that are considered to cause the reduced germination after aging. The maintenance of a functional DNA repair complex appears to be essential for long-term survival (reviewed by Rajjou et al., 2008). Radiation sensitive 23D (RAD23D) in protein spot ID0632 was 2-fold less abundant in AR seeds of NILGAAS5 compared with the better storable Ler seeds (Table III). RAD23 was suggested to participate in DNA damage repair since the two carrot (Daucus carota) RAD23 isoforms rescue the UV-sensitive phenotype of the rad23 deletion mutant in yeast (Saccharomyces cervisiae) (Sturm and Lienhard, 1998). In addition, it was shown by a yeast two hybrid assay that the maize (Zea mays) RAD23 isoform interacts with the abscisic acid Insensitive3 (ABI3) protein, a key regulator in seed maturation (Schultz and Quatrano, 1997; Zhang et al., 2005). The abi3 mutant of Arabidopsis lacks seed dormancy and has an reduced seed longevity (Ooms et al., 1993). Moreover, DOG1 appears to be involved in seed development since its expression peaks during seed development and the loss of function dog1 mutant has a defect in both seed longevity and seed dormancy, similarly to the abi3 mutant (Bentsink et al., 2006). This suggests a potential connection between RAD23, seed development, and seed longevity through DOG1. However, since RAD23D is located in the introgression region of NILGAAS5, we cannot exclude that it affects seed longevity independently from DOG1.

The abundance of late embryo abundant protein (LEA, AT3G17520, ID0976) was 1.6 times higher in aged than in AR seeds of NILGAAS5. An expression correlation network (viewed in GeneMANIA, Fig. 1S) using data of Zuber et al. (2010) showed that NADP-ME1 is connected to DOG1 via AT3G17520. These findings suggest that the acquisition of seed quality traits such as seed dormancy and seed longevity is under the control of seed maturation involving DOG1, LEA and NADP-ME1.

Two isoforms of another LEA protein (AT3G15670) were identified, spot ID0258 accumulated 1.9 fold less in AR seeds of NILGAAS5 than that of Ler, and spot ID1144


63

was 1.5 fold less abundant in aged compared to AR seeds of NILGAAS5 (Table III). This LEA was strongly down-regulated in seedlings of the greening after extended darkness 1 (ged1) mutant compared to wild type (Choy et al., 2008). ABA hypersensitive seed germination was observed for ged1. Elements similar to ABA response elements (ABREs) were detected in the upstream regions of all genes highly affected in ged1 including this LEA. The association of LEA and ABA responsiveness and seed germination after aging was examined by T-DNA knock-out analysis (Table S2). However, the lea mutant exhibited a similar seed longevity compared to Col (Fig. 7). These results indicate that ABA signalling might not be important for seed aging.

Another type of SSPs referred as 2S albumins is represented by five napin isoforms (Krebbers et al., 1988; Van der Klei et al., 1993). One of these AT2S3, was more abundant in aged NILGAAS5 (the most sensitive genotype to aging) (Table III). The AT2S3 protein spot ID1505 might be a degradation product due to its altered MW and isoelectric point (pI) (Table III). To examine if napins, similar to cruciferins, could affect seed longevity, an RNAi-napin line of a gene family encoding napins was analyzed for its seed longevity behavior (Withana-Gamage et al., 2013) (Table S2). The RNAi-napin line, that is depleted in napins and has a reduced protein content mainly in the endosperm, was more sensitive to aging than wild type Col (Fig. 5), which emphasizes the earlier identified role of SSPs in seed longevity.

The specific proteins detected for NILGAAS5 could be explained by the increased sensitivity to aging of this genotype, which is supported by the higher number of differentially accumulated spots identified. The changes in these proteins might occur in the other genotypes as well but after a longer storage time.

Conclusions

The naturally aged genetic material used in our analyses allows the identification of molecular processes involved in natural seed aging. In addition to all earlier proteomics analyses that have been often performed on a single genotype; in this study, the chances of identifying novel seed longevity factors are high due to the use of genetic material with different levels of seed longevity, increased seed longevity in NILGAAS1 and NILGAAS2 and reduced seed longevity in NILGAAS5 when compared to Ler. The dry seed proteome was greatly altered upon aging, despite its metabolic quiescence, which points to the presence of active mechanisms to promote longevity. Common proteins that were identified in all genotypes are SSPs, proteins related to translation and energy metabolism (glycolytic and TCA pathways), and vitamin E antioxidant (VTE1). Our results emphasize the importance of protection and maintenance of functional energetic and metabolic pathways, and antioxidant systems for seed longevity. Interestingly, the different genotypes used appear to express specific seed longevity pathways. NILGAAS1


64

potentially improves protection against aging through ascorbic acid antioxidants, and in NILGAAS5 seed longevity might be associated with seed development via NADP-ME1, LEA and DOG1.

Our findings confirm the results of other studies and also validate the involvement of several genetic factors on seed longevity by analyzing T-DNA knock-out and RNAi lines. Of tested 15 T-DNA gene knock-out lines (Table S1 and S2), five showed a significant seed longevity phenotype. Over-expression analyses of the identified genes or RNAi lines targeting whole gene families might be a better way to examine the effects of these genes on seed longevity, since this overcomes problems of redundant gene families (GAPC and PDC) or genes that encode products involved in redundant functional system (MDAR, GME and TPX). The knock-out lines of CRUs (2 genes), NAPINs, RPS12C and NADP-ME1 showed significant seed longevity phenotypes. Although we did not observe any seed morphological and seed dormancy phenotypes, it should be considered that these genes are expressed during seed development and thus mutations in these genes may have an impact on seed maturation and filling, and thereby indirectly on seed longevity.


Plant materialThe four genotypes, Ler, NILGAAS1(-Cvi), NILGAAS2(-An-1) and NILGAAS5(-Sha), were originally developed as NILDOG2, NILDOG22 and NILDOG1, respectively (Bentsink et al., 2010; Nguyen et al., 2012; Chapter 2). Those genotypes were grown in a randomized complete block design with replicates in soil as described in Bentsink et al. (2010). Seeds of four plants per replicate were bulked. Proteome analyses was conducted for the four genotypes at two physiological stages fully after-ripened (AR) and 4 year-old (aged) seeds. Fully AR seeds were stored in eppendorf tubes at -80oC. Aged seeds were seeds from the same harvest which were stored in cellophane bags under ambient conditions for four years. Four biological replicates were used in the proteomic analyses.

The single, double and triple T-DNA knock-out lines of cruciferin seed storage proteins (SSPs) and the RNAi line of napin SSP family gene were obtained from Withana-Gamage et al. (2013). T-DNA knock-out lines for candidate genes were screened for homozygous insertions and grown with the wild type Columbia (Col) under greenhouse conditions using rock wool supplemented with a Hyponex solution, in a randomized complete block design with four replicates per genotype.

Germination assaysGermination assays were performed according to Joosen at al. (2010) over a period of seven days. Briefly, samples, 50 to 200 seeds each, were sown on two layers of blue germination


65

papers equilibrated with 43 mL of demineralized water in a plastic tray (15 x 21 cm). Trays were piled and wrapped in a closed and transparent plastic bag. Germination piles were incubated in 22oC incubator, under continuous light (30 W.m-2). Pictures of germination assays were taken twice a day over a seven day period. Automatic scoring and curve fitting were analyzed by the Germinator package (Joosen et al., 2010). Germination percentages representing the germination ability were extracted at the end of germination assay. The germinator curve-fitting script, a part of the germinator package, enables to calculate averages and to perform statistic Student’s t-test.

Artificial aging We used artificial aging to evaluate seed longevity of the T-DNA lines. Approximately 200 seeds were aliquoted into 1.5 mL eppendoff tube and stored above a saturated NaCl solution in a closed and ventilated tank (relative humidity of 80% to 85% and temperature of 40oC) for 0 to 10 days. After treatment, germination assays were performed as described above.

Total soluble protein extracts30 mg of dry seeds of each sample (four biological replicates) were ground with mortar and pestle in liquid nitrogen for about one minute. Extraction buffer and protease inhibitor, as previously described by Rajjou et al. (2008), were added into seed powder, followed by a 2-min grinding. The extract was recovered into 1.5 mL eppendorf tube and incubated with DNase I, RNase A, and DTT at 4oC for an hour on rotating disc. The total soluble protein extract was collected as supernatant after centrifugation with 14000 rpm at 4oC for 10 min.

2D gel electrophoresisProtein separation was performed with 20 µL of protein extract, equivalent to about 150 µg of protein. 2D gel electrophoresis was conducted as illustrated in Rajjou et al. (2008; 2011), adapted for gel strips forming an immobilized nonlinear pH gradient from 3 to 11 (Immobilized DryStrip pH 3-11 NL, 24 cm; GE Healthcare).

Protein staining and gel analyses2D gels were stained with silver nitrate according to Rajjou et al. (2008). Silver-stained gels were placed within two layers of cellophane membrane stretched on cassette frames for drying. Images of dry gels were scanned with a Epson perfection V700. Quantitative image analysis was carried out with Progenesis Samespot software (v3.2, NonLinear Dynamics) to quantify proteins spots and detect significant change in protein accumulation over all the samples.

Comparison of proteome profiles and PCA analysisPair-wise statistics was used to detect protein spots that significantly changed their abundance. Two categories were defined: physiological state (4) and genotype (3). The


66

four physiological state comparisons were made between the protein profiles of the two physiological states (aged vs. AR) within each genotype. The three others were based on comparisons of AR seed proteome profiles of the NILs to that of Ler genetic background. Protein spots were considered to have a significantly different abundance when they were equal or more than 1.5 times up or down accumulated and when the P-value was equal or smaller than 0.05 according to one-way ANOVA test (Progenesis Samespot software) between the means of the four replicates.

The PCA was analyzed based on the volumes of the differentially acumulated spots using Progenesis Samespot software.

Protein identificationDue to the high reproducibility of 2D protein patterns, a number of proteins could be identified using previously established reference maps http://www.seed-proteome.com (Chibani et al., 2006) based on their position on gel. Otherwise, proteins spots were isolated from the 2D gels, digested with trypsin and identified by LC-MS/MS as described by Arc et al. (2012). Obtained sequences were submitted to the XTandem Pipeline (http://pappso.inra.fr/bioinfo/xtandempipeline/) databases in order to retrieve the full protein sequence and the gene annotation.

T-DNA knock-out genotype analysesA quick isolation method modified from Cheung et al. (1993) was performed to extract genomic DNA from leaves. In short, samples were ground in an extraction buffer containing 2 M NaCl, 200 mM Tris-HCl (pH 8), 70 mM EDTA and 20 mM Na2S2O5. The grinding was conducted with a stainless steel ball at 30 Hz for 1 min (96-well plate shaker, Mo Bio Laboratory). Then samples were incubated at 65oC for 1 h. Supernatants were collected after centrifugation at maximum speed for 10 min. DNA was precipitated by adding iso-propanol and 10M NH4Ac with ratio of 1:1/2:1 to the supernatant. This mixture was incubated at room temperature for at least 15 min, then centrifuged for 20 min at maximum speed. The DNA pellet was retrieved and rinsed with 70% ethanol followed by centrifugation for 5 min at maximum speed to recover the pellet. After drying, the DNA pellet was dissolved in distilled water.

Homozygous T-DNA insertion lines were screened with gene specific primers (left and right) and insert border primers (Table S1). T-DNA plants that amplified only the insertion product were consider to be homozygous mutants.

Polymerase chain reactions (PCR) were performed in a 12.5 µL-volume containing approximately 30 ng DNA, 25 µM of each dNTP, 25 ng of forward and reverse primers, 0.05U of DNA polymerase (Firepol, Solis BioDyne), 312.5 µM of MgCl2. The reaction protocol was as follows; denaturation at 95oC for 5 min followed by 30 s at 95oC, 30 s annealing at 52 to 57oC and a 45 s to 2 min extension at 72oC, this cycle was repeated for


67

35 times, and ended with last amplification for 10 min at 72oC.

The polymorphism was detected by agarose gel electrophoresis at concentrations from 1.5 % and higher (w/v) depending on size of differences.

Acknowledgement

We thank Dwayne Hegedus (Agriculture and Agri-Food Canada, Saskatoon; Department of Food and Bioproduct Sciences, University of Saskatchewan, Saskatoon) for a kind gift of cruciferin and napin mutant seeds. We also thank our colleagues Magdalena Gamm (Department of Molecular Plant Physiology, Utrecht University) and Farzaneh Yazdanpanah (Laboratory of Plant Physiology, Wageningen University) for providing seeds of rps12C, rack1A and nadp-me1 knock-out lines.

Supplemental data


Figure S1. Co-expression network between DOG1, LEA (AT3G17520) and NADP-ME1 obtained from Zuber et al. (2010)

Table S1. The nomenclature of seed storage proteins

Table S2. The T-DNA knock-out lines of the selected candidate genes

Table S3. Proteins that show an overlapping abundance pattern in after-ripened seeds of NILGAAS1 and NILGAAS2 in comparison with that of Ler

The genetic analyses and fine-mapping of two seed QTLs;

Delay Of Germination2 and Germination Ability After Storage1

Thu-Phuong Nguyen 1,2 and Leónie Bentsink 1,2

1 Wageningen Seed Laboratory, Laboratory of Plant Physiology, Wageningen University, 6708 PB Wageningen, The Netherlands; 2 Department of Molecular Plant Physiology, Utrecht University, 3584 CH Utrecht, The Netherlands.

4Chapter

Fine-mapping of DOG2/GAAS1Chapter 4

70

Abstract

Delay Of Germination2 (DOG2) and Germination Ability After Storage1 (GAAS1) are quantitative trait loci affecting seed dormancy and seed longevity, respectively. DOG2 and GAAS1 are colocated on top chromosome 1 in close vicinity of each other. The effects of both DOG2 and GAAS1 were confirmed by a near isogenic line (NIL), NILDOG2/GAAS1-Cape Verde Islands, that had a lower level of seed dormancy but a better seed longevity than Landsberg erecta. In this chapter, we have performed genetic analyses and showed that DOG2 and GAAS1 were likely two independent loci due to the differences in genetic behavior of DOG2 (maternal) and GAAS1 (co-dominant). The fine-mapping of DOG2 resulted in a 85 Kb-region containing 27 genes. The location of GAAS1 requires further investigation.


71

Introduction

Seed survival is an adaptive strategy that plants use to maintain their life cycle. Seeds need to survive the period in between seed dispersal and germination. Dormancy facilitates this survival by timing germination as such that plants will experience favorite environmental conditions for successful establishment. In addition to seed dormancy, seed longevity mechanisms allow the seed to survive prolonged periods in the (often dry) soil seed bank. Knowledge on the control of seed dormancy and longevity is necessary for various applications, for example, biodiversity conservation and breeding for varieties that have the required level of seed dormancy (preferably low dormancy) but that are storable for a long time. Moreover, knowledge on the genetic bases of seed dormancy and longevity can contribute to the understanding on how evolution shaped the natural variation of these traits.

Seed dormancy is defined as a failure of a viable seed to germinate in favorable conditions and seed longevity (storability) is an ability of seeds to remain viable after long-term storage. Dormant seeds cannot germinate unless environmental cues are applied, for example, dry after-ripening (AR, dry storage after harvest) or stratification (cold treatment in the imbibed state). AR could be considered as the first stage of aging in seeds exhibiting primary dormancy (Bewley et al., 2012), which indicates that both dormancy and longevity might be affected by the same processes. Seed aging is determined by seed moisture content, temperature, and initial seed quality, thus storage in high temperature and high humidity accelerates seed deterioration (Walters, 1998; Walters et al., 2005). In order to shorten the often long periods of natural aging, artificial aging became a common method to assess seed longevity (Tesnier et al., 2002; Rajjou et al., 2008).

Seed dormancy and longevity were discovered to be regulated by quantitative trait loci (QTLs) in several species. Seed dormancy QTLs have been identified in rice (Oryza sativa) (Gu et al., 2006), oilseed rape (Brassica napus) (Schatzki et al., 2013), wheat (Triticum aestivum) (Osa et al., 2003) and Arabidopsis (Arabidopsis thaliana) (Alonso-Blanco et al., 2003; Bentsink et al., 2010). QTLs for seed longevity after both natural and artificial aging have been found in lettuce (Lactuca sativa) (Schwember and Bradford, 2010), maize (Zea mays) (Revilla et al., 2009), wheat (Landjeva et al., 2009), rice (Miura et al., 2002; Sasaki et al., 2005), and Arabidopsis (Bentsink et al., 2000; Clerkx et al., 2004b; Nguyen et al., 2012; Chapter 2).

Moreover, in Arabidopsis, QTL analyses in six recombinant inbred line (RIL) populations, which were generated from crosses between Landsberg erecta (Ler) and six other accessions; Cape Verde Islands (Cvi), Antwerp (An-1), St. Maria do Feira (Fei-0), Kashmir (Kas-2), Kondara (Kond) and Shakdara (Sha), resulted in the identification of eleven seed dormancy loci (Delay Of Germination; DOG) (Bentsink et al., 2010) and nine


72

QTLs for seed longevity after natural aging (Germination Ability After Storage; GAAS) (Nguyen et al., 2012; Chapter 2). Interestingly, seven of the DOG and GAAS loci showed colocation. Noteworthy is the observed negative correlation between the allele effects. Alleles that increase seed dormancy decrease seed longevity and vice versa (Nguyen et al., 2012; Chapter 2). For one of the collocating regions, it was proven that a single gene, DOG1, controlled both seed dormancy and seed longevity (Nguyen et al., 2012; Chapter 2). The other six collocating regions remain to be investigated.

One of these regions is the DOG2/GAAS1 region. DOG2 and GAAS1 are located on top chromosome 1 in close vicinity of each other. The Cvi allele at DOG2/GAAS1 decreases seed dormancy whereas this allele increases seed longevity in comparison to the Ler allele (Nguyen et al., 2012; Chapter 2). Among all DOG loci, DOG2 was specific for the Ler/Cvi population explaining about 12% phenotypic variation (Alonso-Blanco et al., 2003; Bentsink et al., 2010). GAAS1 was the second strongest QTL affecting seed longevity after natural aging also found in Ler/Cvi population, where it accounted for 12.5% of explained variation (Nguyen et al., 2012; Chapter 2). The effect of a QTL is best characterized by a near isogenic line (NIL) that differs for a small genomic region spanning the QTL interval of interest from the genetic background (Alonso-Blanco and Koornneef, 2000). The effects of both DOG2 and GAAS1 were confirmed by NILDOG2-Cvi (Alonso-Blanco et al., 2003), that was later renamed to NILGAAS1-Cvi in the seed longevity study (Nguyen et al., 2012; Chapter 2).

NILs can be used as genetic tools for fine-mapping and chromosome walking towards cloning the gene underlying the QTL (Alonso-Blanco and Koornneef, 2000), which are the next steps after QTL analyses and confirmation of the QTL. In Arabidopsis, this approach is well facilitated by the presence of sequence information and gene annotation knowledge (reviewed by Weigel, 2012). Furthermore, Arabidopsis has enormous advantages in the validation of causal genes due to accessible resources like T-DNA knock-out mutants in almost every gene (http://arabidopsis.org), and due to its transformability which allows RNA interference, over-expression, and complementation studies. DOG1 was the first dormancy QTL successfully identified in such a way, starting with the fine-mapping using recombinants generated from a backcross between NILDOG1 and Ler, and validation by T-DNA knock-out mutants and complementation of the DOG1-Cvi allele into Ler (Bentsink et al., 2006).

This chapter presents the second attempt, after DOG1, to study the negative correlation between seed dormancy and seed longevity in more detail. Here, we have applied a similar strategy to identify candidate genes for the collocating DOG2 and GAAS1 loci, which resulted in 85 Kb-region containing 27 candidate genes for DOG2.


73

Results and discussion

Genetic analysis for DOG2 and GAAS1Seed dormancy and longevity of Ler and NILDOG2/GAAS1DOG2 and GAAS1 are two earlier identified QTLs for respectively seed dormancy (Bentsink et al., 2010) and seed longevity (Nguyen et al., 2012; Chapter 2), and locate in close vicinity of each other. The Ler allele of these QTLs increased the level of seed dormancy (DOG2) and decreased seed longevity (GAAS1). In the above mentioned studies, NILDOG2/GAAS1 was used to validate the effect of both DOG2 and GAAS1. We confirmed the phenotypes of Ler and NILDOG2/GAAS1, Ler required 52 days of seed dry storage to reach 50% germination (DSDS50) while NILDOG2/GAAS1 had a lower seed dormancy level, requiring 29 days (Fig. 1). The germination performance after artificial aging for 9 days of NILDOG2/GAAS1 (66%) was better than that of Ler (22%) (Fig. 1).

Figure 1. The genotypes and phenotypes of Ler and NILDOG2/GAAS1The top and bottom panels show the composite interval mapping position of the seed dormancy locus DOG2 and the seed longevity locus GAAS1 on chromosome 1, respectively. The arrows indicate the positions of the co-factors fixed in the QTL mapping and the direction of the arrowheads the allele effects. Arrowhead pointing up means that the Ler allele increases the phenotype and the arrowhead pointing down that the Ler allele decreases the phenotype. The two middle panels depict the graphical genotype of Ler and NILDOG2/GAAS1 at chromosome 1. The Ler allele is indicated in black and the Cvi alleles in gray. The pale gray region indicates the cross over region. Seed dormancy in DSDS50 (left) and seed longevity after artificial aging treatment in germination percentage (right) were evaluated for Ler and NILDOG2/GAAS1. Standard errors were calculated based on four biological replicates.

Genetic inheritance of DOG2 and GAAS1To study the genetic inheritance of DOG2 and GAAS1, we have analyzed reciprocal F1 and F2 seeds of crosses between Ler and NILDOG2/GAAS1 for seed dormancy and seed longevity after artificial aging. DOG2 inherited in a maternal manner, F1 seeds containing the NIL as a mother were less dormant than those that contained Ler as mother (Fig. 2A). The reciprocal F2 seed populations segregated for DOG2 and therefore showed an intermediate value compared to both parents. Moreover, they showed an equal seed dormancy level (Fig. 2B), which indicated that the maternal effect observed for DOG2 was not caused by a cytoplasmic factor.

The reciprocal F1 seeds did not significantly differ in seed longevity, both having


74

an intermediate germination percentage after artificial aging (Fig. 2C). Similar results were obtained for reciprocal F2 seeds (Fig. 2D), suggesting a co-dominant effect of the Ler and Cvi alleles at GAAS1.

Figure 2. Seed dormancy and seed longevity of Ler, NILDOG2/GAAS1 and reciprocal F1 and F2 seedsSeed dormancy expressed in DSDS50 and seed longevity in germination % after 9 days of artificial aging were measured for reciprocal F1 (A,C) and F2 seeds (B,D) and Ler and NILDOG2/GAAS1 grown in the same experiment. Standard errors were calculated on at least three biological replicates.

The dominance effect of DOG2 and GAAS1A subset of 152 plants of a segregating F2 population (Ler x NILDOG2/GAAS1) were randomly selected to study the dominance effect. The frequency distribution for seed dormancy suggests the dominance of the DOG2-Cvi allele because the number of low dormant lines was almost double that of the more dormant lines (Fig. 3A). However, this segregation ratio might also indicate an additional segregating locus that affects seed dormancy. Therefore, we have categorized the seed dormancy distribution based on the T7A14b marker genotype (located in the introgression) carrying either the Ler, Cvi or both (heterozygous) alleles (Fig. 3B). We could notice that there were two peaks for each genotype group, which indeed suggests two segregating loci affecting seed dormancy in this region. As a consequence we were unable to conclude about the dominance of DOG2.

For seed longevity, the co-dominance was confirmed since plants heterozygous for marker T7A14b reached a peak at germination percentages intermediate compared


75

with the peaks of lines that contained either the Ler or the Cvi allele (Fig. 3C and 3D).

Figure 3. Frequency distributions of seed dormancy and seed longevity, and the correlation of the two traits in a segregating populationThe set containing 152 F2 lines were phenotyped for seed dormancy expressed in DSDS50 (A, B) and longevity in germination percentage after four days of artificial aging treatment (C, D). The lines were grouped based on phenotype (A, C) and based on both phenotype and genotype at marker T7A14b located in the introgression (Ler, Cvi and Heterozygous) (B, D). The scatter plot showed the correlation between seed dormancy and longevity of the 152 F2 lines (E). Arrows indicated the values of two parents with standard errors presented by horizontal lines beneath.

We earlier described a negative correlation between seed dormancy and seed longevity (Nguyen et al., 2012; Chapter 2). Since DOG2/GAAS1 was one of the


76

collocating regions, we have also analyzed the correlation between seed dormancy and seed longevity in the 152 segregating F2 plants (Fig. 3E). However we did not find a correlation between both traits in this population, probably because of the additional locus affecting seed dormancy.

The different genetic behavior of the F1 and F2 seeds for dormancy (maternal) and longevity (co-dominant) and the fact that no correlation was observed for the segregating population might indicate that DOG2 and GAAS1 are two independent, closely linked genes regulating seed dormancy and seed longevity, respectively. In addition, the top of chromosome 1 appears to be a complex region due to another co-segregating factor for seed dormancy.

The fine-mapping of DOG2 and GAAS1Although dealing with two loci, the same mapping population could be used for fine-mapping purposes. In order to narrow DOG2 and GAAS1 regions, we have backcrossed NILDOG2/GAAS1 with Ler and generated a F2 population. Based on sequence polymorphism between Ler and Cvi (available on http://www.arabidopsis.org), molecular markers were developed for fine-mapping (Table S2). A total of 1030 plants were genotyped with four molecular makers (F21B7, T7A14b, T25N20 and F7A19). Recombination frequencies between the first marker F21B7 and T7A14b (46), T25N20 (64), and F7A19 (255) were counted. Based on the number of recombinants, we could calculate the genetic distance of the introgression, which made up to 24.76 cM (Fig. 4B). Compared to the map obtained by Alonso-Blanco et al. (2003) (Fig. 4A), this distance was a little bit larger (24.76 vs. less than 21.77). This could be explained by the differences in the number of lines and markers. For the fine-mapping of DOG2 and GAAS1, phenotyping on homozygous recombinants was required. Thus, 12 plants of the F3 generation were grown to find homozygous lines for each of the heterozygous F2 recombinants. Results will be described and discussed for DOG2 and GAAS1 in separate paragraphs.

Seed dormancy (DOG2)Homozygous recombinants spanning the QTL interval were analyzed for their seed dormancy behavior. A first set of them indicated that DOG2 is positioned between F21B7 and T7A14b (Fig. 4C, Table S3). We next genotyped all recombinants with four additional makers (F20D22, nF19P19, CRY2 and T1G11) (Fig. 4D). Phenotyping for seed dormancy in these recombinants is hampered by the fact that the flowering time locus Cryptochrome2 (CRY2) is located in the QTL interval and therefore segregates in the recombinants. El-Assal et al. (2001) showed that the CRY2-Cvi allele led to a few days earlier flowering compared with the CRY2-Ler allele; in our conditions the difference was 5 days. This results in earlier ripening of seeds that carry the CRY2-cvi allele, which affects the seed dormancy analyses (earlier start of the AR process). We did several


77

attempts to minimize the effect of CRY2 on the phenotyping: (1) synchronization of flowering time by sowing the recombinants containing the CRY2-Cvi allele 5 days after those containing the CRY2-Ler allele, (2) performing the dormancy analyses on seeds developed in a fixed time-window, (3) DSDS50 normalization by calculating DSDS50 values based on the number of days after harvest for each genotype, which allows normalization for genotypes we were unable to correct the difference in flowering time. Despite all these attempts in different experiments, none of the recombinants reached the DSDS50 value of either parent. Meaning that although we could recognize two phenotypic groups of recombinants, the DSDS50 values were always intermediate to the parental values (Table I, S3, and S4), which again indicates a potential additional genetic factor that affects seed dormancy in an additive or epistatic manner.

Figure 4. The fine-mapping of DOG2/GAAS1Graphical genotype of chromosome 1 of NILDOG2/GAAS1, the Ler genetic fragment is indicated in black and the Cvi fragment in gray, recombination region in white. The positions of the molecular markers are indicated in cM (A). The region containing the four markers used to select recombinants in the F2 segregating population and the number of recombinations between marker F21B7 and other markers is presented in detail (B). Graphical genotypes and dormancy phenotypes (in DSDS50) of Ler, NILDOG2/GAAS1 and four recombinants are illustrated (C). Detailed view of the fine-mapped region, molecular markers, the physical position (in Kb) and the number of recombinants between marker and DOG2 phenotypes based on the data shown in Table I (D). The fine-mapped location of DOG2 is depicted by a black bar.


78

Tabl

e I.

Geno

type

s and

phe

noty

pes o

f F4

reco

mbi

nant

sSe

ed d

orm

ancy

(DSD

S50)

and

long

evity

(Ger

min

atio

n %

afte

r 6 d

ays o

f art

ifici

al ag

ing;

Ger

.%) f

or th

e rec

ombi

nant

s (Li

ne) a

re ca

lcul

ated

on

four

bio

logi

cal r

eplic

ates

. Ge

noty

pes

at th

e po

sitio

ns o

f the

mar

kers

are

indi

cate

d. T

he ra

nge

and

mea

n DS

DS50

and

Ger

min

atio

n %

val

ues

of th

e in

divi

dual

repl

icat

es a

re m

entio

ned,

and

re

com

bina

nt p

heno

type

s wer

e as

sign

ed to

eith

er th

e Le

r or t

he N

IL g

roup

. Rec

ombi

nant

s exc

lude

d fr

om th

e an

alys

es a

re in

dica

ted

by x

.

Line

Mar

ker

Dorm

ancy

(DSD

S50)

Long

evity

(Ger

. %)

F21B

7F2

0D22

nF19

P19

CRY2

T1G1

1T7

A14b

T25N

20F7

A19-

2M

ean

Rang

ePh

e.M

ean

Rang

ePh

e.47

5-8

Ler

Cvi

Cvi

Cvi

Cvi

Cvi

Cvi

Cvi

24.0

± 2

.021

-30

NIL

57.5

± 3

.747

-63

NIL

58-5

Ler

Ler

Cvi

Cvi

Cvi

Cvi

Cvi

Cvi

25.3

± 2

.219

-29

NIL

56.5

± 6

.142

-66

NIL

718-

7Le

rLe

rCv

iCv

iCv

iCv

iCv

iCv

i30

.5 ±

0.9

29

-33

x55

.3 ±

2.4

50-6

2NI

L94

3-11

Ler

Ler

Cvi

Cvi

Cvi

Cvi

Cvi

Cvi

24.8

± 1

.721

-28

NIL

57.0

± 6

.339

-69

NIL

228-

6Le

rLe

rLe

rLe

rCv

iCv

iCv

iCv

i33

.3 ±

1.3

31-3

7Le

r39

.0 ±

7.2

24-5

8x

338-

6Le

rLe

rLe

rLe

rCv

iCv

iCv

iCv

i32

.3 ±

1.1

31-3

5Le

r36

.1 ±

5.8

20-4

8x

175-

8Le

rLe

rLe

rLe

rCv

iCv

iCv

iCv

i33

.3 ±

1.7

32-3

6Le

r29

.0 ±

7.2

16-4

9Le

r19

-9Le

rLe

rLe

rLe

rLe

rCv

iCv

iCv

i32

.5 ±

1.2

30-3

5Le

r31

.7 ±

5.5

22-4

7x

498-

8Le

rLe

rLe

rLe

rLe

rCv

iCv

iCv

i33

.8 ±

1.0

29-3

6Le

r36

.0 ±

2.4

30-4

1x

6.02

-7Le

rLe

rLe

rLe

rLe

rCv

iCv

iCv

i34

.0 ±

1.4

32-3

8Le

r35

.6 ±

5.5

25-5

0x

379-

6Le

rLe

rLe

rLe

rLe

rCv

iCv

iLe

r35

.3 ±

3.7

25-4

2Le

r30

.1 ±

5.2

21-4

5x

689-

3Le

rLe

rLe

rLe

rLe

rLe

rLe

rCv

i30

.3 ±

2.0

27-3

5x

46.7

±5.

138

-60

x8.

10-1

1Le

rLe

rLe

rLe

rLe

rLe

rLe

rCv

i36

.3 ±

2.3

31-4

0Le

r37

.9 ±

6.8

16-4

5x

313-

6Cv

iLe

rLe

rLe

rLe

rLe

rLe

rLe

r32

.8 ±

1.4

31-3

7Le

r32

.6 ±

2.6

27-3

9x

617-

7Cv

iLe

rLe

rLe

rLe

rLe

rLe

rLe

r32

.0 ±

1.6

28-3

6Le

r34

.5 ±

9.0

12-4

9x

74-9

Cvi

Ler

Ler

Ler

Ler

Ler

Ler

Ler

31.3

± 3

.123

-38

Ler

35.6

± 3

.128

-43

x71

4-1

Cvi

Cvi

Ler

Ler

Ler

Ler

Ler

Ler

35.0

± 1

.532

-38

Ler

45.8

± 4

.436

-56

Ler

807-

8Cv

iCv

iCv

iLe

rLe

rLe

rLe

rLe

r27

.5 ±

1.0

26-3

0x

49.1

± 5

.339

-63

NIL

99-5

Cvi

Cvi

Cvi

Ler

Ler

Ler

Ler

Ler

33.8

± 2

.129

-39

Ler

51.2

± 6

.340

-68

NIL

180-

8Cv

iCv

iCv

iCv

iLe

rLe

rLe

rLe

r29

.3 ±

1.1

26-3

1x

53.4

± 8

.838

-73

NIL

689-

1Cv

iCv

iCv

iCv

iLe

rLe

rLe

rLe

r25

.8 ±

0.9

24-2

8NI

L62

.8 ±

7.1

47-7

3NI

L92

2-7

Cvi

Cvi

Cvi

Cvi

Ler

Ler

Ler

Ler

27.8

± 1

.026

-30

x62

.6 ±

5.8

51-7

3NI

L16

6-12

Cvi

Cvi

Cvi

Cvi

Cvi

Ler

Ler

Ler

27.5

± 1

.224

-29

x57

.4 ±

3.0

50-6

4NI

L17

8-4

Cvi

Cvi

Cvi

Cvi

Cvi

Ler

Ler

Ler

25.0

± 1

.721

-29

NIL

65.3

± 3

.558

-74

NIL

599-

6Cv

iCv

iCv

iCv

iCv

iLe

rLe

rLe

r24

.5 ±

1.5

21-2

7NI

L55

.4 ±

6.4

40-7

1NI

L17

5-11

Cvi

Cvi

Cvi

Cvi

Cvi

Cvi

Cvi

Ler

25.3

± 2

.817

-29

NIL

66.7

± 6

.349

-79

NIL

691-

10Cv

iCv

iCv

iCv

iCv

iCv

iCv

iLe

r25

.0 ±

1.2

23-2

7NI

L62

.9 ±

3.4

53-6

7NI

LLe

rLe

rLe

rLe

rLe

rLe

rLe

rLe

rLe

r41

.3 ±

2.8

36-4

511

.8 ±

3.8

1-19

NIL

Cvi

Cvi

Cvi

Cvi

Cvi

Cvi

Cvi

Cvi

21.3

± 0

.620

-23

72.3

± 7

.849

-84

84

10

410

1012

No. r

ecom

. for d

orm

ancy

55

02

69

911

No. r

ecom

. for l

onge

vity


79

Because we could not unambiguously determine the position of the DOG2 locus (Table S3 and S4), we used the phenotypes and genotypes of the recombinants for a QTL analysis. We identified, using the data presented in Table S3, a highly significant peak associated with the CRY2 marker. This QTL, for which the Ler allele enhanced the dormancy level, had a LOD score of 9.3 and explained 45.7% of the variance (Fig. 5A). An additional peak at the end of the introgression (marker F7A19) had the same allele direction but was not significant (LOD of 1.26, 4.3% explained variance) (Fig. 5A). Similar results (Fig. 5B) were achieved for data presented in Table S4. It was revealed that the two QTLs did not have an epistatic interaction (Table S5), therefore they regulated seed dormancy in an additive manner.

Since the QTL analyses clearly indicated the presence of a strong locus near CRY2, we further studied the recombinants. We scored seed dormancy based on the mean and range of their DSDS50 value in comparison with the parental lines Ler and NILDOG2/GAAS1, that had been grown at the same time (Table I, S3, and S4). Still several recombinants were excluded because we could not assign to which phenotypic group they belonged. Doing so we could reduce the DOG2 region to 85 Kb between markers nF19P19 and T1G11 (Table I).

Seed longevity (GAAS1)GAAS1 was the second strongest seed longevity QTL identified (Nguyen et al., 2012; Chapter 2), explaining 12.5% of the phenotypic variation in the Ler/Cvi population. Although the seed longevity of Ler and NILDOG2/GAAS1 were significantly different in earlier experiments (Fig. 2C and 2D), that was not always consistent for the parents grown with the recombinants (Table S3 and S4), as a consequence the fine-mapping of GAAS1 was even more difficult than that of DOG2. Also here, we have performed QTL analyses for seed longevity after artificial aging treatment. For data presented in Table S3, a QTL fixed by co-factor F20D22 was detected with a LOD of 4.99 (Fig. 5C). This QTL only explained 28% of phenotypic variance, another 5.6% of the variance was explained by a non-significant peak at the end of the introgression. A similar picture was observed for the data set in Table S4.

In only one of the attempts, we observed a clear seed longevity phenotype for the parents, and were able to assign a part of the recombinants to either parental group (Table I), which narrowed the region of GAAS1 into 116.1 Kb flanked by markers F20D22 and CRY2. However, this is a preliminary result since we had to exclude a number of recombinants which showed intermediate phenotypes (Table I).

It appears that the top of chromosome 1 is a genetically complex region that requires a more detailed study. Possibly, this region harbors so far unknown factors involved in seed longevity. Moreover, it was difficult to obtain unequivocal seed longevity phenotypes for NILDOG2/GAAS1.


80

Figure 5. QTL mapping for seed dormancy and seed longevity using recombinants derived from a cross between Ler and NILDOG2/GAAS1QTL maps for seed dormancy (A, B) and seed longevity (C, D) using F3 (A, C) and F4 (B, D) recombinant populations, respectively. Only the top region of chromosome one has been used in the analyses. Logarithm of Odds (LOD) scores are presented to indicate the strength of the QTL. QTL scores above 1.6 can be considered significant.

Candidate genes for seed dormancy DOG2 locusThe fine-mapped region of DOG2 contains 27 genes (Table II), among which CRY2. It was reported that CRY2 did not have pleiotropic effect on seed germination during AR (El-Assal et al., 2001). However, we noticed differences in growth, harvest and germination conditions between the study of El-Assal et al. (2001) and our study. In addition, seed dormancy was known to be affected by environmental factors (Contreras et al., 2008; Kendall and Penfield, 2012). Therefore we re-analyzed the effect of CRY2 on seed dormancy.

El-Assal et al. (2001) detected that the CRY2 protein of Cvi carries a single amino acid substitution of methionine (M) at position 367 for valine (V) present in Ler. This substitution results in a very early flowering phenotype in short days, but also reduces the photoperiod requirement in long days (El-Assal et al., 2001). In order to determine the effect of this substitution on seed dormancy, we have analyzed the seed dormancy behavior for a set of CRY2 alleles including transformants carrying either the CRY2-Ler (CRY2-Cvi-367V) or the CRY2-Cvi allele (CRY2-Ler-367M) (El-Assal et al., 2001), and the EDI-NIL, that was used for the isolation of CRY2. Limited variation in seed dormancy


81

Gene

IDAn

nota

tion

Tiss

ues

Loca

lizat

ion

AT1G

0432

0tR

NA-G

lyn.

aCe

llula

r com

pone

ntAT

1G04

330

unkn

own

Mat

ure s

eeds

Chlo

ropl

ast

AT1G

0434

0HR

-like

lesio

n-in

ducin

g pro

tein

-rela

ted

Vario

us ti

ssue

s inc

ludi

ng im

mat

ure s

eeds

Endo

mem

bran

e sys

tem

AT1G

0435

0Si

mila

r to 2

-oxo

glut

arat

e-de

pend

ent d

ioxy

gena

seVa

rious

tiss

ues i

nclu

ding

seed

sCy

topl

asm

AT1G

0436

0RI

NG/U

-box

supe

rfam

ily p

rote

inSe

ed co

at, h

ypoc

otyl

and

root

Nucle

usAT

1G04

370

Ethy

lene-

Resp

onsiv

e Elem

ent B

idin

g Fac

tor1

4n.

aNu

cleus

AT1G

0438

0Si

mila

r to 2

-oxo

glut

arat

e-de

pend

ent d

ioxy

gena

seFl

ower

, hyp

ocot

yl an

d se

ed en

dosp

erm

Cyto

plas

mAT

1G04

390

BTB/

POZ

dom

ain-

cont

aini

ng p

rote

inVa

rious

tiss

ues i

nclu

ding

seed

sCh

loro

plas

tAT

1G04

400

Cryp

toch

rom

e2Va

rious

tiss

ues i

nclu

ding

seed

sNu

cleus

, vac

uole

AT1G

0441

0Cy

toso

lic-N

AD-D

epen

dent

Mal

ate D

ehyd

roge

nase

1Va

rious

tiss

ues i

nclu

ding

imm

atur

e see

dsNo

t in

Golg

i app

arat

us, c

ytos

olAT

1G04

420

NAD(

P)-li

nked

oxid

ored

ucta

se su

perfa

mily

pro

tein

Vario

us ti

ssue

sCh

loro

plas

tAT

1G04

430

S-ad

enos

yl-L

-met

hion

ine-

depe

nden

t met

hyltr

ansfe

rase

s sup

erfa

mily

pro

tein

Vario

us ti

ssue

sGo

lgi a

ppar

atus

, tran

s-Go

lgi n

etw

ork

AT1G

0444

0Ca

sein

Kin

ase L

ike1

3Va

rious

tiss

ues i

nclu

ding

seed

sCy

topl

asm

AT1G

0444

5C2

H2-li

ke zi

nc fin

ger p

rote

inn.

aIn

trace

llula

rAT

1G04

450

ROP-

Inte

ract

ive C

RIB

Mot

if-Co

ntan

ing P

rote

in3

n.a

Cyto

plas

m, n

ucle

usAT

1G04

457

50S r

ibos

omal

pro

tein

L28

n.a

n.a

AT1G

0447

0M

unc1

3 ho

mol

ogy 1

n.a

n.a

AT1G

0448

0Ri

boso

mal

pro

tein

L14p

/L23

e fam

ily p

rote

inVa

rious

tiss

ues i

nclu

ding

seed

sCh

loro

plas

t, rib

osom

eAT

1G04

490

unkn

own

root

, sta

men

Nucle

usAT

1G04

500

CCT

mot

if fam

ily p

rote

inFl

ower

and

polle

nNu

cleus

AT1G

0450

1un

know

nn.

an.

aAT

1G04

510

U-bo

x pro

tein

sVa

rious

tiss

ues

Nucle

usAT

1G04

520

Plas

mod

esm

ata-

Loca

ted P

rote

in2

Vario

us ti

ssue

sPl

asm

odes

ma

AT1G

0453

0Te

tratri

cope

ptid

e Rep

eat4

Vario

us ti

ssue

s inc

ludi

ng se

eds

Nucle

usAT

1G04

540

Calci

um-d

epen

dent

lipid

-bin

ding

fam

ily p

rote

inVa

rious

tiss

ues

Mem

bran

eAT

1G04

550

Indo

le-3-

Acet

ic Ac

id In

ducib

le12

Vario

us ti

ssue

s inc

ludi

ng se

eds

Nucle

usAT

1G04

555

unkn

own

n.a

Mito

chon

drio

n

Tabl

e II

. Lis

t of c

andi

date

gen

es in

85

Kb fi

ne-m

appe

d re

gion

cont

aini

ng 2

7 ge

nes

Gene

iden

tity a

nd an

nota

tion

are l

iste

d in

the f

irst

two

colu

mns

. Inf

orm

atio

n on

whe

re th

e gen

e is e

xpre

ssed

(tis

sues

) and

whe

re th

e pro

tein

is lo

caliz

ed (l

ocal

izat

ion)

ar

e ob

tain

ed fr

om T

AIR

and

eFP

brow

sers

. Gen

es in

dica

ted

in b

old

are

the

DOG2

cand

idat

es p

rese

nted

in F

igur

e 7.

n.a

indi

cate

s inf

orm

atio

n no

t ava

ilabl

e.


82

among the different CRY2 variants was observed (Fig. 6B), in agreement with El-Assal et al. (2001), confirming that CRY2 was not DOG2 regulating seed dormancy. We also observed that the EDI-NIL does not contain the DOG2-Cvi phenotype, whereas based on our fine-mapping it should include the dormancy locus (Fig. 6A). Since we do not have detailed genome-wide genotyping information of the EDI-NIL, we cannot exclude that this is caused by an additional introgression elsewhere in this NIL.

Figure 6. The effect of CRY2 on seed dormancyA schematic representation of the Cvi introgression with markers and physical distance in Kb and representative genotypes of Ler, NILDOG2/GAAS1 and EDI-NIL (A). Seed dormancy level presented in DSDS50 for Ler, NILDOG2, EDI-NIL, Ler lines transformed with different CRY2 variants: CRY2-Ler, CRY2-Ler-367M, CRY2-Cvi, CRY2-Cvi-367V) (B). Standard errors were calculated on four biological replicates.

In order to determine which of the genes in the DOG2 region (85 Kb) could explain the dormancy difference between the two alleles, we had a look at the expression of these genes (eFP browser source, http://bar.utoronto.ca/efp). Genes affecting seed dormancy are likely expressed in the seed. Due to the maternal nature of DOG2, genes expressed in maternal integuments, i.e. seed coat and endosperm, appeared to be interesting candidates. Two of the identified genes AT1G04360, a RING/U-box superfamily protein, and AT1G04380 encoding a protein similar to 2-oxoglutarate-dependent dioxygenase areexpressed in seed coat and endosperm, respectively (Fig. 7). However, at this moment we cannot exclude other genes to encode for DOG2.


83

Figure 7. Seed gene expression profiles of two DOG2 candidate genesGene expression during imbibition treatment (h) in whole seed, and seed sections which include cotyledons, radicles, micropylar endosperm and lateral endosperm is depicted for AT1G04360 (A) and AT1G04380 (B). Along the imbibition treatment, it is indicated when testa is ruptured (TR) or non-ruptured (NR), and when endosperm is ruptured (ER). Expression values are absolute and represented by color scale. Data were obtained from http://vseed.nottingham.ac.uk/ and Dekkers et al. (2013).

Conclusions

The genetic analyses for seed dormancy and seed longevity indicated that DOG2 and GAAS1 are two genes located in a very close vicinity of each other. The fine-mapping of DOG2 and GAAS1 was hindered by various above discussed reasons. The top chromosome 1 appears to be very complex, and there might be several genetic factors affecting the two traits. However, we could reduce the DOG2 region to 85 Kb containing 27 genes. Reverse genetic analyses, such as complementation and T-DNA knockout mutant analyses will conclusively identify the DOG2 gene. To fine-map GAAS1, optimization of seed longevity


84

phenotyping will be required because recombinants did not show a clear phenotype. This can be facilitated by seed development at higher temperatures (25°C), which has been described to increased seed longevity and thereby the difference between Ler and NILDOG2/GAAS1 (He et al., submitted).


Plant materialNILDOG2-Cvi contains a 25 cM Cvi introgression on top chromosome 1 in a Ler genetic background developed by Alonso-Blanco et al. (2003). This NIL was later renamed to NILGAAS1 because of its confirmed seed longevity phenotype of the GAAS1 QTL (Nguyen et al., 2012; Chapter 2). In this chapter, we will name this line NILDOG2/GAAS1. Reciprocal F1 and F2 seeds of a cross between Ler and NILDOG2/GAAS1 and a segregating population of 1032 F2 plants were generated for genetic analyses and fine-mapping of both DOG2 and GAAS1.

To analyze the effect of CRY2 on the studied traits, we also included EDI-NIL, that carries a 7 cM Cvi introgression from the top chromosome 1 in Ler background (El-Assal et al., 2001), and a group of transformants (CRY2-Ler, CRY2-Ler-367M, CRY2-Cvi, CRY2-Cvi-367V) obtained from El-Assal et al. (2001).

Reciprocal crossesReciprocal F1 seeds were produced by hand pollination, anthers were removed from the mother of the cross and pollens of the father were artificially placed on top of the stigma of mature flowers. Approximately 19 days after pollination, seeds were harvested. Reciprocal F2 seeds were harvested from reciprocal F1 plants.

Growth, harvest and storage conditionsAt the vegetative stage (before inflorescence emergence), plants were grown in a controlled climate chamber at 20°C/18°C (day/night) under a 16-h photoperiod of artificial light (150 μmol·m-2·s-1) and 70% relative humidity, in either soil or hyponex solution (Table S1). When the infloresence emerged, growing temperature was either kept the same for experiments using soil or changed in to 25°C/22°C (day/night) for experiments using hyponex.

In order to reduce the effect of flowering time differences between the recombinants in experiment WUR12-8 (data presented in Table S4), only seeds developed at the same period of 10 days during seed development were harvested. Whereas, in experiment WUR12-15 (data presented in Table I), growing time of recombinants was adjusted according to the CRY2 allele to synchronize flowering time, meaning that recombinants with the CRY2-Ler allele were grown 5 days earlier than those with the CRY2-Cvi allele.


85

This adjustment enabled seeds of most genotypes to mature at the same time.

Plants were grown and harvested as a single plant or as a block of four plants depending on the experiment (Table S1). Seeds were stored in 6 x 13 cm cellophane flat bags (Hera papierverarbeiting, Nidda, Germany) or 6.5 x 10.5 cm paper bags (Koninklijke Van der Most BV, Heerde, The Netherlands) at ambient condition.

Seed germination, dormancy and longevity assaysGermination assays were performed according to Joosen et al. (2010). Briefly, six samples, approximately 50 to 100 seeds per sample, were sown on two layers of blue germination paper equilibrated with 50 mL of demineralized water in plastic tray (15 x 21 cm). Trays were piled and wrapped in a closed transparent plastic bag. Germination was incubated in 22oC incubator, under continuous light (30 W.m-2). Pictures of germination assays were taken over 7-day period. Automatic scoring and curve fitting were analyzed by the Germinator package (Joosen et al., 2010) to obtain final germination percentage.

Seed dormancy measurement: Germination assays were performed every week from fresh harvest until seeds were fully after-ripened (100% germination). Generalized linear model with a logit link as described in Hurtado et al. (2012) was adapted to calculate the days of seed dry storage required to reach 50% germination (DSDS50). Germination data were adjusted to the data situation by choosing n = 100 and fitted one smooth curve per line. The observed germination proportion was re-interpreted as having observed y “successes” in n binomial trials (e.g. 75% germinated means y = 75 out of 100 possible “trials”). DSDS50 was the closest time point to where a horizontal line with y = 50 crosses the fitted curve (He et al., submitted). DSDS50 for each genotype was calculated based on the number of days after harvest, which allows the normalization for genotypes we were unable to correct the difference of flowering time.

Seed longevity measurement: when seeds had been fully after-ripened, seed longevity was assessed after artificial aging. A small amount of seeds (ca. 150 seeds) were aliquoted in 1.2 mL opened tube and stored above a saturated NaCl solution in a closed and ventilated tank (relative humidity of 85% and temperature of 40oC) for a period varying from 0 to 10 days. After treatment, germination assays were performed as described above.

DNA isolation, PCR condition and gel electrophoresisA quick isolation method modified from Cheung et al. (1993) was performed to extract genomic DNA from leaves and seedlings. In short, samples were ground in an extraction buffer containing 2 M NaCl, 200 mM Tris-HCl (pH 8), 70 mM EDTA, and 20 mM Na2S2O5. The grinding was conducted with a stainless steel ball at 30 Hz for 1 min (96-well plate


86

shaker, Mo Bio Laboratory). Then samples were incubated at 65oC for 1 h. Supernatants were collected after centrifugation at maximum speed for 10 min. DNA was precipitated by adding iso-propanol and 10M NH4Ac with ratio of 1:1/2:1 to the supernatant. This mixture was incubated at room temperature for at least 15 min, then centrifuged for 20 min at maximum speed. The DNA pellet was retrieved and rinsed with 70% ethanol followed by centrifugation for 5 min at maximum speed to recover the pellet. After drying, the DNA pellet was dissolved in distilled water.

Polymerase chain reactions (PCR) were performed in a 12.5 µL-volume containing approximately 30 ng DNA, 25 µM of each dNTP, 25 ng of forward and reverse primers, 0.05U of DNA polymerase (Firepol, Solis BioDyne), 312.5 µM of MgCl2. The reaction protocol was as follows; denaturation at 95oC for 5 min followed by 30 s at 95oC, 30 s annealing at 50 to 58oC and a 45 s to 2 min extension at 72oC, this cycle was repeated for 35 times, and ended with last amplification for 10 min at 72oC.

The polymorphism was detected by agarose gel electrophoresis concentrated from 1.5 to 2.5 % (w/v) depending on size of differences.

QTL analysesSeed dormancy expressed in DSDS50 and seed longevity determined as germination percentage after artificial aging treatment of the recombinants were phenotypic data for QTL analyses. The analyses including interval mapping and multiple-QTL-model were conducted with MapQTL software version 6.0 (Van Ooijen, 2009) using eight markers in the introgressed region (Table S2).

Epistatic interactions between sets of two markers was examined using the Generalized Linear Model (IBM SSPS statistics 19).

Supplemental data

Supplemental information can be downloaded from http://www.wageningenseedlab.nl/thesis/tpnguyen/chapter4Table S1. Overview of growth, harvest and storage conditions for the different experimentsTable S2. Markers used for the fine-mappingTable S3. Genotypes and phenotypes of the F3 homozygous recombinants grown in experiment WUR11-26Table S4. Genotypes and phenotypes of F4 homozygous recombinants (Line) grown in experiment WUR12-8Table S5. Analysis for epistatic interaction between CRY2 and F7A19 in the control of seed dormancy


Physiological and genetic characterization of novel seed dormancy mutants

Thu-Phuong Nguyen 1,2, Marieke Van Bolderen-Veldkamp 1,2, Jeroen Wolfkamp 1, and Leónie Bentsink 1,2

1 Wageningen Seed Laboratory, Laboratory of Plant Physiology, Wageningen University, 6708 PB Wageningen, The Netherlands; 2 Department of Molecular Plant Physiology, Utrecht University, 3584 CH Utrecht, The Netherlands.

5Chapter

Identification of seed dormancy modifiersChapter 5

88

Abstract

Seeds are fully competent to germinate once seed dormancy is released. Delay Of Germination6 (DOG6) was identified to be one of the strongest quantitative trait loci (QTL) controlling seed dormancy in Arabidopsis recombinant inbred line populations. The effect of DOG6 on seed dormancy was confirmed by a near isogenic line containing an introgression fragment of the Shakdara accession (NILDOG6-Sha) spanning the QTL interval in the Landsberg erecta background. We used EMS-induced mutagenesis, to screen for novel reduced seed dormancy mutants in the dormant NILDOG6-Sha background. To exclude well-known non-dormant mutants, such as aba and abi mutants, assays for seed germination sensitivity to paclobutrazol and ABA, respectively, were performed. Several mutants allelic to the dog1 mutant were found by allelism test, suggesting that DOG1 is epistatic over DOG6 in the control of seed dormancy. In total, six novel mutants, named seed dormancy modifier1 (sdm1) to sdm6 were isolated. In addition to reduced dormancy, sdm1, sdm3, sdm4, sdm5, and sdm6 showed reduced seed longevity as a pleiotropic effect of the mutated seed dormancy gene. Sdm2 and sdm4 exhibited abnormal seed mucilage. Mapping populations were generated and phenotyped in order to identify the underlying genes by a mapping-by-sequencing approach.


89

Introduction

Seed dormancy is an important factor of plant fitness because it determines the environmental conditions, in which a seed is able to germinate to ensure plant establishment success (Donohue et al., 2005; Huang et al., 2010). Seed dormancy is defined as a temporary failure of a viable seed to germinate in conditions that favor germination. Mutant screens revealed that gibberellins (GA) and abscisic acid (ABA) are important hormones for seed dormancy and germination. GA releases seed dormancy and promotes seed germination while ABA induces and maintains seed dormancy; the balance of these two hormones determines whether a seed germinates or not (reviewed in Finch-Savage and Leubner-Metzger, 2006; Holdsworth et al., 2008). The first Arabidopsis mutants isolated with a seed germination phenotype are a group of GA-deficient (ga) mutants. These mutants cannot germinate without the addition of exogenous GA and, therefore, express an increased seed dormancy (Koornneef and Van der Veen, 1980; Koornneef et al., 1982). One of these GA-deficient mutants, ga1, was used to isolate the ABA-deficient aba1 mutant, that shows reduced seed dormancy, and does not require GA to germinate (Koornneef et al., 1982). The aba2 and aba3 mutants were isolated in a screen for germination in the presence of the GA biosynthesis inhibitor paclobutrazol (Léon-Kloosterziel et al., 1996a). Such aba mutants are affected in ABA biosynthesis and, therefore, contain less ABA, resulting in a non-dormant seed phenotype. Seeds of these aba mutants are more resistant to paclobutrazol than wild type seeds since they do not require GA to overcome the ABA-induced block of germination. Another type of ABA related mutants are ABA insensitive (abi) mutants. They are derived from a screen for seed germination in the presence of ABA, thus seeds of abi mutants are able to germinate in the presence of a higher ABA concentration than wild type (Koornneef et al., 1984).

In addition to hormones, other factors that affect seed dormancy have been identified. Screens for mutants that do not require after-ripening (seed dry storage that overcomes seed dormancy) allowed the identification of the reduced-dormancy (rdo) mutants, rdo1 to rdo4 (in the Landsberg erecta [Ler] genetic background) (Léon-Kloosterziel et al., 1996b; Peeters et al., 2002), and the delay of germination1 (dog1) mutant (in the NILDOG1-Cvi background; a dormant near isogenic line [NIL] that carries the DOG1 Cape Verde Island [Cvi] allele) (Bentsink et al., 2006). Moreover, seed germination is determined by both the embryo and the tissues surrounding the embryo which are the seed coat (testa) and the endosperm (Bewley, 1997; Dekkers et al., 2013). Thus, also mutants with seed coat defects have reduced seed dormancy. Examples of the seed testa mutants are the transparent testa (tt) and transparent testa glabra (ttg) mutants, that are defective in flavonoid pigmentation, resulting in seed colors ranging from yellow to pale brown (Koornneef, 1981, 1990; Debeaujon et al., 2000), and mutants altered in testa structure such as aberrant testa shape (ats) and apetala2 (ap2) (Jofuku et al., 1994;


90

Léon-Kloosterziel et al., 1994). An exception to the seed testa mutants is the glabra2 (gl2) mutant, that shows a deformed testa surface structure (Bowman and Koornneef, 1994), and has a slightly increased seed dormancy level (Debeaujon et al., 2000). Seed coat mutants also presents pleiotropic effects on seed mucilage. Seed mucilage consists of viscous polysaccharides that encapsulate the seed during imbibition. There are different hypotheses for the role of seed mucilage in seed germination and thus seed dormancy. Mucilage may support seed hydration especially under water or salt stresses and serve as a water reservoir for germination, or inhibit germination by preventing oxygen permeation (reviewed in Western, 2012).

Most mutants that show a seed dormancy phenotype are also affected in seed longevity. Leafy Cotyledon1 (LEC1) and ABI3 are key regulators of seed maturation (Meinke et al., 1994; To et al., 2006) and seeds of the lec1 and abi3 mutants show a reduced seed dormancy and a rapid loss of viability upon storage (Ooms et al., 1993; Clerkx et al., 2004a; Sugliani et al., 2009). Both are the result of a defective seed maturation during which seed dormancy and longevity (desiccation tolerance) are not induced. The testa mutants, tts, ttg, ats, and ap2 also have reduced seed dormancy and reduced seed longevity levels (Debeaujon et al., 2000), likely due to the effect of a reduced capacity of the seed coat to function as a structural barrier in protecting the embryo and seed reserves from biotic and abiotic stresses (reviewed in Rajjou and Debeaujon, 2008). The dog1-1 mutant, which is a loss-of-function mutant, has a severely reduced seed longevity (Bentsink et al., 2006) but the mechanism of how DOG1 affects seed dormancy and longevity is still unclear.

The mutants discussed above are defective in seed dormancy and testify the power of forward genetics using mutagenesis-induced variation in identifying genes underlying seed dormancy. Seed dormancy in Arabidopsis thaliana was also studied using natural variation analysis. These analyses resulted in the identification of several DOG quantitative trait loci (QTL) (Alonso-Blanco et al., 2003; Bentsink et al., 2010). DOG6 was detected as the second strongest QTL affecting seed dormancy after DOG1, and the DOG6 dormancy effect was confirmed by NILDOG6-Sha (a NIL carrying a Shakdara [Sha] introgression at the position of DOG6 in Ler background) (Bentsink et al., 2010).

In this study, we report the identification of six novel seed dormancy mutants that have been isolated after ethyl methanesulfonate (EMS)-induced mutagenesis in the dormant NILDOG6-Sha background. These mutants display a reduced seed dormancy phenotype, and were named seed dormancy modifier (sdm), sdm1 to sdm6. The physiological and genetic characterization of these sdm mutants is presented.


91

Results

The isolation of seed dormancy mutantsDOG6 was one of the strongest QTLs controlling seed dormancy in Arabidopsis (Bentsink et al., 2010). The effect of DOG6 on seed dormancy was validated by NILDOG6-Sha. Seeds of NILDOG6-Sha are more dormant than seeds of the Ler genetic background (Bentsink et al., 2010). In order to identify modifiers of DOG6, mutagenesis induced by EMS was carried out on 10.000 NILDOG6-Sha seeds. Initially, the screen revealed 600 putative mutants with reduced dormancy, of which 106 and 68 were confirmed in, respectively, the M2 and M3 generation. To exclude false positive isolation of non-dormant mutants, the presence of DOG6-Sha were tested. Seeds of these mutants all germinated over 50% directly after harvest, while the control NILDOG6-Sha seeds did not germinate. Previous screens for reduced seed dormancy (Koornneef et al., 1982; 1984; Léon-Kloosterziel et al., 1996a) revealed mutants in ABA biosynthesis (ABA-deficient mutants, aba) and ABA sensing (abi) genes. To exclude these types of mutants, we examined the M3 and M4 mutants for sensitivity to 1 and 3 µM ABA (to detect abi mutants), and 1, 10 and 80 µM paclobutrazol, a chemical which inhibits GA biosynthesis (to distinguish aba mutants), respectively. These screens resulted in 31 mutants that had a reduced seed dormancy but were not abi either aba mutants. These 31 selected mutants were classified into seven groups based on their germination behavior to paclobutrazol. Allelism tests within and between these groups of mutants were performed in order to identify allelic mutations. This test revealed 25 mutants that after crossing to dog1-1 all appeared to be allelic to the non-dormant dog1 (Bentsink et al., 2006). The remaining six mutants are likely novel mutants that we have named seed dormancy modifier, sdm1 to sdm6.

Physiological characterization of the six seed dormancy modifier mutantsSeed dormancy and longevity analysesSeed dormancy levels of the six sdm mutants were measured during after-ripening (seed dry storage) to obtain the DSDS50 (days of seed dry storage required to reach 50% germination) values. All mutants showed reduced seed dormancy levels compared to NILDOG6, DSDS50 values of less than 15 compared with 45 days, respectively (Fig. 1A). Since reduced dormancy mutants often show a reduced seed longevity, we have also investigated the germination ability of the sdm mutants after artificial aging (0, 2, 4, 5, 6, and 7 days of treatment). Mutants sdm1, sdm3, sdm4, sdm5, and sdm6 exhibited a reduced seed longevity. Seeds of these mutants hardly germinated compared to NILDOG6 seeds that germinated for almost 80% after 4 days of treatment (Fig. 1C). Sdm2 did not show any defect in seed longevity in comparison with NILDOG6 (Fig. 1C).

Sensitivity to ABA and paclobutrazolTo study the relationship of the sdm mutants with ABA in more detail we analyzed the germination in response to a range concentrations of ABA and paclobutrazol. The


92

mutants displayed a similar ABA sensitivity as NILDOG6, in which 1 µM ABA could inhibit seed germination, only mutant sdm2 required higher ABA concentrations to stop germination (10 µM) (Fig. 1D). Although sdm2 seeds were still able to germinate 79% in 1 µM ABA, the majority of germinated seeds could not establish seedlings within 14 days after imbibition (DAI) (Fig. S1A).

Figure 1. Seed phenotypes of novel mutantsSeed dormancy in DSDS50 (A) and seed longevity in germination (%) after artificial aging for 0, 2, 4, 5, 6, and 7 days of treatment (C) were measured for NILDOG6 (NIL) and the six sdm mutants. ABA content in dry seeds (ng gDS-1) was also determined (B). Standard errors were calculated on four biological replicates. Seed mucilage of the six mutants in comparison with NILDOG6 when imbibed with water and stained with Ruthenium red 0.1% (F) are presented. Seed coat layers of the six mutants were imaged with scanning electron microscopy with magnification of 6000 (G) and 10000 (H) times.


93

Paclobutrazol inhibited the seed germination of sdm1, sdm3, sdm5, and sdm6 more than that of NILDOG6. Seed germination of those lines was strongly reduced to less than 25% in 1 µM paclobutrazol, whereas NILDOG6 seeds germinated for 73% (Fig. 1E). Seeds of sdm4 showed a sensitivity that was similar to NILDOG6. Seeds of sdm2 were more resistant to paclobutrazol; they were still able to germinate for 20% at 10 µM, however also here poor seedling establishment after 14 DAI was observed (Fig. S1B).

Endogenous ABA contentTo investigate if the seed dormancy and hormone sensitivity phenotypes can be explained by the endogenous ABA concentration, ABA measurement were performed in dry seeds of the six mutants. The sdm1, sdm3, and sdm4 mutant had a similar ABA level while sdm5 and sdm6 contained a slightly lower amount than NILDOG6 (Fig. 1B). Notably, sdm2 contained almost 1.5 times more ABA than NILDOG6 (Fig. 1B).

Seed mucilageMucilage is a pectinaceous layer excreted upon seed imbibition, which can be visualized by staining with ruthenium red. NILDOG6 seeds formed a normal mucilage layer, which is slimy and turns red when stained. The sdm2 and sdm4 seeds showed an altered phenotype (Fig. 1F). For seeds of sdm2, no mucilage layer was observed during seed imbibition, and also the seed coat did not stained red. Scanning electron microscopy (SEM) images of the seed coat of this mutant revealed a complete absence of the columella, which is the mucilage storage compartment (Fig. 1G and 1H). This suggests that mutant sdm2 carries a mutation in a gene affecting the mucilage synthesis pathway. In the case of sdm4, we only saw a thin, slimy layer of mucilage. The layer did stain with rhuthenium red, however, it was not formed firmly around the seed coat (Fig. 1E). Also the columella of the sdm4 seed coat cells was not mounted as high as that of NILDOG6 (Fig. 1G and 1H). The mucilage phenotype of sdm4 seeds could be caused by a gene having a role in the secretion or structural composition of mucilage.

Genetic analyses of sdm1 to sdm4Seed dormancy behavior of the F2 mapping populationsTo study the genetic behavior of genes underlying the seed phenotypes, F2 mapping populations were generated by crossing NILDOG6 with the mutants, sdm1, sdm2, sdm3, and sdm4. Mutant sdm5 and sdm6 were excluded from further analyses because they showed a certain level of male sterility. Seed dormancy behavior of the F3 seeds was evaluated during after-ripening. The frequency distributions of seed germination for mutant sdm1, sdm2, and sdm4 suggested monogenic segregations and a recessive behavior (3:1) of the mutant; heterozygous lines exhibited a low seed germination percentage similar to that of NILDOG6 (Fig. 2A, 2B and 2D). The F3 seed population of mutant sdm3 suggested a monogenic segregation, however, in a co-dominant manner, since there was a peak of germination percentages intermediate between the two parental peaks (Fig. 2C).


94

Figure 2. Seed dormancy distributions of F2 mapping populations for sdm mutantsSeed dormancy was measured in germination (%) during after-ripening of F2 mapping populations for the four mutants: sdm1 (A), sdm2 (B), sdm3 (C), and sdm4 (D) at 7, 1, 43 and 34 days after harvest, respectively. For each population, parental phenotypes are indicated by arrows with standard errors (horizontal bar beneath), black for NILDOG6 (NIL) and gray for the mutant.

Reciprocal crosses between NILDOG6 and each of the four mutants were performed to clarify the dominance effect of these novel seed dormancy genes. Reciprocal F1 seeds of all mutants had similar seed dormancy phenotypes to that of NILDOG6 (Fig. 3), which confirms that the mutant alleles are recessive. F1 seeds of the reciprocal crosses were only after-ripened for a short time, therefore, the co-dominant effect between NILDOG6 and sdm3 might not be seen (Fig. 3C).

Bulk segregant analyses for seed longevityIn addition to a dormancy phenotype, seeds of mutants sdm1, sdm3, and sdm4 also showed a seed longevity phenotype (Fig. 1C). In order to determine whether both phenotypes were a pleiotropic effect of the seed dormancy gene, we have performed bulk segregant analyses for seed longevity using artificial aging. F3 seeds from the mapping population that showed parental seed dormancy phenotypes of either NILDOG6 (Bulk-NIL) or the mutant (Bulk-mutant) were pooled. In all cases, seeds of Bulk-NIL germinated as good as those of NILDOG6, and Bulk-mutant resembled the seed longevity of the mutant (Fig. 4). These results indicated that seed dormancy and longevity were likely regulated by a single gene.


95

Figure 3. Seed dormancy of NILDOG6, sdm mutants and reciprocal F1 seedsSeed dormancy presented by germination (%) during after-ripening at 10 and 17 days after harvest (DAH) were measured for the parental NIL, sdm mutants and the corresponding reciprocal F1 between NIL and sdm1 (A), sdm2 (B), sdm3 (C), and sdm4 (D). Standard errors were calculated on at least three biological replicates.

Bulk segregant analyses for seed mucilageSdm2 and sdm4 showed altered seed mucilage phenotypes. In order to examine if these phenotypes were also caused by the seed dormancy gene, we used the before-mentioned bulks to phenotype for their seed mucilage. Seed mucilage was quantified as abnormal (0), segregating (1) or normal (2). For sdm2, Bulk-mutant exhibited a significantly lower score for seed mucilage compared to Bulk-NIL (Fig. 5A). Neither of the two bulks reached the NIL and sdm2 scores which could indicate that not sdm2 but a gene closely linked to sdm2 caused the seed mucilage phenotype. For sdm4, we did not observe a difference in seed mucilage score between Bulk-NIL and Bulk-mutant (Fig. 5B), indicating that also here seed dormancy and seed mucilage are regulated by two independent genes.

Discussion

Mutagenesis approaches are helpful in identifying genes that affect processes of interest, including germination (Koornneef and Van der Veen, 1980; Koornneef, 1981; 1990; Koornneef et al., 1982; 1984; Léon-Kloosterziel et al., 1996a; 1996b; Debeaujon et al.,


96

2000; Peeters et al., 2002; Bentsink et al., 2006). However the induction of mutations (modifier) in specific backgrounds (i.e. mutants) can also help to understand the genetic and molecular pathways by which genes in this particular background affect the trait of interest (Soppe et al., 1999). Moreover, mutants can facilitate the cloning of QTLs as in the case of DOG1. NILDOG1 was the template for a mutagenesis screen for non-dormant mutants. One of the mutants identified could genetically not be separated from DOG1 and therefore likely presented a mutation within the DOG1 gene (Bentsink et al. 2006).

Figure 4. Bulk segregant analyses for seed longevitySeed longevity was analyzed on bulk segregants for sdm1, sdm3, and sdm4. A bulk of lines showing similar seed dormancy to NILDOG6 (Bulk-NIL) and to the mutant (Bulk-mutant) were generated from an F2 mapping population. Seed dormancy during after-ripening (7, 43, and 34 days after harvest for sdm1, sdm3 and sdm4, respectively) and seed longevity after artificial aging were measured in germination (%) for NILDOG6 (NIL) and mutants (sdm1, A; sdm3, B; and sdm4, C) together with corresponding Bulk-NIL and Bulk-mutant. Artificial aging treatment was 5, 3, and 6 days for sdm1, sdm3, and sdm4, respectively. Averages and standard errors are presented.

Here we have mutagenized the dormant NILDOG6 line and screened for non-dormant mutants. In the M2, all mutants that showed a strong plant or seed (seed color) phenotype in addition to the seed dormancy phenotype were excluded. Furthermore, aba and abi mutants were also excluded by testing germination in the presence of paclobutrazol and ABA, respectively. Of the 31 mutants selected in this way 25 were


97

carrying mutations in the DOG1 gene. These dog1 mutants overcome the DOG6 dormant behavior, which indicates that DOG1 is genetically upstream of DOG6. The remaining six novel mutants (sdm1 to sdm6) are possible modifiers of DOG6. Although mutants exhibiting obvious plant phenotypes were eliminated, the sdm mutants showed slightly altered plant phenotypes compared with NILDOG6 (Fig. S2), which might be a result of background mutations or of pleiotropic effects of the seed dormancy gene.

Figure 5. Bulk segregant analyses for seed mucilageSeed mucilage was analyzed on bulk segregants for sdm2 and sdm4. A bulk of lines showing similar seed dormancy to NILDOG6 (Bulk-NIL) and to the mutant (Bulk-mutant) were generated from an F2 mapping population. Seed dormancy is presented as germination (%) at 1 and 34 days after harvest for, respectively, sdm2 and sdm4. Seed mucilage is presented on a quantitative scale: abnormal seed mucilage (0), mixture of normal and abnormal seed mucilage (1), and normal seed mucilage (2) for NILDOG6 (NIL) and the mutants (sdm1, A; and sdm5, B) together with corresponding Bulk-NIL and Bulk-mutant. Averages and standard errors are presented.

The seed dormancy modifier mutantsAll the sdms exhibited a reduced seed dormancy phenotype, that was caused by a single mutation as suggested by the segregation pattern of the F2 mapping populations for seed dormancy (Fig. 2). The seed dormancy alleles of sdm1, sdm2 and sdm4 are recessive, while that of sdm3 is co-dominant. This confirms that EMS can induce different dominance effects and not only recessive mutations as is often thought.

The reduced seed longevity observed in sdm1, sdm3, and sdm4 was caused by the seed dormancy gene as suggested by bulk segregant analyses (Fig. 4). This finding was as expected, and was reported for various mutants, such as seed development mutants (lec and abi) (Ooms et al., 1993; Clerkx et al., 2004a; Sugliani et al., 2009), seed testa mutants (tts, ttg, and ats) (Debeaujon et al., 2000), and seed dormancy mutants (rdo3 and dog1) (Bentsink et al., 2006; Liu et al., 2007). The identification of the genes underlying the mutations will reveal how seed dormancy and seed longevity are regulated and whether this is due to e.g. a general effect during seed maturation.


98

Sdm1 and sdm3The mutants sdm1 and sdm3 had similar physiological characteristics including similar seed longevity and ABA and paclobutrazol sensitivity levels (Fig. 1), however they showed a different genetic behavior, recessive for sdm1 and co-dominant for sdm3 (Fig. 2A and 2C). It is noted that these two mutants were more sensitive to paclobutrazol than NILDOG6 (Fig. 1D), indicating that they required more GA for seed germination than NILDOG6. This enhanced sensitivity is a remarkable finding since higher GA requirements have always been related to higher dormancy levels, as for ga1 (Koornneef and Van der Veen, 1980). Because of a similar ABA content and a similar ABA sensitivity to those of NILDOG6 (Fig. 1B), sdm1 and sdm3 mutants might be involved in GA signaling (over-sensitivity) rather than GA synthesis. However, the increased GA requirement can also be a result of an additional background mutation.

Sdm2Sdm2 is the only sdm mutant that exhibited wild type (NILDOG6) seed longevity levels (Fig. 1B) which indicates that dormancy is affected by a different mechanism in the sdm2 mutant. This is supported by the fact that sdm2 shows a lower sensitivity to both ABA and paclobutrazol compared to NILDOG6 and the other sdm mutants (Fig. 1D and 1E). The sdm2 mutant possesses a higher ABA content than NILDOG6 (Fig. 1B), and possibly the lower sensitivity to ABA and paclobutrazol is caused by an additional mutation in the sdm2 background that leads to the mucilage phenotype. Probably, the lack of mucilage hampered the membrane permeability and therefore of ABA and paclobutrazol uptake. Consequently, these compounds could not pass through the seed coat in order to block germination. Only upon testa rupture both compounds reached the embryo and thereby blocked seedling establishment (Fig. S1). These phenotypes are different from those of abi mutants, which can both germinate and produce normal seedlings in the presence of ABA.

Sdm4Sdm4 seeds were more sensitive to artificial aging than NILDOG6, however they were less severely affected than sdm1 and sdm3 (Fig. 1C). Sdm4 had a similar response to ABA and paclobutrazol, and a similar ABA content compared to NILDOG6 (Fig. 1B, 1D and 1E), thus it is very likely a modifier of the DOG6 dormancy phenotype. In the sdm4 seeds, an additional seed mucilage phenotype was observed (Fig. 1F), however the bulk segregant analyses indicated an independently mutated gene (Fig. 5B). Backcrossing sdm4 with NILDOG6 can separate the seed dormancy from the seed mucilage gene.

Conclusions and perspectives

In this chapter, the isolation and physiological characterization of the six reduced seed


99

dormancy mutants, sdm1 to sdm6, is presented. Unraveling the mechanism underlying seed dormancy and other seed related phenotypes such as seed longevity and seed mucilage requires gene identification. Efforts to identify the underlying genes have been started and a mapping-by-sequencing (SHOREmap) approach will be used (Schneeberger et al., 2009; James et al., 2013). This approach allows the identification of mutations by subjecting DNA pools of plants expressing mutant phenotypes to deep sequencing. The candidate mutations will be non-segregating alleles that differ from that of wild type. Once the genes underlying the mutation have been identified, further functional analyses will be performed, which will start with retrieving additional alleles of the mutants (T-DNA knock-out lines). Such additional alleles will reveal whether the seed longevity, ABA and paclobutrazol sensitivities, and plant phenotypes are indeed pleiotropic effects of the identified gene. The unambiguous gene confirmation will be performed by complementation experiments, for which we will transfer a wild type allele (i.e. from Columbia or Ler) to the recessive mutant background. In addition, expression analyses will be done to identify when and where the genes are expressed. Further analyses will depend on the nature of the identified gene.


MaterialNILDOG6 carries a Sha genomic fragment introgressed in Ler background spanning the DOG6 QTL interval (Bentsink et al., 2010). The F2 mapping populations for each of the sdm1, sdm2, sdm3, and sdm4 mutants were generated from F1 plants of crosses between NILDOG6 and the mutants.

Growth, harvest and storage conditionsPlants were grown in a controlled climate chamber at 20°C/18°C (day/night) under a 16-h photoperiod of artificial light (150 μmol m-2 s-1) and 70% relative humidity, on rockwool supplied with a hyponex solution. Plants of the F2 mapping population were grown in an air-conditioned greenhouse supplemented with additional light.

Plants were grown and harvested as single plants. Seeds were stored at ambient conditions, in 6 x 13 cm cellophane flat bags (Hera papierverarbeiting, Nidda, Germany) or in 6.5 x 10.5 cm paper bags (Koninklijke Van der Most BV, Heerde, The Netherlands) for the F3 seeds of mapping populations.

Mutant induction10.000 NILDOG6-Sha seeds were packed in miracloth that allows EMS treatment and washing without losing seeds. Seeds were incubated for 15 h at 4°C in 50 mL 0.1% KCL, after which they were washed with 100 mL demineralized water. Mutagenesis was performed by continuously shaking the seeds for 17 h in 100 mL of 30 mM EMS solution


100

in the dark. After this incubation, the EMS solution was removed and the seeds were washed three times with 100 mL of 100 mM Sodium thiosulphate, which breaks down the EMS. This washing is followed by rinsing the seeds in 500 mL of demineralized water for 30 min. Finally, seeds were added to an 0.2% (w/v) agar solution, in which they were evenly spread on the soil. In total, 86 trays with approximately 120 M1 seeds per pot have been grown into plants. Seeds of these plants (M2 seeds) have been harvested as bulks. Directly after harvest, the bulks have been tested for their germination behavior, individuals that germinated within 5 days (non-dormant mutants) have been selected and grown until seed harvest. The seeds of these plants (M3 seeds) were tested for their dormancy behavior again. In total, 106 mutants were confirmed (over 50% of germination at harvest compared to 0% of wild type NILDOG6-Sha).

Genotype analysesDNA of 106 plants was isolated from leaves according to the protocol described in Chapter 4. These 106 plants have been tested for the presence of the NILDOG6-Sha introgression by marker “nit1.2” amplified with forward primer “CGGAATTGATGTTTTGGACC” and reverse primer “CCCTACATTCTACAACCATGTAGCC”. Polymerase chain reactions (PCR) were performed as described in Chapter 4 with annealing temperature of 58°C. The polymorphic PCR products were visualized in 1.5% (w/v) agarose gel.

Phenotyping for seed germination, dormancy, longevity, and seed germination sensitivity to ABA and paclobutrazolGermination assays were performed according to Joosen et al. (2010) and are described in Chapter 4.

Seed dormancy was measured as germination percentage of seeds undergoing after-ripening, the higher the germination percentage, the lower the seed dormancy level. For the physiological characterization of the six sdms, seed dormancy was measured as DSDS50. DSDS50 was calculated according to Hurtado et al. (2012), described in Chapter 4.

Seed longevity was assessed after artificial aging when seeds had been fully after-ripened. A small amount of seeds (ca. 150 seeds) were aliquoted in 1.2 mL open tubes and stored above a saturated NaCl solution in a closed and ventilated tank (relative humidity of 85% and temperature of 40oC) for a period varying from 0 to 7 days. After treatment, germination assays were performed as described above, the germination percentage was a parameter for seed longevity, the higher the germination the better the seed longevity.

The level of sensitivity to ABA and pacloputrazol was evaluated by germination assays in the presence of 0, 1, 10, and 50 µM ABA, and of 0, 1, 10, and 80 µM paclobutrazol, respectively. ABA was dissolved in absolute ethanol to make an ABA 200 mM stock solution, from which the ABA solutions at different concentrations were obtained by


101

diluting with demineralized water. Paclobutrazol was dissolved in a few drops of acetone, after that demineralized water was added to make an stock of 80 µM. Thus appropriate concentrations were obtained by diluting stock solution with demineralized water.

Phenotyping for seed mucilageSeed mucilage can be revealed by staining with 0.1% (w/v) rhuthenium red. Seeds were imbibed on filter paper (Sartorius filter discs 3hw) wetted with distilled water in 6 cm-petri dish for about 15 min. Following, 100 µL of rhuthenium red was added, seeds were incubated for ca. 1 h. The presence of mucilage was visualized by red staining.

For the F2 mapping population, individuals were scored as either wild type, mutant or a segregating seed mucilage phenotype using a binocular after approximately 30 min of imbibition.

Scanning electron microscopy (SEM) revealed the surface structures of the seed coat. Dry seed were mounted on a stubs and coated with palladium-gold. Images were made with SEM model Leica VCT100 cryo-system.

ABA extraction and detectionABA purification was performed by adapting the protocol described by Zhou et al. (2003). 10 mg of dry seeds were frozen in liquid nitrogen and ground with a tissue lyser (Mo Bio Laboratory) at 25 Hz for 1.5 min with help of stainless steel beads. ABA was extracted with 1.5 mL of methanol/water/acetic acid (80:19:1) and 25 pmol/mL of [2H6]-ABA as internal standard in a 2 mL centrifuge tube. The tubes were vortexed and sonicated for 10 min in a Branson 3510 ultrasonic bath (Branson Ultrasonics, Danbury, CT, US). Samples were centrifuged for 10 min at 2500 g and the liquid phase was carefully transferred to a 4 mL glass vial. The samples were re-extracted with 1.5 mL of methanol/water/acetic acid (80:19:1). Both fractions were combined in a 4 mL glass vial and dried in a speedvac concentrator coupled tot a refrigerated vapour trap (Thermo Fisher Scientific). The residue was dissolved in 100 µL methanol/acetic acid 99:1 (v/v) and 900 µL 1% acetic acid in UPLC grade water. Undissolved particles were pelleted before loading the samples on HLB columns (Oasis®, Waters, 30 mg 1 cc) which were previously equilibrated with 1 mL 100% methanol (HPLC supra gradient) followed by 1 mL of methanol/water/acetic acid 10:89:1. The columns/samples were washed with 1 mL methanol/water/acetic acid 10:89:1. After washing, 1mL methanol/water/acetic acid 80:19:1 was added to the columns and the flow through was collected. The samples were dried in a speedvac and re-suspended in 100 μL UPLC grade water/acetonitrile/formic acid 94.9:5:0.1. The samples were stored at -20°C until measurement.

ABA analysis was performed with a Waters Xevo tandem quadrupole mass spectrometer equipped with an electrospray ionization source and coupled to an Acquity UPLC system (Waters, USA). Chromatographic separation was achieved using an Acquity UPLC BEH C18 column (100 x 2.1 mm, 1.7 μm) (Waters, USA), applying a


102

water/acetonitrile gradient, starting at 5% acetonitrile for 1.0 min, raised to 98% (v/v) acetonitrile in 5 min which was then maintained for 4 min before returning to 5% acetonitrile in water using a 0.13 min gradient. The column was equilibrated at this solvent composition for 1.87 min prior to the next injection. Total run time was 11 min. The column was operated at 50oC with a flow-rate of 0.2 mL min-1 and sample injection volume was 10 μL. The mass spectrometer was operated in positive electrospray ionization (ESI) mode. The cone and desolvation gas flows were 50 and 1000 L h-1, respectively. Argon was used for fragmentation by collision-induced dissociation in the ScanWave collision cell. The capillary voltage was set at 3 kV, the source temperature at 150oC and the desolvation gas temperature at 550oC. The cone voltage (CV), collision energy (CE) and parent daughter transitions were optimized by injecting pure ABA and [2H6]-ABA in the Waters IntelliStart MS console. Transitions were selected based on the most abundant and specific fragment ions. For ABA, the Multiple reaction monitoring (MRM) transitions m/z 265.16 > 173.07 at a CE of 18 eV, 265.16 > 187.05 at 18 eV, 265.16 > 229.15 at 10 eV and 265.16 > 247.17 at 6 eV; and for [2H6]-ABA, the transitions m/z 271.22 > 234.49 at CE of 10 eV, 271.22 > 253.24 at CE of 6 eV. The cone voltage for all transitions was 10V. ABA was quantified using a calibration curve with known amount of ABA based on the ratio of the summed area of the MRM transitions for ABA to those for [2H6]-ABA. Data acquisition was performed using MassLynx 4.1 software (Waters, USA).

Genetic analysis of the isolated mutantsReciprocal crosses between NILDOG6 and sdm1, sdm2, sdm3, and sdm4 were performed. Reciprocal F1 seeds were produced by hand pollination, anthers were removed from the mother of the cross and pollens of the father were artificially placed on top of the stigma of mature flowers. Approximately 19 days after pollination, seeds were harvested.

Bulk segregant analysesSeeds of 12 to 17 plants of the F2 mapping population, that resemble either the parental seed dormancy phenotype, were pooled to generate Bulk-NIL (dormant as NILDOG6) and Bulk-mutant (non-dormant as the mutant).The bulks were analyzed for seed longevity after artificial aging. For mutant sdm1, seed longevity of the bulks were averages of seed longevity from 15 individuals of the bulk. Seed mucilage was quantified in the range of 0, 1, and 2. Thus plants that had all seeds with abnormal mucilage were scored 0; plants that had all seeds with normal mucilage were scored 2; and plants that had mixture of seeds with both abnormal and abnormal mucilage were scored 1. The mutants and NILDOG6 were also scored based on these criteria. Mucilage of the bulks were averages of mucilage scores from individual plants selected in the bulks.


103

Supplemental data


Figure S1. Seed germination and seedling establishment 14 days after imbibition of mutant sdm2 in response to 1 µM ABA and 10 µM paclobutrazol

Figure S2. Plant phenotypes of NILDOG6 and sdm mutants.

General discussion

6Chapter

General discussion Chapter 6

106

Seed dormancy and seed longevity are the most important survival traits in the soil seed bank. Both traits are induced during seed maturation and evolved to assure seed survival during environmental conditions that cannot support the regular course of life. Seed dormancy is related to the timing of germination while seed longevity is involved in retaining germination ability that is gradually lost as a result of aging. The germination ability during the seed life span in laboratory dry storage conditions can be presented as a bell-shaped curve, with on one side seed dormancy processes (maintenance and/or release) and on the other side seed longevity related processes (Box 1). The work presented in this thesis aims to identify novel genetic factors that regulate seed dormancy and seed longevity by generating and exploiting genetic variation.

Quantitative trait locus (QTL) analyses in six recombinant inbred line (RIL) populations for seed longevity measured as seed germination ability after long-term storage at ambient conditions, revealed nine QTLs (Chapter 2). These QTLs were named Germination Ability After Storage (GAAS). Comparison between GAAS loci and seed dormancy Delay Of Germination (DOG) QTLs earlier established in the same RIL populations (Bentsink et al., 2010), led to the identification of collocating loci for which the allele effects were negatively correlated. The QTL effects of the GAAS loci and their colocation with the DOGs were validated in near isogenic lines (NILs). To our knowledge this is the first time that such a negative correlation between dormancy and longevity was reported, deep seed dormancy correlating with low seed longevity and vice versa. Detailed analysis on the collocating GAAS5 and DOG1 loci revealed that the DOG1-Cape Verde Islands (Cvi) allele both reduces seed longevity and increases seed dormancy.

Chapter 3 combines a proteomics and genetics approach to identify molecular mechanisms as well as genetic factors that are involved in seed aging by performing proteome profiling on seeds of four genotypes at two physiological states (after-ripened [AR] and aged). For that, we have investigated the three strongest GAAS loci (GAAS1, GAAS2, and GAAS5) present in the NILs and the Landsberg erecta (Ler) genetic background. During aging, dry seed proteomes are markedly changed in a genotype specific manner, implying that different mechanisms are involved in seed longevity conferred by the three NILs. The shared pathways revealed the importance of seed reserves (seed storage proteins, SSPs), antioxidant systems notably vitamin E, as well as the protection and maintenance of the translation machinery, and energy related pathways in seed longevity. Reverse genetics using T-DNA knock-out mutants validated that SSPs cruciferins and napins, the small ribosomal subunit RPS12C, and the NADP-dependent malic enzyme1 (NADP-ME1) significantly affect seed longevity.

Chapter 4 presents investigations on the negative correlation between seed dormancy and seed longevity at the DOG2/GAAS1 locus on top of chromosome 1. The effects of both DOG2 and GAAS1 were confirmed in NILDOG2/GAAS1-Cvi, that had a lower level of seed dormancy but a better seed longevity than Ler. Genetic analyses


107

were performed, which showed that DOG2 and GAAS1 were likely two independent loci due to the differences in genetic behavior of DOG2 (maternal) and GAAS1 (co-dominant). The fine-mapping of DOG2 resulted in a 85 Kb-region containing 27 genes. The location of GAAS1 requires further investigation.

DOG6 is one of the strongest seed dormancy QTLs found in the six RIL populations, of which the effect was confirmed by NILDOG6-Shakdara (Sha). Genetic variation for seed dormancy was generated by ethyl methanesulfonate-induced mutagenesis in the dormant NILDOG6-Sha background in order to identify novel reduced seed dormancy mutants (Chapter 5). Six seed dormancy modifier mutants (sdm1 to sdm6) were isolated. The sdm1, sdm3, sdm4, sdm5, and sdm6 mutants showed reduced seed longevity as a pleiotropic effect of the mutated seed dormancy gene. Mutants sdm2 and sdm4 exhibited defects in seed mucilage. Mapping populations were generated and phenotyped in order to identify the underlying sdm genes by a mapping-by-sequencing approach.

The relation between seed dormancy and seed longevity and its ecological significance

The effects of seed dry storage on seed dormancyFor many species, dormancy can be alleviated after a prolonged period of seed dry storage (AR), which results in a widening of the environmental conditions that permit germination. Loss of dormancy during AR is generally promoted by the same factors that affect seed aging, thus AR and seed aging are two physiological processes which could be occurring concurrently (Baskin and Baskin, 1998). In fact, AR could be considered as the first stage of aging in seeds exhibiting primary dormancy.

Dry AR is generally effective in a range of seed moisture contents (between 5% and 15 to 18% on a fresh weight basis), that corresponds to region two of the sorption isotherms (i.e., weakly bound water) (Bazin et al., 2011). Hardly anything is known about the molecular mechanisms that control AR. However, non-enzymatic mechanisms are likely candidates for seed dormancy release in anhydrobiotic conditions. Oxygen can diffuse within glasses, such as the vitreous cytoplasm at low seed moisture levels, and ultimately lead to reactive oxygen species (ROS) accumulation, as has been shown by Oracz et al. (2007). This accumulation of ROS was associated with seed dormancy release by AR in sunflower (Helianthus annuus). It was demonstrated that ROS caused lipid peroxidation and carbonylation of a specific subset of proteins that were associated with seed dormancy release. The most pronounced post-translational modification (PTM) targeting proteins during AR is carbonylation (Arc et al., 2011). This type of PTM can lead to changes in the biochemical properties of proteins and thereby affect enzymatic and binding activities and as a result promote protein degradation or increase


108

the sensitivity to proteolysis. Job et al. (2005) suggested that specific carbonylation of seed storage proteins facilitates proteolytic reactions to remobilize these nutrition resources during germination. Nakabayashi et al. (2012) showed that the DOG1 protein was altered during AR, which led to the relieve of seed dormancy. Lipids and proteins are not the only targets of oxidative modifications by ROS. Nucleic acids, in particular, are very sensitive to free radicals, in which RNA is more susceptible than DNA; and among all RNA species, mRNA is the most sensitive to oxidation resulting in 8-oxo-7,8-dihydroguanine (8-OHG) formation (Bazin et al., 2011). Selective oxidation of a specific subset of stored mRNAs was observed during AR in wheat (Triticum aestivum) (Gao et al., 2013) and sunflower (Bazin et al., 2011). In addition, the antioxidant defense system in wheat was shown to be associated with the maintenance of seed dormancy (Bykova et al., 2011). Controlled and active mechanisms operate during AR in dry seeds and are reflected by changes in transcripts and abundance of proteins. Transcriptome analyses have shown that AR affects the abundance of specific transcripts (Finch-Savage et al., 2007; Carrera et al., 2008) hypothetically occurring in humid pockets in dry seed. Leubner-Metzger (2005) showed that the transient transcription and translation of β-1,3-glucanase, that contributed to endosperm weakening and testa rupture, were increased in dry tobacco (Nicotiana tabacum) seeds during dormancy release. Proteins related to metabolic functions and energy metabolism were up-accumulated during AR of dry seeds (Chibani et al., 2006).

The effects of seed dry storage on seed longevityOxidation appears to be beneficial for dormancy release, however, accumulation of oxidative stresses during extended seed dry storage can ultimately cause deterioration (aging) and loss of viability as a consequence of nucleic acid, protein, and lipid modification and degradation. Prevention of deleterious oxidation reactions depends on active molecular mechanisms. DNA ligases (LIG4 and LIG6) are in charge of genome integrity maintenance, thus mutants of these genes show a reduced seed longevity in Arabidopsis (Waterworth et al., 2010) whether these mutants also show a dormancy phenotype has not been reported. Recently, Chen et al. (2012) reported that OGG1 over-expression enhanced seed longevity in Arabidopsis. OGG1 encodes a bi-functional DNA glycosylase/AP lyase, which removes 8-OHG to repair DNA damage. The protein repairing enzyme, protein-L-isoaspartate methyltransferase (PIMT), limits and repairs age induced damage to aspartyl and asparaginyl residues in proteins. Over-expression of PIMT1 and PIMT2 enhanced seed longevity in Arabidopsis (Oge et al., 2008; Verma et al., 2013). Protein oxidation occurs also at methionine residues, that are oxidized to methionine sulfoxide. This oxidation is reversed by methionine sulfoxide reductases (MSRs), of which protein abundance and enzymatic capacity are strongly linked to seed longevity in both Medicago (Medicago truncatula) and Arabidopsis (Chatelain et al., 2013). Non-enzymatic lipid oxidation causes oxidative stress and reduction of


109

lipid oxidation by antioxidants, such as vitamin E, prevents seed deterioration. Mutants affected in the vitamin E synthesis genes, vte1 and vte2, exhibited a reduced seed longevity (Sattler et al., 2004).

Whether the above-mentioned enzymes are already functional during seed dry storage or whether they become active upon seed imbibition remains to be addressed.

The relation between dormancy and longevityIt was previously thought that seed dormancy and seed longevity are positively correlated. This notion was mainly based on mutant studies. The leafy cotyledon1, abscisic acid insensitive3 (Ooms et al., 1993; Clerkx et al., 2004a; Sugliani et al., 2009), transparent testa, and aberrant testa shape mutants (Debeaujon et al., 2000), as well as the loss-of-function mutant in the DOG1 gene (Bentsink et al., 2006), and the green seed mutant (enhancer of abi3-1) (Clerkx et al., 2003) all have reduced dormancy levels that correlate with reduced seed longevity. The same was observed for the novel sdm1, sdm3, sdm4, sdm5, and sdm6 mutants, for which the isolation is presented in this thesis (Chapter 5). Moreover, the earlier mentioned factors (ROS, mRNA oxidation, carbonylation, and other PTMs) that relieve seed dormancy also induce seed deterioration and as a consequence reduce seed longevity (Box 1). The detoxification by antioxidants prolongs seed AR and also prevents damages during aging to extend seed longevity (Box 1).

In contrast to the above-reported positive correlation, we have recently revealed a negative relation between seed dormancy and seed longevity for natural alleles of DOG and GAAS loci (Nguyen et al., 2012; Chapter 2). At the DOG1 (the only seed dormancy QTL cloned) and GAAS5 position, it was shown that a single gene increases seed dormancy whereas it decreases seed longevity (Nguyen et al., 2012; Chapter 2). The negative correlation is not only limited to collocating loci. Recently, the group of Barazani reported a negative correlation between seed dormancy and seed longevity in the winter annual Eruca sativa (Barazani et al., 2012; Hanin et al., 2013). Their first study showed that Eruca sativa populations, collected in Israel along a climate gradient, display increased seed dormancy levels with increasing aridity (Barazani et al., 2012). In a second paper, it was revealed that the more dormant seeds of the desert population deteriorated significantly faster than those of the less dormant semiarid population (Hanin et al., 2013). Thus, also here the more dormant populations are less storable.

Ecological roleThe presence of natural variation for seed dormancy and seed longevity suggests a role of the environment in the selection for these traits. This makes sense since the timing of germination also influences the expression of other life-history traits in the plant’s life cycle (Evans and Cabin, 1995; Donohue, 2002; Wilczek et al., 2009). Ecophysiological studies have shown the complexity of the conditions involved in the regulation of


110

Box 1. Overview of processes that affect the seed life span.The germination ability of dry seeds is presented as a bell-shaped curve (the dashed-line curve). Seed DORMANCY RELEASE is reflected by an increased germination ability during seed dry storage; and seed longevity is reflected by a loss of germination ability after even longer storage (AGING). The main factors that affect both seed dormancy and seed longevity are indicated by separate boxes (Maturation and Storage). During seed maturation, low temperature (T) enhances seed dormancy and reduces seed longevity. Mutations in key seed maturation regulators (abi3, fus3, and lec1) reduce both seed dormancy and seed longevity. During seed dry storage, high temperature and relative humidity (RH) facilitates dormancy release as well as the rate of aging, likely because of oxidative stress. Oxidative stresses are increased due to the accumulation of reactive oxygen species (ROS), which cause oxidation of proteins, DNA, mRNA, and lipid. Oxidation appears to have beneficial effects on seed dormancy release, however oxidative stress facilitates seed aging. Antioxidants (VTC, VTE, GHS, TPX, CAT, SOD, APX, and MSR) scavenge ROS to prevent oxidation, as a result, inhibit dormancy release as well as seed aging. DOG1/GAAS5 enhances seed dormancy and reduces seed longevity. Effects of oxidation stress on seed longevity is counteracted by PIMT1 and PIMT2, that repair oxidized (ox.) proteins; by OGG1 that repairs oxidized genetic material (DNA and mRNA); by LIG6 and LIG4 that repair DNA breakage; and by VTE that prevents lipid oxidation. It remains unclear when these genetic factors are induced (in maturation or in dry storage) and when their effects emerge (in dry storage or during seed imbibition).Abbreviation: abi3, abscisic acid insensitive; fus3, fusca3; lec1, leafy cotyledon1; VTC, vitamin C; VTE, vitamin E; GHS, glutathione; TPX, thioredoxin-dependant peroxidase; CAT, catalase; SOD, superoxide dismutase; APX, ascorbate peroxidase, MSR, methionine sulfoxide reductases; DOG, Delay Of Germination; GAAS, Germination Ability After Storage; PIMT, protein-L-isoaspartate methyltransferase; OGG1, DNA glycosylase/AP lyase; and LIG, DNA ligase.

seed germination and plant survival in the Mediterranean environment (Lloret et al., 1999; Gutterman, 2002; Petru and Tielborger, 2008), and highlighted the importance


111

of seed dormancy as an adaptive trait that enables plant species to persist in extreme environments (Freas and Kemp, 1983; Kigel and Galili, 1995; Gutterman, 2002). Studies on adaptation processes in Arabidopsis revealed that effects of seed dormancy on plant fitness were due to fixing new beneficial mutations (Huang et al., 2010). One of these studies focused on DOG1, which provides a unique opportunity to study allelic diversity since it was found to collocate with QTLs for germination timing and fitness in the field, performed on Spanish, French, Norwegian, and Central Asian populations (Huang et al., 2010). Moreover, Kronholm et al. (2012) suggested that dormancy is adaptive and probably acts through DOG1. The question whether the variation for seed longevity is also due to ecological adaptation has not been addressed so far. However, the fact that we found colocation between DOG and GAAS loci, and that in the case of DOG1, one gene regulates both seed dormancy and seed longevity, indicates that the selection can be on either one of them.

Ecological adaptation is driven by environmental factors that, during seed maturation, have been reported to affect seed dormancy (Fenner, 1991; Baskin and Baskin, 1998; Tollenaar, 1999; Gutterman, 2000). For example, high temperatures during seed development are generally associated with lower dormancy levels in lettuce (Lactuca sativa) (Drew and Brocklehurst, 1990) and wild oat (Avena fatua) (Sawhney et al., 1985), which could be caused by reduced synthesis of inhibitory compounds at high temperature (e.g., abscisic acid), or greater synthesis of promoting substances (e.g., gibberellins). Day-length and light quality (wavelength composition) have been reported to influence germination ability during seed development (Fenner, 1991; Baskin and Baskin, 1998; Gutterman, 2000). Longer days have, in most cases, been associated with decreased germination ability and higher dormancy, although exceptions have been reported (Fenner, 1991; Baskin and Baskin, 1998). Seeds developed under light environments with reduced red to far-red light ratios have lower germination in the dark, compared with seed developed under red light rich conditions (McCullough and Shropshire, 1970; Hayes and Klein, 1974). Recently He et al. (submitted) have reported on the relation between seed dormancy and seed longevity in various seed maturation environments. The authors showed that temperature, light and nitrate regimes affect seed dormancy and seed longevity in a similar way to the earlier described genetic relation; high temperature, high light, and high nitrate reduce seed dormancy and increase seed longevity, and vice versa low temperature, low light, and low nitrate increase seed dormancy and decrease seed longevity.

The trade-off between seed dormancy and seed longevityIn this section, three possible hypotheses are presented to explain the negative correlation (trade-off) between seed dormancy and seed longevity. The first two are formulated based on the role of oxygen in seed dormancy release and seed aging (Bewley and Black, 1982; Bazin et al., 2011). Hypothesis one proposes that dormancy release


112

requires seeds that are sensitive to oxidation. Likely, the higher oxidation sensitivity makes dormant seeds more prone to further damage, as dry storage continues, and therefore seeds are more sensitive to aging. Hypothesis one is supported by the DOG1 case, where the modification of the DOG1 protein during AR is expected to lead to a non-functional variant in AR seeds (Nakabayashi et al., 2012). As a result, these seeds become non-dormant and are also less storable. This shows again the importance of DOG1 in both seed dormancy and seed longevity, which is confirmed by the loss-of-function dog1 mutant, that is non-dormant and badly storable (Bentsink et al., 2006). Hypothesis two is based on the assumption that seeds contain a certain level of antioxidants and pro-oxidants for scavenging and buffering oxidation effects. Dry storage leads to increased ROS levels, thus dormant seeds employ their anti- and pro-oxidant properties already during AR, resulting in reduced protection during further storage. Consequently, dormant seeds are rapidly exposed to oxidative stresses, and more sensitive to aging. Hypothesis three is related to energy consumption. It has been debated whether dry seeds consume energy because enzymatic reactions take place in water activities above 0.5 (corresponding to approximately 50% relative humidity) (Labuza, 1971). Dry seeds contain up to 7% moisture content and therefore likely do not use energy (Baud et al., 2002). However, seed dry proteome investigations demonstrated an increase in glyceraldehyde-3-P dehydrogenase enzyme abundance in aged seeds (Rajjou et al., 2008). This enzyme is involved in the production of 3-phosphoglycerate that also increases in oxidative stressed cells (Baxter et al., 2007). Chibani et al. (2006) observed accumulation of proteins related to metabolic functions and energy metabolisms in dry AR seeds. The enzyme activity of β-1,3-glucanase was associated with dormancy release in dry tobacco seed (Leubner-Metzger, 2005). We hypothesize that the release of dormancy is an energy requiring process. Therefore more dormant seeds will invest more energy in the AR. The remaining energy is insufficient for deterioration-related repair and maintenance during further storage or germination, making dormant seeds less storable.

We have suggested that the negative correlation between seed dormancy and seed longevity is a pleiotropic effect of a single gene, which explains the colocation of the two traits, and describes a reduced longevity as a negative side effect of an increased dormancy level. So far this hypothesis has only been proven for DOG1. This pleiotropic effect is also relevant for the less dormant NILDOG22/GAAS2 genotype presented in Chapter 2 (Nguyen et al., 2012). The better longevity could be a positive site effect of the reduced dormancy of this genotype, taking the three earlier discussed hypotheses in mind. Pleiotropism could also explain the effect of the maternal environment (He et al., submitted). DOG1 expression and dormancy behavior was tested in a set of accessions that were grown in different maternal environments and revealed an increase in DOG1 expression when seeds matured at low temperature, which was associated with


113

an increased dormancy and a reduced seed longevity (Chiang et al., 2011; Kendall et al., 2011; He et al., submitted). This finding indicates the role for the environment in determining the appropriate dormancy or longevity level after seed dispersal.

Independent genetic factors that either affect seed dormancy or seed longevity

Although, seed dormancy and longevity share several underlying mechanisms, as suggested by the colocation of seed dormancy and seed longevity loci (Nguyen et al., 2012; Chapter 2), only seven of eleven identified DOG and seven of twelve GAAS loci collocate, indicating that there are independent genetic elements that affect either one of the two traits. Moreover, the fine-mapping of DOG2 and GAAS1, which are in close vicinity, revealed that these are two independent genes that control seed dormancy and seed longevity, respectively (Chapter 4). Studies on non-collocating DOG and GAAS loci, together with the reduced-dormant sdm2 mutant that is not defected in seed longevity (Chapter 5) will help to understand these specific mechanisms.

Seed proteome investigation in dry seeds also revealed seed longevity factors such as seed storage proteins (SSPs), the small ribosomal subunit RPS12C and NADP-ME1 (Chapter 3). The SSP cruciferein crua cruc double mutant and crua crub cruc triple mutant, the SSP napin RNAi-napin line, and the nadp-me1 mutant are more sensitive to seed aging. The rps12C mutant has a better seed performance after artificial aging treatment. Seed dormancy of these mentioned mutants have been examined; and are not different from that of wild type Columbia. Muntz et al. (2001) suggested that SSPs contributed to seed germination vigor and support early seedling growth when mobilized upon germination. Surprisingly, the markedly reduced amount of SSPs in the crua cruc double and the crua crub cruc triple mutant and the RNAi-napin line (Withana-Gamage et al., 2013) do not affect seed dormancy. Possibly, the effect of reduced amounts of SSPs only becomes apparent after prolonged storage during which they have a role in ROS buffering. This also explained their high abundance in Arabidopsis seeds and the wide range of PTMs (Job et al., 2005; Wan et al., 2007). Rajjou et al. (2004) demonstrated the importance of translation in seed germination. The translation related rps12C mutant expressed an increased seed longevity probably due to the activation of redundant genes in the family. Oxidation during AR might not severely damage functional proteins required for germination, thus synthesis of new elements can be sustained at a low but functional level, which limits the effect of a mutation in a single translation factor on the dormancy phenotype. However, aging stress might increase the need for newly synthesized proteins to renew damaged proteins. A competent translation machinery is required for aged seeds to prepare for germination. The reduced seed longevity of the


114

nadp-me1 might be related to energy deficiency in repair, maintenance, and synthesis of germination systems in aged seeds.

Concluding remark

Results presented in this thesis contribute a novel concept to the current notion on the relation between seed dormancy and seed longevity. The power of natural and induced genetic variation in the identification of underlying mechanisms is exploited, which will facilitate the future understanding of seed dormancy and seed longevity.


References

ReferencesReferences

116

Ali-Rachedi S, Bouinot D, Wagner MH, Bonnet M, Sotta B, Grappin P, Jullien M (2004) Changes in endogenous abscisic acid levels during dormancy release and maintenance of mature seeds: studies with the Cape Verde Islands ecotype, the dormant model of Arabidopsis thaliana. Planta 219: 479-488

Alonso-Blanco C, Bentsink L, Hanhart CJ, Vries HBE, Koornneef M (2003) Analysis of natural allelic variation at seed dormancy loci of Arabidopsis thaliana. Genetics 164: 711-729

Alonso-Blanco C, Koornneef M (2000) Naturally occurring variation in Arabidopsis: an underexploited resource for plant genetics. Trends Plant Sci 5: 22-29

Arc E, Chiban K, Grappin P, Jullien M, Godin B, Cueff G, Valot B, Balliau T, Job D, Rajjou L (2012) Cold stratification and exogenous nitrates entail similar functional proteome adjustments during Arabidopsis seed dormancy release. J Proteome Res 11: 5418-5432

Arc E, Galland M, Cueff G, Godin B, Lounifi I, Job D, Rajjou L (2011) Reboot the system thanks to protein post-translational modifications and proteome diversity: how quiescent seeds restart their metabolism to prepare seedling establishment. Proteomics 11: 1606-1618

Bailly C (2004) Active oxygen species and antioxidants in seed biology. Seed Sci Res 14: 93-107

Bailly C, Benamar A, Corbineau F, Come D (1996) Changes in malondialdehyde content and in superoxide dismutase, catalase and glutathione reductase activities in sunflower seeds as related to deterioration during accelerated aging. Physiol Plant 97: 104-110

Bailly C, El-Maarouf-Bouteau H, Corbineau F (2008) From intracellular signaling networks to cell death: the dual role of reactive oxygen species in seed physiology. C R Biol 331: 806-814

Barazani O, Quaye M, Ohali S, Barzilai M, Kigel J (2012) Photo-thermal regulation of seed germination in natural populations of Eruca sativa Miller (Brassicaceae). J Arid Environ 85: 93-96

Baskin CC, Baskin JM (1998) Seed: ecology, biogeography, and evolution of dormancy and germination. Academic Press, San Diego

Baud S, Boutin JP, Miquel M, Lepiniec L, Rochat C (2002) An integrated overview of seed development in Arabidopsis thaliana ecotype WS. Plant Physiol Biochem 40: 151-160

Baxter CJ, Redestig H, Schauer N, Repsilber D, Patil KR, Nielsen J, Selbig J, Liu J, Fernie AR, Sweetlove LJ (2007) The metabolic response of heterotrophic Arabidopsis cells to oxidative stress. Plant Physiol 143: 312-325


117

Bazin J, Langlade N, Vincourt P, Arribat S, Balzergue S, El-Maarouf-Bouteau H, Bailly C (2011) Targeted mRNA oxidation regulates sunflower seed dormancy alleviation during dry after-ripening. Plant Cell 23: 2196-2208

Beaudoin N, Serizet C, Gosti F, Giraudat J (2000) Interactions between abscisic acid and ethylene signaling cascades. Plant Cell 12: 1103-1115

Bentsink L, Alonso-Blanco C, Vreugdenhil D, Tesnier K, Groot SPC, Koornneef M (2000) Genetic analysis of seed-soluble oligosaccharides in relation to seed storability of Arabidopsis. Plant Physiol 124: 1595-1604

Bentsink L, Hanson J, Hanhart CJ, Blankestijn-de Vries H, Coltrane C, Keizer P, El-Lithy M, Alonso-Blanco C, de Andres MT, Reymond M, van Eeuwijk F, Smeekens S, Koornneef M (2010) Natural variation for seed dormancy in Arabidopsis is regulated by additive genetic and molecular pathways. Proc Natl Acad Sci USA 107: 4264-4269

Bentsink L, Jowett J, Hanhart CJ, Koornneef M (2006) Cloning of DOG1, a quantitative trait locus controlling seed dormancy in Arabidopsis. Proc Natl Acad Sci USA 103: 17042-17047

Bettey M, Finch-Savage WE (1998) Stress protein content of mature Brassica seeds and their germination performance. Seed Sci Res 8: 347-355

Bewley JD (1997) Seed germination and dormancy. Plant Cell 9: 1055-1066

Bewley JD, Black M (1982) Physiology and Biochemistry of Seeds. Plantum Press, New York

Bewley JD, Bradford KJ, Hilhorst HWM, Nonogaki H (2012) Seeds: Physiology of Development, Germination and Dormancy, Ed 3rd edition. Springer-Verlag, New York

Bleecker AB, Estelle MA, Somerville C, Kende H (1988) Insensitivity to ethylene conferred by a dominant mutation in Arabidopsis thaliana. Science 241: 1086-1089

Blokhina O, Virolainen E, Fagerstedt KV (2003) Antioxidants, oxidative damage and oxygen deprivation stress: a review. Ann Bot 91 Spec No: 179-194

Bowman JL, Koornneef M (1994) Seed morphology. Springer-Verlag, New York

Bray CM, West CE (2005) DNA repair mechanisms in plants: crucial sensors and effectors for the maintenance of genome integrity. New Phytol 168: 511-528

Browse J, Somerville CR (1994) Glycerolipids. In E Meyerowitz, C Somerville, eds. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, pp 881-912

Buitink J, Leprince O, Hemminga MA, Hoekstra FA (2000) Molecular mobility in the cytoplasm: an approach to describe and predict lifespan of dry germplasm. Proc Natl Acad Sci USA 97: 2385-2390


118

Bykova NV, Hoehn B, Rampitsch C, Banks T, Stebbing JA, Fan T, Knox R (2011) Redox-sensitive proteome and antioxidant strategies in wheat seed dormancy control. Proteomics 11: 865-882

Cadman CSC, Toorop PE, Hilhorst HWM, Finch-Savage WE (2006) Gene expression profiles of Arabidopsis Cvi seeds during dormancy cycling indicate a common underlying dormancy control mechanism. Plant J 46: 805-822

Cao DN, Cheng H, Wu W, Soo HM, Peng JR (2006) Gibberellin mobilizes distinct DELLA-dependent transcriptomes to regulate seed germination and floral development in Arabidopsis. Plant Physiol 142: 509-525

Carrera E, Holman T, Medhurst A, Dietrich D, Footitt S, Theodoulou FL, Holdsworth MJ (2008) Seed after-ripening is a discrete developmental pathway associated with specific gene networks in Arabidopsis. Plant J 53: 214-224

Carrera E, Holman T, Medhurst A, Peer W, Schmuths H, Footitt S, Theodoulou FL, Holdsworth MJ (2007) Gene expression profiling reveals defined functions of the ATP-binding cassette transporter COMATOSE late in phase II of germination. Plant Physiol 143: 1669-1679

Carroll AJ, Heazlewood JL, Ito J, Millar AH (2008) Analysis of the Arabidopsis cytosolic ribosome proteome provides detailed insights into its components and their post-translational modification. Mol Cell Proteomics 7: 347-369

Chatelain E, Satour P, Laugier E, Ly Vu B, Payet N, Rey P, Montrichard F (2013) Evidence for participation of the methionine sulfoxide reductase repair system in plant seed longevity. Proc Natl Acad Sci USA 110: 3633-3638

Chen H, Chu P, Zhou Y, Li Y, Liu J, Ding Y, Tsang EW, Jiang L, Wu K, Huang S (2012) Overexpression of AtOGG1, a DNA glycosylase/AP lyase, enhances seed longevity and abiotic stress tolerance in Arabidopsis. J Exp Bot 63: 4107-4121

Cheung WY, Hubert N, Landry BS (1993) A simple and rapid DNA microextraction method for plant, animal, and insect suitable for RAPD and other PCR analyses. PCR Methods Appl 3: 69-70

Chiang GCK, Bartsch M, Barua D, Nakabayashi K, Debieu M, Kronholm I, Koornneef M, Soppe WJJ, Donohue K, de Meaux J (2011) DOG1 expression is predicted by the seed-maturation environment and contributes to geographical variation in germination in Arabidopsis thaliana. Mol Ecol 20: 3336-3349

Chibani K, Ali-Rachedi S, Job C, Job D, Jullien M, Grappin P (2006) Proteomic analysis of seed dormancy in Arabidopsis. Plant Physiol 142: 1493-1510

Choy MK, Sullivan JA, Theobald JC, Davies WJ, Gray JC (2008) An Arabidopsis mutant able to green after extended dark periods shows decreased transcripts of seed protein genes and altered sensitivity to abscisic acid. J Exp Bot 59: 3869-3884


119

Clerkx EJ, Vries HB, Ruys GJ, Groot SP, Koornneef M (2003) Characterization of green seed, an enhancer of abi3-1 in Arabidopsis that affects seed longevity. Plant Physiol 132: 1077-1084

Clerkx EJM, Blankestijn-De Vries H, Ruys GJ, Groot SPC, Koornneef M (2004a) Genetic differences in seed longevity of various Arabidopsis mutants. Physiol Plant 121: 448-461

Clerkx EJM, El-Lithy ME, Vierling E, Ruys GJ, Blankestijin-De Vries H, Groot SPC, Vreugdenhil D, Koornneef M (2004b) Analysis of natural allelic variation of Arabidopsis seed germination and seed longevity traits between the accessions Landsberg erecta and Shakdara, using a new recombinant inbred line population. Plant Physiol 135: 432-443

Cone JW, Spruit CJP (1983) Imbibition conditions and seed dormancy of Arabidopsis thaliana. Physiol Plant 59: 416-420

Contreras S, Bennett MA, Metzger JD, Tay D (2008) Maternal light environment during seed development affects lettuce seed weight, germinability, and storability. HortScience 43: 845-852

Davies MJ (2005) The oxidative environment and protein damage. Biochim Biophys Acta 1703: 93-109

Debeaujon I, Leon-Kloosterziel KM, Koornneef M (2000) Influence of the testa on seed dormancy, germination, and longevity in Arabidopsis. Plant Physiol 122: 403-413

Dekkers BJ, Pearce S, van Bolderen-Veldkamp RP, Marshall A, Widera P, Gilbert J, Drost HG, Bassel GW, Muller K, King JR, Wood AT, Grosse I, Quint M, Krasnogor N, Leubner-Metzger G, Holdsworth MJ, Bentsink L (2013) Transcriptional dynamics of two seed compartments with opposing roles in Arabidopsis seed germination. Plant Physiol 163: 205-215

Derkx MPM, Karssen CM (1993a) Effects of light and temperature on seed dormancy and gibberellin-stimulated germination in Arabidopsis thaliana: studies with gibberellin-deficient and -insensitive mutants. Physiol Plant 89: 360-368

Derkx MPM, Karssen CM (1993b) Variability in light gibberellin and nitrate requirement of Arabidopsis thaliana seeds due to harvest time and conditions of dry storage. J Plant Physiol 141: 574-582

Doerge RW (2002) Mapping and analysis of quantitative trait loci in experimental populations. Nat Rev Genet 3: 43-52

Donohue K (2002) Germination timing influences natural selection on life-history characters in Arabidopsis thaliana. Ecology 83: 1006-1016


120

Donohue K, Dorn L, Griffith C, Kim E, Aguilera A, Polisetty CR, Schmitt J (2005) Environmental and genetic influences on the germination of Arabidopsis thaliana in the field. Evolution 59: 740-757

Drew RLK, Brocklehurst PA (1990) Effects of temperature of mother plant environment on yield and germination of seeds of lettuce (Lactuca sativa). Ann Bot 66: 63-71

El-Assal SED, Alonso-Blanco C, Peeters AJM, Raz V, Koornneef M (2001) A QTL for flowering time in Arabidopsis reveals a novel allele of CRY2. Nat Genet 29: 435-440

Eltayeb AE, Kawano N, Badawi GH, Kaminaka H, Sanekata T, Shibahara T, Inanaga S, Tanaka K (2007) Overexpression of monodehydroascorbate reductase in transgenic tobacco confers enhanced tolerance to ozone, salt and polyethylene glycol stresses. Planta 225: 1255-1264

Evans AS, Cabin RJ (1995) Can dormancy affect the evolution of post-germination traits? The case of Lesquerella fendleri. Ecology 76: 344-356

Fait A, Angelovici R, Less H, Ohad I, Urbanczyk-Wochniak E, Fernie AR, Galili G (2006) Arabidopsis seed development and germination is associated with temporally distinct metabolic switches. Plant Physiol 142: 839-854

Farnsworth E (2000) The ecology and physiology of viviparous and recalcitrant seeds. Annu Rev Ecol Syst 31: 107-138

Fenner M (1991) The effects of the parent environment on seed germinability. Seed Sci Res 1: 75-84

Finch-Savage WE, Cadman CSC, Toorop PE, Lynn JR, Hilhorst HWM (2007) Seed dormancy release in Arabidopsis Cvi by dry after-ripening, low temperature, nitrate and light shows common quantitative patterns of gene expression directed by environmentally specific sensing. Plant J 51: 60-78

Finch-Savage WE, Leubner-Metzger G (2006) Seed dormancy and the control of germination. New Phytol 171: 501-523

Finkelstein RR, Gampala SS, Rock CD (2002) Abscisic acid signaling in seeds and seedlings. Plant Cell 14 Suppl: S15-45

Focks N, Benning C (1998) Wrinkled1: a novel, low-seed-oil mutant of Arabidopsis with a deficiency in the seed-specific regulation of carbohydrate metabolism. Plant Physiol 118: 91-101

Freas KE, Kemp PR (1983) Some relationships between environmental reliability and seed dormancy in desert annual plants. J Ecol 71: 211-217


121

Gallardo K, Job C, Groot SP, Puype M, Demol H, Vandekerckhove J, Job D (2002) Importance of methionine biosynthesis for Arabidopsis seed germination and seedling growth. Physiol Plant 116: 238-247

Gallardo K, Job C, Groot SPC, Puype M, Demol H, Vandekerckhove J, Job D (2001) Proteomic analysis of Arabidopsis seed germination and priming. Plant Physiol 126: 835-848

Gao F, Rampitsch C, Chitnis VR, Humphreys GD, Jordan MC, Ayele BT (2013) Integrated analysis of seed proteome and mRNA oxidation reveals distinct post-transcriptional features regulating dormancy in wheat (Triticum aestivum L.). Plant Biotechnol J 11: 921-932

Goldberg RB, De Paiva G, Yadegari R (1994) Plant embryogenesis: zygote to seed. Science 266: 605-614

Graeber K, Linkies A, Muller K, Wunchova A, Rott A, Leubner-Metzger G (2010) Cross-species approaches to seed dormancy and germination: conservation and biodiversity of ABA-regulated mechanisms and the Brassicaceae DOG1 genes. Plant Mol Biol 73: 67-87

Graeber K, Nakabayashi K, Miatton E, Leubner-Metzger G, Soppe WJJ (2012) Molecular mechanisms of seed dormancy. Plant Cell Environ 35: 1769-1786

Grene R (2002) Oxidative stress and acclimation mechanisms in plants. Arabidopsis Book 1: e0036

Groot SP, Surki AA, de Vos RC, Kodde J (2012) Seed storage at elevated partial pressure of oxygen, a fast method for analysing seed ageing under dry conditions. Ann Bot 110: 1149-1159

Gu XY, Kianian SF, Foley ME (2006) Isolation of three dormancy QTLs as Mendelian factors in rice. Heredity 96: 93-99

Guo JJ, Wang JB, Xi L, Huang WD, Liang JS, Chen JG (2009) RACK1 is a negative regulator of ABA responses in Arabidopsis. J Exp Bot 60: 3819-3833

Gutierrez L, Van Wuytswinkel O, Castelain M, Bellini C (2007) Combined networks regulating seed maturation. Trends Plant Sci 12: 294-300

Gutterman Y (2000) Genotypic and phenotypic germination survival strategies of ecotypes and annual plant species in the Negev desert of Israel. Seed Biology: Advances and Applications: 389-399

Gutterman Y (2002) Survival strategies of annual desert plants. Springer Berlin Heidelberg

Hanin N, Quaye M, Westberg E, Barazani O (2013) Soil seed bank and among-years genetic diversity in arid populations of Eruca sativa Miller (Brassicaceae). J Arid Environ 91: 151-154


122

Hayes RG, Klein WH (1974) Spectral quality influence of light during development of Arabidopsis thaliana plants in regulating seed germination. Plant Cell Physiol 15: 643-653

He H, De Souza Vidigal D, Snoek BL, Schnabel S, Nijveen H, Hilhorst H, Bentsink L (submitted) Interaction between parental environment and genotype affects plant and seed performance in Arabidopsis.

Heath JD, Weldon R, Monnot C, Meinke DW (1986) Analysis of storage proteins in normal and aborted seeds from embryo lethal mutants of Arabidopsis thaliana. Planta 169: 304-312

Hoekstra FA, Golovina EA, Buitink J (2001) Mechanisms of plant desiccation tolerance. Trends Plant Sci 6: 431-438

Hoffmann MH (2002) Biogeography of Arabidopsis thaliana (L.) Heynh. (Brassicaceae). J Biogeogr 29: 125-134

Holdsworth MJ, Bentsink L, Soppe WJJ (2008) Molecular networks regulating Arabidopsis seed maturation, after-ripening, dormancy and germination. New Phytol 179: 33-54

Hoque TS, Uraji M, Tuya A, Nakamura Y, Murata Y (2012) Methylglyoxal inhibits seed germination and root elongation and up-regulates transcription of stress-responsive genes in ABA-dependent pathway in Arabidopsis. Plant Biol 10: 1438-8677

Hossain MA, Hossain MZ, Fujita M (2009) Stress-induced changes of methylglyoxal level and glyoxalase I activity in pumpkin seedlings and cDNA cloning of glyoxalase I gene. Aust J Crop Sci 3: 53-64

Hou A, Liu K, Catawatcharakul N, Tang X, Nguyen V, Keller WA, Tsang EW, Cui Y (2005) Two naturally occurring deletion mutants of 12S seed storage proteins in Arabidopsis thaliana. Planta 222: 512-520

Huang X, Schmitt J, Dorn L, Griffith C, Effgen S, Takao S, Koornneef M, Donohue K (2010) The earliest stages of adaptation in an experimental plant population: strong selection on QTLS for seed dormancy. Mol Ecol 19: 1335-1351

Hundertmark M, Buitink J, Leprince O, Hincha DK (2011) The reduction of seed-specific dehydrins reduces seed longevity in Arabidopsis thaliana. Seed Sci Res 21: 165-173

Hurtado PX, Schnabel SK, Zaban A, Vetelainen M, Virtanen E, Eilers PHC, van Eeuwijk FA, Visser RGF, Maliepaard C (2012) Dynamics of senescence-related QTLs in potato. Euphytica 183: 289-302

ISTA (2012) International rules for seed testing, Ed 2012. The International Seed testing Association (ISTA), Baaserdorf, Switzerland


123

James GV, Patel V, Nordstrom KJ, Klasen JR, Salome PA, Weigel D, Schneeberger K (2013) User guide for mapping-by-sequencing in Arabidopsis. Genome Biol 14: R61

Job C, Rajjou L, Lovigny Y, Belghazi M, Job D (2005) Patterns of protein oxidation in Arabidopsis seeds and during germination. Plant Physiol 138: 790-802

Jofuku KD, Denboer BGW, Vanmontagu M, Okamuro JK (1994) Control of Arabidopsis flower and seed development by the homeotic gene Apetala2. Plant Cell 6: 1211-1225

Joosen RVL, Arends D, Willems LAJ, Ligterink W, Jansen RC, Hilhorst HWM (2012) Visualizing the genetic landscape of Arabidopsis seed performance. Plant Physiol 158: 570-589

Joosen RVL, Kodde J, Willems LAJ, Ligterink W, van der Plas LHW, Hilhorst HWM (2010) GERMINATOR: a software package for high-throughput scoring and curve fitting of Arabidopsis seed germination. Plant J 62: 148-159

Joshi V, Joung JG, Fei Z, Jander G (2010) Interdependence of threonine, methionine and isoleucine metabolism in plants: accumulation and transcriptional regulation under abiotic stress. Amino Acids 39: 933-947

Kagaya Y, Toyoshima R, Okuda R, Usui H, Yamamoto A, Hattori T (2005) LEAFY COTYLEDON1 controls seed storage protein genes through its regulation of FUSCA3 and ABSCISIC ACID INSENSITIVE3. Plant Cell Physiol 46: 399-406

Karssen CM, Brinkhorstvanderswan DLC, Breekland AE, Koornneef M (1983) Induction of dormancy during seed development by endogenous abscisic acid - Studies on abscisic acid deficient genotypes of Arabidopsis Thaliana (L) Heynh. Planta 157: 158-165

Kendall S, Penfield S (2012) Maternal and zygotic temperature signalling in the control of seed dormancy and germination. Seed Sci Res 22: S23-S29

Kendall SL, Hellwege A, Marriot P, Whalley C, Graham IA, Penfield S (2011) Induction of dormancy in Arabidopsis summer annuals requires parallel regulation of DOG1 and hormone metabolism by low temperature and CBF transcription factors. Plant Cell 23: 2568-2580

Kigel J, Galili G (1995) Seed development and germination. M. Dekker, New York

Kirch HH, Bartels D, Wei Y, Schnable PS, Wood AJ (2004) The ALDH gene superfamily of Arabidopsis. Trends Plant Sci 9: 371-377

Koornneef M (1981) The complex syndrome of ttg mutants. Arabidopsis Info Serv 18: 45-51

Koornneef M (1990) Mutations affecting the testa colour in Arabidopsis. Arabidopsis Info Serv 27: 1-4


124

Koornneef M, Elgersma A, Hanhart CJ, Vanloenenmartinet EP, Vanrijn L, Zeevaart JAD (1985) A gibberellin insensitive mutant of Arabidopsis thaliana. Physiol Plant 65: 33-39

Koornneef M, Jorna ML, Derswan DLCB, Karssen CM (1982) The isolation of abscisic acid (ABA) deficient mutants by selection of induced revertants in non-germinating gibberellin sensitive lines of Arabidopsis thaliana (L) Heynh. Theor Appl Genet 61: 385-393

Koornneef M, Reuling G, Karssen CM (1984) The isolation and characterization of abscisic-acid insensitive mutants of Arabidopsis thaliana. Physiol Plant 61: 377-383

Koornneef M, Van der Veen JH (1980) Induction and analysis of gibberellin sensitive mutants in Arabidopsis thaliana (L) Heynh. Theor Appl Genet 58: 257-263

Krebbers E, Herdies L, De Clercq A, Seurinck J, Leemans J, Van Damme J, Segura M, Gheysen G, Van Montagu M, Vandekerckhove J (1988) Determination of the processing sites of an Arabidopsis 2S albumin and characterization of the complete gene family. Plant Physiol 87: 859-866

Kroj T, Savino G, Valon C, Giraudat J, Parcy F (2003) Regulation of storage protein gene expression in Arabidopsis. Development 130: 6065-6073

Kronholm I, Pico FX, Alonso-Blanco C, Goudet J, de Meaux J (2012) Genetic basis of adaptation in Arabidopsis thaliana: local adaptation at the seed dormancy QTL DOG1. Evolution 66: 2287-2302

Kucera B, Cohn MA, Leubner-Metzger G (2005) Plant hormone interactions during seed dormancy release and germination. Seed Sci Res 15: 281-307

Kushiro T, Okamoto M, Nakabayashi K, Yamagishi K, Kitamura S, Asami T, Hirai N, Koshiba T, Kamiya Y, Nambara E (2004) The Arabidopsis cytochrome P450 CYP707A encodes ABA 8’-hydroxylases: key enzymes in ABA catabolism. EMBO J 23: 1647-1656

Labuza TP (1971) Kinetics of lipid oxidation in foods. CRC Crit Rev Food Sci Nutr 2: 355-405

Landjeva S, Lohwasser U, Börner A (2009) Genetic mapping within the wheat D genome reveals QTL for germination, seed vigour and longevity, and early seedling growth. Euphytica 171: 129-143

Lee BH, Lee HJ, Xiong LM, Zhu JK (2002) A mitochondrial complex I defect impairs cold-regulated nuclear gene expression. Plant Cell 14: 1235-1251


125

Lee YP, Baek KH, Lee HS, Kwak SS, Bang JW, Kwon SY (2010) Tobacco seeds simultaneously over-expressing Cu/Zn-superoxide dismutase and ascorbate peroxidase display enhanced seed longevity and germination rates under stress conditions. J Exp Bot 61: 2499-2506

Léon-Kloosterziel KM, Gil MA, Ruijs GJ, Jacobsen SE, Olszewski NE, Schwartz SH, Zeevaart JAD, Koornneef M (1996a) Isolation and characterization of abscisic acid-deficient Arabidopsis mutants at two new loci. Plant J 10: 655-661

Léon-Kloosterziel KM, Keijzer CJ, Koornneef M (1994) A seed shape mutant of Arabidopsis that is affected in integument development. Plant Cell 6: 385-392

Léon-Kloosterziel KM, vandeBunt GA, Zeevaart JAD, Koornneef M (1996b) Arabidopsis mutants with a reduced seed dormancy. Plant Physiol 110: 233-240

Leprince O, Hoekstra FA (1998) The responses of cytochrome redox state and energy metabolism to dehydration support a role for cytoplasmic viscosity in desiccation tolerance. Plant Physiol 118: 1253-1264

Leubner-Metzger G, Petruzzelli L, Waldvogel R, Vogeli-Lange R, Meins F (1998) Ethylene-responsive element binding protein (EREBP) expression and the transcriptional regulation of class I beta-1,3-glucanase during tobacco seed germination. Plant Mol Biol 38: 785-795

Lin SY, Sasaki T, Yano M (1998) Mapping quantitative trait loci controlling seed dormancy and heading date in rice, Oryza sativa L., using backcross inbred lines. Theor Appl Genet 96: 997-1003

Liu YX, Geyer R, van Zanten M, Carles A, Li Y, Horold A, van Nocker S, Soppe WJJ (2011) Identification of the Arabidopsis REDUCED DORMANCY 2 gene uncovers a role for the Polymerase Associated Factor 1 Complex in seed dormancy. PLoS One 6: e22241

Liu YX, Koornneef M, Soppe WJJ (2007) The absence of histone H2B monoubiquitination in the Arabidopsis hub1 (rdo4) mutant reveals a role for chromatin remodeling in seed dormancy. Plant Cell 19: 433-444

Lloret F, Casanovas C, Penuelas J (1999) Seedling survival of Mediterranean shrubland species in relation to root : shoot ratio, seed size and water and nitrogen use. Funct Ecol 13: 210-216

Lubzens E, Cerdà J, Clark MS (2010) Introduction. E Lubzens, J Cerdà, MS Clark eds. Topics in Current Genetics: Dormancy and Resistance in Harsh Environments Vol 21. Springer, Heidelberg, Dordrecht, London, New York: pp 1-4


126

Maia J, Dekkers BJ, Provart NJ, Ligterink W, Hilhorst HW (2011) The re-establishment of desiccation tolerance in germinated Arabidopsis thaliana seeds and its associated transcriptome. PLoS One 6: e29123

Masubelele NH, Dewitte W, Menges M, Maughan S, Collins C, Huntley R, Nieuwland J, Scofield S, Murray JAH (2005) D-type cyclins activate division in the root apex to promote seed germination in Arabidopsis. Proc Natl Acad Sci USA 102: 15694-15699

Mayer U, Ruiz RAT, Berleth T, Misera S, Jurgens G (1991) Mutations affecting body organization in the Arabidopsis embryo. Nature 353: 402-407

McCullough JM, Shropshire WJ (1970) Physiological predetermination of germination responses in Arabidopsis thaliana (L.) Heynh. Plant Cell Physiol 11: 139-148

Meinke DW, Franzmann LH, Nickle TC, Yeung EC (1994) Leafy cotyledon mutants of Arabidopsis. Plant Cell 6: 1049-1064

Miura K, Lin SY, Yano M, Nagamine T (2002) Mapping quantitative trait loci controlling seed longevity in rice (Oryza sativa L.). Theor Appl Genet 104: 981-986

Moller IM, Jensen PE, Hansson A (2007) Oxidative modifications to cellular components in plants. Annu Rev Plant Biol 58: 459-481

Moore RP (1985) Handbook on tetrazolium testing. International Seed Testing Association, Zurich, Switzerland: 99

Morales-Ruiz T, Birincioglu M, Jaruga P, Rodriguez H, Roldan-Arjona T, Dizdaroglu M (2003) Arabidopsis thaliana OGG1 protein excises 8-hydroxyguanine and 2,6-diamino-4-hydroxy-5-formamidopyrimidine from oxidatively damaged DNA containing multiple lesions. Biochemistry 42: 3089-3095

Mudgett MB, Lowenson JD, Clarke S (1997) Protein repair L-isoaspartyl methyltransferase in plants. Phylogenetic distribution and the accumulation of substrate proteins in aged barley seeds. Plant Physiol 115: 1481-1489

Muntz K, Belozersky MA, Dunaevsky YE, Schlereth A, Tiedemann J (2001) Stored proteinases and the initiation of storage protein mobilization in seeds during germination and seedling growth. J Exp Bot 52: 1741-1752

Nakabayashi K, Bartsch M, Xiang Y, Miatton E, Pellengahr S, Yano R, Seo M, Soppe WJJ (2012) The time required for dormancy release in Arabidopsis is determined by DELAY OF GERMINATION1 protein levels in freshly harvested seeds. Plant Cell 24: 2826-2838

Nakabayashi K, Okamoto M, Koshiba T, Kamiya Y, Nambara E (2005) Genome-wide profiling of stored mRNA in Arabidopsis thaliana seed germination: epigenetic and genetic regulation of transcription in seed. Plant J 41: 697-709


127

Nambara E, Marion-Poll A (2003) ABA action and interactions in seeds. Trends Plant Sci 8: 213-217

Nelson DC, Scaffidi A, Dun EA, Waters MT, Flematti GR, Dixon KW, Beveridge CA, Ghisalberti EL, Smith SM (2011) F-box protein MAX2 has dual roles in karrikin and strigolactone signaling in Arabidopsis thaliana. Proc Natl Acad Sci USA 108: 8897-8902

Nguyen TP, Keizer P, van Eeuwijk F, Smeekens S, Bentsink L (2012) Natural variation for seed longevity and seed dormancy are negatively correlated in Arabidopsis. Plant Physiol 160: 2083-2092

Nocter G, Foyer CH (1998) Ascorbate and glutathione: keeping active oxygen under control. Ann Rev Plant Physiol Plant Mol Biol 49: 249-279

Oge L, Bourdais G, Bove J, Collet B, Godin B, Granier F, Boutin JP, Job D, Jullien M, Grappin P (2008) Protein repair L-isoaspartyl methyltransferase 1 is involved in both seed longevity and germination vigor in Arabidopsis. Plant Cell 20: 3022-3037

Ooms JJJ, Léon-Kloosterziel KM, Bartels D, Koornneef M, Karssen CM (1993) Acquisition of desiccation tolerance and longevity in seeds of Arabidopsis thaliana: a comparative study using abscisic acid-insensitive abi3 mutants. Plant Physiol 102: 1185-1191

Oracz K, El-Maarouf Bouteau H, Farrant JM, Cooper K, Belghazi M, Job C, Job D, Corbineau F, Bailly C (2007) ROS production and protein oxidation as a novel mechanism for seed dormancy alleviation. Plant J 50: 452-465

Osa M, Kato K, Mori M, Shindo C, Torada A, Miura H (2003) Mapping QTLs for seed dormancy and the Vp1 homologue on chromosome 3A in wheat. Theor Appl Genet 106: 1491-1496

Osborne DJ, Dellaquila A, Elder RH (1984) DNA-Repair in plant cells. An essential event of early embryo germination in seeds. Folia Biol (Praha) 30: 155-169

Osborne DJ, Sharon R, Benishai R (1981) Studies on DNA Integrity and DNA-Repair in germinating embryos of rye (Secale cereale). Isr J Plant Sci 29: 259-272

Page DR, Grossniklaus L (2002) The art and design of genetic screens: Arabidopsis thaliana. Nat Rev Genet 3: 124-136

Pang PP, Pruitt RE, Meyerowitz EM (1988) Molecular-cloning, genomic organization, expression and evolution of 12S-seed storage protein genes of Arabidopsis thaliana. Plant Mol Biol 11: 805-820


128

Peeters AJM, Blankestijn-de Vries H, Hanhart CJ, Leon-Kloosterziel KM, Zeevaart JAD, Koornneef M (2002) Characterization of mutants with reduced seed dormancy at two novel rdo loci and a further characterization of rdo1 and rdo2 in Arabidopsis. Physiol Plant 115: 604-612

Petru M, Tielborger K (2008) Germination behaviour of annual plants under changing climatic conditions: separating local and regional environmental effects. Oecologia 155: 717-728

Prieto-Dapena P, Castano R, Almoguera C, Jordano J (2006) Improved resistance to controlled deterioration in transgenic seeds. Plant Physiol 142: 1102-1112

Quick WA, Hsiao AI, Hanes JA (1997) Dormancy implications of phosphorus levels in developing caryopses of wild oats (Avena fatua L.). J Plant Growth Regul 16: 27-34

Rajjou L, Belghazi M, Catusse J, Oge L, Arc E, Godin B, Chibani K, Ali-Rachidi S, Collet B, Grappin P, Jullien M, Gallardo K, Job C, Job D (2011) Proteomics and posttranslational proteomics of seed dormancy and germination. Methods Mol Biol 773: 215-236

Rajjou L, Belghazi M, Huguet R, Robin C, Moreau A, Job C, Job D (2006) Proteomic investigation of the effect of salicylic acid on Arabidopsis seed germination and establishment of early defense mechanisms. Plant Physiol 141: 910-923

Rajjou L, Debeaujon I (2008) Seed longevity: survival and maintenance of high germination ability of dry seeds. C R Biol 331: 796-805

Rajjou L, Duval M, Gallardo K, Catusse J, Bally J, Job C, Job D (2012) Seed germination and vigor. Annu Rev Plant Biol 63: 507-533

Rajjou L, Gallardo K, Debeaujon I, Vandekerckhove J, Job C, Job D (2004) The effect of alpha-amanitin on the Arabidopsis seed proteome highlights the distinct roles of stored and neosynthesized mRNAs during germination. Plant Physiol 134: 1598-1613

Rajjou L, Lovigny Y, Groot SPC, Belghaz M, Job C, Job D (2008) Proteome-wide characterization of seed aging in Arabidopsis: a comparison between artificial and natural aging protocols. Plant Physiol 148: 620-641

Rajjou L, Miche L, Huguet R, Job C, Job D (2007) The use of proteome and transcriptome profiling in the understanding of seed germination and identification of intrinsic markers determining seed quality, germination efficiency and early seedling vigour. Seeds: Biology, Development and Ecology: 149-158

Ravanel S, Gakiere B, Job D, Douce R (1998) The specific features of methionine biosynthesis and metabolism in plants. Proc Natl Acad Sci USA 95: 7805-7812


129

Raz V, Bergervoet JHW, Koornneef M (2001) Sequential steps for developmental arrest in Arabidopsis seeds. Development 128: 243-252

Revilla P, Butron A, Rodriguez VM, Malvar RA, Ordas A (2009) Identification of genes related to germination in aged maize seed by screening natural variability. J Exp Bot 60: 4151-4157

Rouhier N, Jacquot JP (2005) The plant multigenic family of thiol peroxidases. Free Radic Biol Med 38: 1413-1421

Sasaki K, Fukuta Y, Sato T (2005) Mapping of quantitative trait loci controlling seed longevity of rice (Oryza sativa L.) after various periods of seed storage. Plant Breed 124: 361-366

Sattler SE, Gilliland LU, Magallanes-Lundback M, Pollard M, DellaPenna D (2004) Vitamin E is essential for seed longevity and for preventing lipid peroxidation during germination. Plant Cell 16: 1419-1432

Sawhney R, Quick WA, Hsiao AI (1985) The effect of temperature during parental vegetative growth on seed germination of wild oats (Avena Fatua L.). Ann Bot 55: 25-28

Schatzki J, Allam M, Kloppel C, Nagel M, Borner A, Mollers C (2013) Genetic variation for secondary seed dormancy and seed longevity in a set of black-seeded European winter oilseed rape cultivars. Plant Breed 132: 174-179

Schneeberger K, Ossowski S, Lanz C, Juul T, Petersen AH, Nielsen KL, Jorgensen JE, Weigel D, Andersen SU (2009) SHOREmap: simultaneous mapping and mutation identification by deep sequencing. Nat Methods 6: 550-551

Schultz TF, Quatrano RS (1997) Characterization and expression of a rice RAD23 gene. Plant Mol Biol 34: 557-562

Schwember AR, Bradford KJ (2010) Quantitative trait loci associated with longevity of lettuce seeds under conventional and controlled deterioration storage conditions. J Exp Bot 61: 4423-4436

Shen-Miller J (2002) Sacred lotus, the long-living fruits of China Antique. Seed Sci Res 12: 131-143

Simons AM (2011) Modes of response to environmental change and the elusive empirical evidence for bet hedging. Proc R Soc B 278: 1601-1609

Soeda Y, Konings M, Vorst O, van Houwelingen A, Stoopen GM, Maliepaard CA, Kodde J, Bino RJ, Groot SPC, van der Geest AHM (2005) Gene expression programs during Brassica oleracea seed maturation, osmopriming, and germination are indicators of progression of the germination process and the stress tolerance level. Plant Physiol 137: 354-368


130

Soppe WJ, Bentsink L, Koornneef M (1999) The early-flowering mutant efs is involved in the autonomous promotion pathway of Arabidopsis thaliana. Development 126: 4763-4770

Sturm A, Lienhard S (1998) Two isoforms of plant RAD23 complement a UV-sensitive rad23 mutant in yeast. Plant J 13: 815-821

Sugimoto K, Takeuchi Y, Ebana K, Miyao A, Hirochika H, Hara N, Ishiyama K, Kobayashi M, Ban Y, Hattori T, Yano M (2010) Molecular cloning of Sdr4, a regulator involved in seed dormancy and domestication of rice. Proc Natl Acad Sci USA 107: 5792-5797

Sugliani M, Brambilla V, Clerkx EJM, Koornneef M, Soppe WJJ (2010) The conserved splicing factor SUA controls alternative splicing of the developmental regulator ABI3 in Arabidopsis. Plant Cell 22: 1936-1946

Sugliani M, Rajjou L, Clerkx EJM, Koornneef M, Soppe WJJ (2009) Natural modifiers of seed longevity in the Arabidopsis mutants abscisic acid insensitive3-5 (abi3-5) and leafy cotyledon1-3 (lec1-3). New Phytol 184: 898-908

Tesnier K, Strookman-Donkers HM, Van Pijlen JG, Van der Geest AHM, Bino RJ, Groot SPC (2002) A controlled deterioration test for Arabidopsis thaliana reveals genetic variation in seed quality. Seed Sci Technol 30: 149-165

Tiedemann J, Rutten T, Monke G, Vorwieger A, Rolletschek H, Meissner D, Milkowski C, Petereck S, Mock HP, Zank T, Baumlein H (2008) Dissection of a complex seed phenotype: novel insights of FUSCA3 regulated developmental processes. Dev Biol 317: 1-12

To A, Valon C, Savino G, Guilleminot J, Devic M, Giraudat J, Parcy F (2006) A network of local and redundant gene regulation governs Arabidopsis seed maturation. Plant Cell 18: 1642-1651

Tollenaar M (1999) Duration of the grain-filling period in maize is not affected by photoperiod and incident PPFD during the vegetative phase. Field Crop Res 62: 15-21

Tommasi F, Paciolla C, Arrigoni O (1999) The ascorbate system in recalcitrant and orthodox seeds. Physiol Plant 105: 193-198

Van der Klei H, Vandamme J, Casteels P, Krebbers E (1993) A 5th 2S albumin isoform is present in Arabidopsis Thaliana. Plant Physiol 101: 1415-1416

Van Ooijen JW (2009) MapQTL6, software for the mapping of quantitative trait loci in experimental populations of diploid species. Kyazma B.V., Wageningen, Netherlands.


131

Verdier J, Lalanne D, Pelletier S, Torres-Jerez I, Righetti K, Bandyopadhyay K, Leprince O, Chatelain E, Vu BL, Gouzy J, Gamas P, Udvardi MK, Buitink J (2013) A regulatory network-based approach dissects late maturation processes related to the acquisition of desiccation tolerance and longevity of Medicago truncatula seeds. Plant Physiol 163: 757-774

Verma P, Kaur H, Petla BP, Rao V, Saxena SC, Majee M (2013) Protein L-isoaspartyl methyltransferase2 is differentially expressed in chickpea and enhances seed vigor and longevity by reducing abnormal isoaspartyl accumulation predominantly in seed nuclear proteins. Plant Physiol 161: 1141-1157

Walters C (1998) Understanding the mechanisms and kinetics of seed aging. Seed Sci Res 8: 223-244

Walters C, Hill LM, Wheeler LJ (2005) Dying while dry: kinetics and mechanisms of deterioration in desiccated organisms. Integr Comp Biol 45: 751-758

Wan LL, Ross ARS, Yang JY, Hegedus DD, Kermode AR (2007) Phosphorylation of the 12S globulin cruciferin in wild-type and abi1-1 mutant Arabidopsis thaliana (thale cress) seeds. Biochem J 404: 247-256

Waterworth WM, Masnavi G, Bhardwaj RM, Jiang Q, Bray CM, West CE (2010) A plant DNA ligase is an important determinant of seed longevity. Plant J 63: 848-860

Wechsberg GE, Probert RJ, Bray CM (1994) The relationship between ‘dehydrin-like’ proteins and seed longevity in Ranunculus sceleratus L. J Exp Bot 45: 1027-1030

Wehmeyer N, Vierling E (2000) The expression of small heat shock proteins in seeds responds to discrete developmental signals and suggests a general protective role in desiccation tolerance. Plant Physiol 122: 1099-1108

Weigel D (2012) Natural variation in Arabidopsis: from molecular genetics to ecological genomics. Plant Physiol 158: 2-22

Weigel D, Glazebrook J (2006) In planta transformation of Arabidopsis. CSH Protoc 2006

Weigel D, Mott R (2009) The 1001 genomes project for Arabidopsis thaliana. Genome Biol 10: 107

Western TL (2012) The sticky tale of seed coat mucilages: production, genetics, and role in seed germination and dispersal. Seed Sci Res 22: 1-25

Wilczek AM, Roe JL, Knapp MC, Cooper MD, Lopez-Gallego C, Martin LJ, Muir CD, Sim S, Walker A, Anderson J, Egan JF, Moyers BT, Petipas R, Giakountis A, Charbit E, Coupland G, Welch SM, Schmitt J (2009) Effects of genetic perturbation on seasonal life history plasticity. Science 323: 930-934

SamenvattingReferences

132

Withana-Gamage TS, Hegedus DD, Qiu X, Yu P, May T, Lydiate D, Wanasundara JP (2013) Characterization of Arabidopsis thaliana lines with altered seed storage protein profiles using synchrotron-powered FT-IR spectromicroscopy. J Agric Food Chem 61: 901-912

Wolucka BA, Persiau G, Van Doorsselaere J, Davey MW, Demol H, Vandekerckhove J, Van Montagu M, Zabeau M, Boerjan W (2001) Partial purification and identification of GDP-mannose 3’,5’-epimerase of Arabidopsis thaliana, a key enzyme of the plant vitamin C pathway. Proc Natl Acad Sci USA 98: 14843-14848

Wolucka BA, Van Montagu M (2003) GDP-mannose 3’,5’-epimerase forms GDP-L-gulose, a putative intermediate for the de novo biosynthesis of vitamin C in plants. J Biol Chem 278: 47483-47490

Wu XL, Liu HY, Wang W, Chen SN, Hu XL, Li CH (2011) Proteomic analysis of seed viability in maize. Acta Physiol Plant 33: 181-191

Xin X, Lin XH, Zhou YC, Chen XL, Liu X, Lu XX (2011) Proteome analysis of maize seeds: the effect of artificial ageing. Physiol Plant 143: 126-138

Zhang XR, Garreton V, Chua NH (2005) The AIP2 E3 ligase acts as a novel negative regulator of ABA signaling by promoting ABI3 degradation. Genes Dev 19: 1532-1543

Zhou R, Squires TM, Ambrose SJ, Abrams SR, Ross AR, Cutler AJ (2003) Rapid extraction of abscisic acid and its metabolites for liquid chromatography-tandem mass spectrometry. J Chromatogr A 1010: 75-85

Zhou YL, Chu P, Chen HH, Li Y, Liu J, Ding Y, Tsang EWT, Jiang LW, Wu KQ, Huang SZ (2012) Overexpression of Nelumbo nucifera metallothioneins 2a and 3 enhances seed germination vigor in Arabidopsis. Planta 235: 523-537

Zuber H, Davidian JC, Aubert G, Aime D, Belghazi M, Lugan R, Heintz D, Wirtz M, Hell R, Thompson R, Gallardo K (2010) The seed composition of Arabidopsis mutants for the group 3 sulfate transporters indicates a role in sulfate translocation within developing seeds. Plant Physiol 154: 913-926

SamenvattingReferences

Samenvatting

SamenvattingSamenvatting

134

Kiemrust en zaadbewaarbaarheid zijn de twee belangrijkste eigenschappen voor het overleven van zaden in de grond (zaadbanken). Beide eigenschappen worden geïnduceerd tijdens de zaadrijping en zorgen ervoor dat zaden de condities waarin ze niet kunnen uitgroeien tot levensvatbare planten kunnen overleven. Kiemrust bepaalt het moment waarop zaden kiemen terwijl de zaadbewaarbaarheid zorgt voor voldoende kiemingspotentiaal (kiempercentage). Het is bekend dat deze kiemingspotentiaal gedurende bewaring langzaam achteruit gaat (veroudering). De kiemingspotentiaal tijdens zaadbewaring in laboratoriumcondities kan worden weergegeven als een normale verdeling, met aan de ene kant kiemrust en aan de andere kant zaadbewaarbaarheid. Het werk beschreven in dit proefschrift is gericht op het identificeren van nieuwe genetische factoren die kiemrust en zaadbewaarbaarheid reguleren, door het induceren en uitbaten van genetische variatie.

QTL (Quantitative Trait Locus) analyses voor zaadbewaarbaarheid in zes genetische populaties (‘recombinant inbred lines’; RILs) gemeten als kiempercentage na bewaring in laboratorium condities heeft geleid tot de identificatie van negen QTLs (Germination Ability After Storage; GAAS; Hoofdstuk 2). De locatie van deze GAAS QTLs is vergeleken met die van de eerder geïdentificeerde kiemrust QTLs (Delay Of Germination; DOGs; Bentsink et al., 2010) wat heeft geleid tot de identificatie van co-lokaliserende loci met tegengestelde allel effecten. De effecten van de GAAS loci en de co-locatie met de DOGs zijn bevestigd door gebruik te maken van bijna isogene lijnen (NILs). Voor zover wij weten is dit de eerste rapportage van een negatieve correlatie tussen kiemrust en zaadbewaarbaarheid: diepe kiemrust correlerend met korte bewaarbaarheid en andersom. Gedetailleerde analyses aan de co-lokaliserende GAAS5 en DOG1 loci heeft laten zien dat het DOG1 allel van de accessie van de Kaapverdische Eilanden zowel de zaadbewaarbaarheid verlaagt als de kiemrust verhoogt.

In Hoofdstuk 3 hebben wij eiwit- en genetische analyses gecombineerd voor het vinden van genetische factoren die betrokken zijn bij zaadveroudering. Hiertoe hebben wij eiwitanalyses uitgevoerd met behulp van tweedimensionale gel electroforese op zaden van vier verschillende genotypen in twee fysiologische stadia (nagerijpt en verouderd zaad). Dit onderzoek is gedaan aan de drie sterkste GAAS loci (GAAS1, GAAS2 en GAAS5) gerepresenteerd door NILs met een Landsberg erecta (Ler) genetische achtergrond en aan Ler. Tijdens veroudering vinden er opmerkelijke veranderingen plaats in de eiwitprofielen van droog zaad. Deze veranderingen zijn genotype-specifiek, wat erop wijst dat zaadbewaring in de verschillende genotypen via verschillende mechanismen gereguleerd wordt. De veranderingen die wél overeen komen tussen de genotypen wijzen op een rol voor voedingsreserves (opslageiwitten) en anti-oxidant systemen zoals vitamine E, het belang van het beschermen en het instant houden van de eiwitvertaling, en energie en metabolisme gerelateerde processen. De rol van opslageiwitten (cruciferines en napines), het kleine ribosomale RPS12C en het NADP-afhankelijke malaatenzym 1 (NADP-ME1) is onomstotelijk vastgesteld doordat mutaties

SamenvattingSamenvatting

135

in deze genen leidden tot een significante verandering in de zaadbewaarbaarheid.

In Hoofdstuk 4 wordt de negatieve correlatie tussen kiemrust en zaadbewaarbaarheid op het DOG2/GAAS1 locus bovenaan chromosoom 1 nader onderzocht. Het effect van zowel DOG2 als GAAS1 zijn bevestigd met NILDOG2/GAAS1-Cvi. Zaden van deze NIL hebben een verminderde kiemrust en een betere bewaarbaarheid dan Ler. Genetische analyse wijst uit dat DOG2 en GAAS1 zeer waarschijnlijk twee onafhankelijke genen zijn. DOG2 beïnvloedt kiemrust op een maternale wijze en beide GAAS1 allelen reguleren bewaarbaarheid co-dominant. De locatie van het DOG2 locus is uiteindelijk teruggebracht tot een gebied van 85 Kb waarin 27 genen liggen. Het lokaliseren van GAAS1 vergt nog verdere analyse.

DOG6 is één van de sterkste kiemrust QTLs die zijn geïdentificeerd in de zes RIL populaties. Het effect van deze QTL is bevestigd door NILDOG6-Shakdara. We hebben deze NIL gebruikt voor het identificeren van non-dormante mutanten door het induceren van mutaties met behulp van ethylmethaansulfonaat in deze dormante achtergrond (Hoofdstuk 5). Op deze manier hebben wij zes kiemrust ‘modifiers’ (sdm1-6) geïdentificeerd. De mutanten sdm1, sdm3, sdm4, sdm5, en sdm6 hebben een verminderde zaadbewaarbaarheid als pleiotroop effect van het gemuteerde kiemrustgen. Sdm2 en sdm4 hebben een afwijkend fenotype voor de slijmlaag om de zaadhuid. Voor de genetische analyse van de mutanten en het identificeren van het gemuteerde gen door middel van een sequentieanalyse zijn genetische populaties gemaakt.

Het werk beschreven in dit proefschrift geeft een nieuwe kijk op de relatie tussen kiemrust, zaadbewaring en zaadveroudering en geeft een eerste zicht op de genen die hierbij mogelijk een rol spelen.

Acknowledgments

Acknowledgements Acknowledgements

138

This thesis is the fruit of many efforts of, besides myself, a number of people who I would like to acknowledge gratefully.

The first I would like to mention is Leónie, my supervisor, without whom this thesis would not have been possible. Leónie; this thesis is also yours. Do you know that it would take me quite some paper and ink to express my thanks to you? However, I chose to be short, precise and honest in order to avoid that readers get bored when reading this paragraph. Thank you for giving me the opportunity to work with you, for teaching me from beginning to the end and for your willingness to support me in the past as well as in the future. I cannot list everything that happened in these past years, but what I will keep with me is that I have enjoyed the time we were working together. Although sometimes discussions did not satisfy both sides, I enjoyed these also but in a different way. Saying without proof may not be very convincing, thus giving an example of how I enjoyed working with you would be nice. Writing the thesis is not an easy task. For some people, including myself, it is like a battle that one needs to fight days and nights to finish well in time. When the deadline was approaching, you kept fighting incredibly with me regardless of time and date, which made me feel that I was not alone. This greatly increased my motivation and as a result, we have nicely won the battle. There were also some highlighted moments at which we have made each other proud. These were not many, but they were enough to make me very happy to be your student and to have you as my supervisor. For now, I know that you are an example of the scientist I would like to become. Going beyond the relationship between a supervisor and a student is our friendship. When my passions for work and for life sank, you were there for me to abate all the frustrations, worries and stress, except when you were the source of these of course. That was always helpful to boost my inspirations and to keep me moving forward. Myself, my parents and my aunt are also very grateful to you and your family to welcome us at your house; that was such a beautiful experience. I truly deeply thank you.

I was so fortunate to be embed in the Seed Group led by Henk. In such a supportive scientific environment and the network given by the group, I have grown up fast. Through group work discussions in all these years, I have equipped myself with knowledge on broad aspects of seeds from which I will benefit a lot for my future in seed science. The progress I have made is also attributable to my two promotors; Sjef and Harro. I very much appreciate their critical reading and correcting of the manuscripts, as well as their supervision to guarantee my graduation. My external supervisor, Maarten, who very enthusiastically gave advices and suggestions for experiments; thank you very much. I am also thankful to Bas, who dedicates himself to science, for his discussions and proofreading of my manuscripts.

I am grateful to all the colleagues of the Seed Group and the Laboratory of Plant Physiology for their help in one way or another and to make my work and life more

Acknowledgements Acknowledgements

139

enjoyable. I will always remember your company when we had group and lab activities; they are such fantastic memories. I would take the chance here to address special acknowledgements to colleagues who turned into my dear friends. Hanzi, Deborah and Maria-Cecília, who not only helped me at work but also gave me so much joy and fun when we were together having our dinner circus and chatting. I wish we will keep our tradition as long as we can. Wilco, the biggest friend that I ever had; thank you for being such a helpful paranymph and for all funny moments when I was talking to you. My gratitude also goes to Lidiya; your sweetness and kindness for me will never let me forget you. Yanxia, a small but strong friend; from you I have learned to grow stronger. Farzaneh and Bing; thanks for sharing your sweets with me.

Although not physically situated in the Molecular Plant Physiology Group in Utrecht, I am thankful to all my colleagues there; thanks for always remembering me as a member of the group.

The experiments would not have gone quickly without help of my students, Quan, Jeroen and Jelle, and of the technician Marieke. From them I also learned how to be a supervisor, how to delegate tasks and how to work in a team.

Being away from my home country, I very much missed talking in Vietnamese and eating Vietnamese food. This miss was compensated by my Vietnamese friends, who are students living in Wageningen. I appreciate the time we shared together a lot. I also thank a special friend, Tung, who taught me science and life from the day I started my journey abroad. Without your company I would never have made it this far.

Waiting for me all the time, thousands miles away from the Netherlands, in Vietnam, are my parents and my sister. I am deeply indebted to them for their unconditional love and spiritual support. My success today is dedicated to you because it was sown by you. I love you so much.

I know words cannot express all appreciations that I have for all of you. I will always keep them in my heart and will always miss you. Many thanks and much love!

In the last words I would like to thank my research work for making me be recognized in the scientific world.

This research is supported by the Dutch Technology Foundation STW, which is part of the Netherlands Organisation for Scientific Research (NWO) and partly funded by the Ministry of Economic Affairs (project number STW 10328)

Curriculum vitae

140

Curriculum Vitae

Thu-Phuong Nguyen (Nguyên Thu Phương) was born on 27th of November 1983 in Hanoi, Vietnam. After graduating from high school, she had difficulties to make a choice between studying economics or agriculture. Finally, she decided to go for the Hanoi University of Agriculture with a bachelor major in Plant Breeding. Graduating in 2005, she started working at the Institute of Agricultural Genetics in Hanoi. After one year of working, she knew that she wanted to pursue a Master degree abroad. Being granted a scholarship from the Vietnamese government for a two-year study, she went to Germany and began her wished journey abroad. Phuong studied Agrobiotechnology at the Justus Liebig University in Giessen, Germany. During these two years, she discovered that the passion for plant science had been strongly growing in her, and also that she wanted to explore and experience more countries she had never visited. Therefore she started searching for a suitable PhD project outside Germany. Late 2009, Phuong became a PhD student in the STW project “Seed dormancy and storability: towards the identification of markers and genes” under the supervision of Dr. Leónie Bentsink. This project was based at the Molecular Plant Physiology Group of Utrecht University, but was conducted in the Wageningen Seed Lab, Laboratory of Plant Physiology, Wageningen University, The Netherlands.

Publication

Nguyen TP, Keizer P, van Eeuwijk F, Smeekens S, Bentsink L (2012) Natural variation for seed longevity and seed dormancy are negatively correlated in Arabidopsis. Plant Physiol 160: 2083-2092

Documents

Seed Dormancy and Seed Longevitywageningenseedlab.nl/thesis/tpnguyen/Thesis Thu-Phuong...expression of seed storage proteins (Gutierrez et al., 2007). Seeds of abi3 , lec1 , and fus3