Plant Physiology and Biochemistry 83 (2014) 346e355

Contents lists available at ScienceDirect

Plant Physiology and Biochemistry journal homepage: www.elsevier.com/locate/plaphy

Research article

Preferred carbon precursors for lipid labelling in the heterotrophic endosperm of developing oat (Avena sativa L.) grains Åsa Grimberg* €xtskyddsva €gen 1, P.O. Box 101, SE-230 53 Alnarp, Sweden Department of Plant Breeding, Swedish University of Agricultural Sciences, Va

a r t i c l e i n f o

a b s t r a c t

Article history: Received 10 April 2014 Accepted 21 August 2014 Available online 30 August 2014

Oat (Avena sativa L.) is unusual among the cereal grains in storing high amounts of oil in the endosperm; up to 90% of total grain oil. By using oat as a model species for oil metabolism in the cereal endosperm, we can learn how to develop strategies to redirect carbon from starch to achieve high-oil yielding cereal crops. Carbon precursors for lipid synthesis were compared in two genetically close oat cultivars with different endosperm oil content (about 6% and 10% of grain dw, medium-oil; MO, and high-oil; HO cultivar, respectively) by supplying a variety of 14C-labelled substrates to the grain from both up- and downstream parts of glycolysis, either through detached oat panicles in vitro or by direct injection in planta. When supplied by direct injection, 14C from acetate was identified to label the lipid fraction of the grain to the highest extent among substrates tested; 46% of net accumulated 14C, demonstrating its applicability as a marker for lipids in the endosperm. Time course analyses of injected 14C acetate during grain development suggested a more efficient transfer of fatty acids from polar lipids to triacylglycerol in the HO as compared to the MO cultivar, and turnover of triacylglycerol was suggested to not play a major role for the final oil content of oat grain endosperm despite the low amount of protective oleosins in this tissue. Moreover, availability of light was shown to drastically affect grain net carbon accumulation from 14 C-sucrose when supplied through detached panicles for the HO cultivar. © 2014 Elsevier Masson SAS. All rights reserved.

Keywords: Carbon partitioning Cereals Endosperm Fatty acid Oat grain Oil Starch

1. Introduction The endosperm of oat grains contains the highest concentration of oil among the cereals, therefore making it an excellent model species for studying carbon allocation into oil in the cereal endosperm (Banas et al., 2007; Liu, 2011; Price and Parsons, 1975; Barthole et al., 2012). The potential to develop new oil crops by redirecting carbon from starch to oil in already high-productive crops like cereals has been suggested as a possibility to meet the increased demand for a higher global plant oil production (Barthole et al., 2012; Carlsson et al., 2011). If the oil content of maize (Zea mays), today on average 4%, was increased to 25% by grain weight (assuming no change in total energy accumulation but taking into account the reduced weight loss due to the difference in energy density between starch and oil), the annual total global plant oil commodity available would almost double from today's 160 Mton (Carlsson et al., 2011; FAOSTAT, 2012; Vanhercke et al., 2013).

Abbreviations: TAG, triacylglycerol; FA, fatty acid; MO, medium-oil; HO, high-oil. * Tel.: þ46 40 415541. E-mail address: [email protected]. http://dx.doi.org/10.1016/j.plaphy.2014.08.018 0981-9428/© 2014 Elsevier Masson SAS. All rights reserved.

The cereal endosperm is a heterotrophic storage tissue that is dependent on long-distance transported fixed carbon from photosynthesizing leaves during the seed filling period. Sucrose unloading into wheat and barley grain endosperms occurs from the phloem along the single vascular bundle in the crease of the grain (Thorne, 1985). The biosynthesis of triacylglycerols (TAG; oil) from sucrose includes many more steps and requires a higher investment of energy (ATP) and reducing power (NAD(P)H) but the final product is a more reduced dense form of carbon storage, as compared to starch (Baud and Lepiniec, 2010; De Vries et al., 1974). To be available for fatty acid (FA) synthesis, carbon in the cytosol must first be transformed into pyruvate through either cytosolic or plastidial glycolysis and at some point imported into the plastid for de novo FA synthesis. FAs are then exported to the endoplasmatic reticulum where assembly with glycerol-3-phosphate subsequently yields TAG (Bates et al., 2013). Breeding efforts to develop a high oil phenotype of maize that resulted in 20% oil in kernels was explained by both increased oil content of and enlarged embryo (Moose et al., 2004; Dudley and Lambert, 2010), whereas in oat, recurrent selection resulted in 18% oil in grains which was caused by increased oil content of the endosperm at the expense of starch (Peterson and Wood, 1997). To

Å. Grimberg / Plant Physiology and Biochemistry 83 (2014) 346e355

achieve a drastic increase of oil yields from cereals, a radical reprogramming of the metabolism of the starch storing cereal endosperm, which makes out the largest part (up to 95%) of the grain, is needed. In attempts to increase the oil content by genetic engineering of plant storage tissues, transcription factors are attractive gene targets in contrast to single-gene approaches. WRINKLED1 (WRI1) is a transcription factor from Arabidopsis known to regulate the flux of carbon into oil in seeds by controlling the expression of genes involved in glycolysis and FA synthesis (Baud et al., 2007; Focks and Benning, 1998). When overexpressing a maize embryo homolog of the Arabidopsis WRI1 in the maize endosperm, oil concentration of this tissue was not increased in contrast to when overexpressing it in the embryo where it increased oil content from approximately 32% up to 37% (Shen et al., 2010). Interestingly, a homolog to the Arabidopsis WRI1 was highly expressed in the endosperm of oat but with no difference in expression levels between two oat cultivars with different endosperm oil content (supplementary material in (Hayden et al., 2011)), indicating that this transcription factor could not explain the difference in oil content. All together this emphasizes the importance of studying lipid metabolism of the developing oat endosperm to elucidate why carbon in the storage organ of this cereal grain is not only committed for starch and protein accumulation as in other cereals, but also for significant amounts of oil. To understand the flow of carbon into different storage products in seeds with different proportions of starch, oil and protein, usually two types of analyses have been adapted. Plastids from seeds have been isolated and their ability to incorporate a variety of exogenously supplied metabolites into storage products has been determined. A limitation to the studies with isolated plastids is that they do not quantitatively reflect carbon fluxes in vivo (Alonso et al., 2007). More recently, 13C metabolic flux analysis has been done by feeding developing embryos cultured in vitro with different substrates to determine the relative proportions of carbon flux through different pathways (Chen and Shachar-Hill, 2012). These in vitro embryo culture systems were optimized to carefully mimic embryo growth and storage composition in vivo (Schwender and Ohlrogge, 2002) and even though also developed for maize kernels with intact endosperm (Alonso et al., 2011), no such system has yet been developed for oat grains. In general, data from these types of studies made on embryos from sunflower (Helianthus annuus), soybean (Glycine max), maize (Zea mays), oil-seed rape (Brassica napus), and Arabidopsis show that the metabolic fluxes and carbon precursors for FA synthesis differ from one oilseed species to another, but usually with similarities among green or non-green seeds (reviewed in Baud and Lepiniec (2010)). An experimental system for developing oat grains on detached panicles in liquid cultures was previously developed and was shown to mimic grain composition during different developmental stages in planta (Ekman et al., 2008). That study also showed significant differences in total 14C accumulation from sucrose in the grain and allocation of 14C into lipids, between two genetically very close oat cultivars that had different endosperm oil concentration. Comparative transcriptome and metabolome analyses of the same two oat cultivars revealed large differences at the level of metabolites, rather than transcripts, of core metabolic pathways such as glycolysis indicating that posttranscriptional regulation is probably important for the final phenotype (Hayden et al., 2011). These results illustrate the complexity in finding key reasons to why one cereal endosperm tissue allocates a higher share of carbon into oil compared to another and emphasizes the need for using complementary analyses to gain a deeper understanding of the metabolism of the cereal endosperm. The aim of this study was to biochemically characterize the oat endosperm by supplying a variety of 14C-labelled precursors for FA

347

synthesis from both up- and downstream part of glycolysis using the previously developed detached panicle system (Ekman et al., 2008) and also a method that in some aspects even better mimics in vivo conditions; by directly injecting labelled substrates into the endosperm using a syringe in planta (Vigeolas and Geigenberger, 2004; Rolletschek et al., 2005). By using both these methods in parallel, carbon sources for lipid synthesis in oat endosperms were compared in two closely related oat cultivars with different endosperm oil concentration: 6% oil of grain dw (medium-oil, MO) and 10% oil (high-oil, HO). Differences between the two supply methods used are also discussed. 2. Material and methods 2.1. Plant material and growth conditions Two oat cultivars, cv. Matilda with 10% oil in mature grains, and €nnen, Svalo € v, Sweden), were grown in cv. Freja with 6% oil (Lantma controlled growth chambers (Biotronen, SLU-Alnarp, Sweden) under fluorescent light (200 mmol m2 s1 photosynthetically active radiation) under a 16/8 h light/dark photoperiod at 21/18  C temperature and 70% humidity. 2.2. In vitro culture of grains on detached oat panicles Oat grains were developed on detached panicles in vitro as described previously (Ekman et al., 2008). In brief, panicles on plants grown in controlled growth chambers were detached at anthesis, surface sterilized and put in a nutrient medium containing 15 g L1 sucrose, 0.8 g L1 glutamine, 0.5 g L1 [N-morpholino] ethane sulphonic acid, and 1.47 g L1 Murashige Skoogh medium (all chemicals from Duchefa, Harleem, The Netherlands). Panicles were further incubated in environmental growth cabinets (Sanyo Electric Co, Japan) with growth parameters comparable to plants grown in vivo described above (but without humidity control) until grains reached mid developmental stage. Nutrient medium was changed and fresh cuts of stems were done to optimize nutrient uptake every second day. 2.3. Radioactive isotope labelling of oat grains Detached oat panicles in vitro: When grains on detached panicles reached mid developmental stage (approximately 14 days after detachment), radioactive 14C substrates (see below) were added to the nutrient medium in the morning. Panicles (two individual panicles per substrate tested) were incubated for 48 h in growth cabinets where after grains were sampled (still green with a milky endosperm and fresh weights of approximately 35 mg grain1, ‘stage E’ according to Ekman et al. (2008)) into tubes on ice. Exposure to darkness was achieved by covering the panicles with aluminum foil. Fresh weights of individual grains were measured to confirm developmental stage and then pooled (six grains from each panicle), snap freezed in liquid nitrogen and then saved in 80  C freezer until extraction. Grains labelled with 14C-acetate were first split into endosperm and embryo þ scutellum. 14C-substrates were added to fresh nutrient medium to a concentration of 444,000 dpm (mL)1 corresponding to 0.3, 66.7, 4.1, 12.5, and 3.5 nmol (mL)1, respectively according to the substrate list below. Total radioactivity accumulating in grains when substrates were fed through the panicle reached values of approximately 20,000e50,000 dpm grain1 corresponding to substrate amounts of 0.24e6.0 nmol grain1 for glucose, malate, pyruvate, and acetate but as much as 3500e4800 nmol grain1 for sucrose (which was also present as un-labelled carbon source in the liquid growth media), after 48 h.

348

Å. Grimberg / Plant Physiology and Biochemistry 83 (2014) 346e355

Direct injection in planta: Grains on intact plants in controlled growth chambers in planta of mid developmental stage (stage E, see above) were identified and marked using a pen. 14C substrates in 1 mL were injected in the morning using a syringe (SGE, outer €teborg, Sweden) through diameter 0.47 mm, Scantec Nordic AB, Go the pericarp into the endosperm at half of the grain length, next to the crease. Due to the increased pressure of the grain at the moment of injection, a part of this volume (mixed with some milky endosperm) was squeezed out of the grain but was removed using tissue paper. After 24 h of incubation, marked grains were sampled into tubes on ice, fresh weights of individual grains measured to confirm developmental stage and then pooled (two to three grains from each panicle, from three individual panicles), and then snap freezed in liquid nitrogen and saved in 80  C freezer until extraction. 14C-substrates were diluted in sterile millipore water to 20,000 dpm (mL)1 and the amount (1 mL) injected into each grain corresponded to 0.01, 3.00, 0.18, 0.56, and 0.16 nmol grain1, respectively, according to list below. Substrates used: D-[U-14C]-sucrose; 200 mCi (mL)1, 630 mCi (mmol)1, D-[U-14C]-glucose; 200 mCi (mL)1, 3 mCi (mmol)1, L[1,4(2,3)-14C]malic acid; 200 mCi (mL)1, 49 mCi (mmol)1, 2e14C pyruvic acid; 500 mCi (mL)1, 16 mCi (mmol)1, 2-14C-acetic acid; 200 mCi (mL)1, 57 mCi (mmol)1. All substrates were from Amersham (Buckinghamshire, UK) or Perkin Elmer (Massachusetts, USA). The specific activity of 14C-sucrose used in the detached panicle system (diluted by the sucrose in the nutrient medium) was 4.6 mCi mmol1. 2.4. Extraction and radioactivity measurements Bligh and Dyer extractions (Bligh and Dyer, 1959) of grain samples resulted in one chloroform phase containing the lipids, and one water/methanol phase containing other compounds. An aliquot of the chloroform phase was transferred to a scintillation vial. The water methanol phase (with all chloroform remnants removed) was vortexed and an aliquot was immediately taken out to a scintillation vial before solid particles started to sediment. The rest of the water/methanol phase was centrifuged (1000 g, 3 min) to get a totally clear phase from which an aliquot was taken for scintillation counting. Chloroform samples were evaporated under nitrogen gas on hot sand and then resuspended in scintillation solvent for organic samples (Ultima Gold F, Perkin Elmer, Shelton, USA) and the water/methanol samples were used directly in scintillation solvent (Ultima Flo-M, Perkin Elmer, Shelton, USA). Radioactivity was measured using a liquid scintillation counter (PW 4700, Philips, Almelo, The Netherlands) to calculate the amounts of radioactivity in lipids (from the chloroform phase), soluble compounds (from the clear water/methanol phase) and starch/protein/ cellulose (from the subtraction of radioactivity in the clear from the vortexed water/methanol phase). Aliquots of total lipid chloroform extracts were separated using thin layer chromatography (TLC) on silica 60 plates (Merck, Darmstadt, Germany) in hexane/diethylether/acetic acid (35/15/ 0.01, v/v/v). Lipids (TAGs, diacylglycerols, polar lipids, and the rest) were identified using authentic standards and scraped from plates to scintillation vials, dried under nitrogen gas on hot sand, and then counted for radioactivity in scintillation solvent (Ultima Gold F).

pairwise comparisons were made for all combinations of treatments. All the stated differences are significant at P < 0.05 if not else is specified. 3. Results 3.1. Net accumulation of carbon in the oat grain Both feeding methods used in this study give data on net accumulation of 14C from five different substrates and its distribution between different storage compounds in the grain, but under two different conditions. While feeding substrates through detached panicles gives an increase in total grain 14C accumulation over time, the predefined amount of injected substrates using a syringe give a pulse of 14C in the grain. The two methods of supplying substrates into the grain endosperm therefore resulted in very different total amounts of net accumulated 14C (Fig. 1). It is important to note that the amounts of substrates fed to grains in all experiments should be regarded as tracer most likely not affecting the endogenous total metabolite pool sizes, except for when feeding 14C-sucrose to grains through detached panicles where also the nutrient medium itself contained high amounts of sucrose as carbon source. For grains developing on detached panicles, an incubation time of 48 h was used since 24 h was not long enough to get a reasonable amount of 14C in the liquid media to reach into the grain through the stem (data not shown). Total radioactivity accumulating in grains when substrates were fed through the panicle reached values of approximately 20,000e50,000 dpm grain1 after 48 h. Among substrates tested, only sucrose showed significant cultivar difference in total net accumulated 14C per grain when supplied through detached panicles with 40% higher levels in the HO compared to in the MO cultivar (Fig. 1(A), p < 0.075). When supplying 14C substrates by direct injection, a total radioactivity of 20,000 dpm was loaded into the grain endosperm. A part of this total radioactivity was squeezed out from the grain (and removed) due to the increased pressure in the grain induced at the moment of injection which probably explains the high variability (Fig. 1(B)). 14C from sucrose and glucose showed the highest net accumulation of 14C per grain (10,000e13,000 dpm grain1) as compared to from malate, pyruvate and acetate (4000e8000 dpm grain1) 24 h after injections (Fig. 1(B)).

2.5. Statistical analysis of data Data were analysed by analysis of variance (ANOVA) using the general linear model in which all treatment factors were fixed (MINITAB 15; Minitab, State College, PA, USA). Pairwise comparisons of respons data were made for treatment factors shown to significantly affect response data using the method of Tukey at the 5% level. If there was interaction between treatment factors,

Fig. 1. Total 14C net accumulation in oat grains from five different 14C-labelled substrates supplied either through detached oat panicles in liquid culture incubated for 48 h (A) or by direct injection into the grain endosperm using a syringe and 24 h of incubation (B). Sucr; sucrose, Gluc; glucose, Mal; malate, Pyr; pyruvate, Acet; acetate, MO; medium-oil cultivar, HO; high-oil cultivar. Results are the mean ± standard deviation for two (A) or three (B) samples.

Å. Grimberg / Plant Physiology and Biochemistry 83 (2014) 346e355

349

3.2. Carbon precursors for lipid synthesis in the oat grain The relative distributions of accumulated 14C in grains between lipids, starch/protein/cellulose, and water soluble compounds from substrates supplied through detached panicles to the oat grain were determined (for details see 2.4) at a mid-developmental stage. This stage was previously determined to be linear in oil accumulation (green grains with fresh weights 35 mg, (Ekman et al., 2008)). The non-lipid fraction of the grain was centrifuged at low speed to separate large macromolecules like starch granules, insoluble/denatured proteins and cell walls from the clear water/methanol phase that should include all other soluble metabolites and molecules (which can mean both unused and metabolized substrate, among other soluble compounds). The overall distribution pattern of net accumulated 14C in the grain looked fairly similar for all substrates with the majority (60e73%) found in the starch/protein/ cellulose fraction, 6e20% in lipids, and 8e29% in water soluble compounds (Fig. 2(A)). The highest proportions of 14C found in lipids out of net accumulated 14C in the grain were from sucrose and glucose, which are more upstream precursors for FA synthesis as compared to malate, pyruvate, and acetate. Significantly higher proportions of 14C in lipids (3e8% higher of total recovered radioactivity in grain) were observed for the HO compared to MO cultivar for all substrates tested, with largest relative difference for acetate. The overall 14C distribution pattern was more diverse for substrates supplied to the oat endosperm by direct injection using a syringe compared to through detached panicles, but with no significant cultivar differences (Fig. 2(B)). A very high proportion of 14C incorporation into lipids was observed for acetate, 46% of total recovered radioactivity in grain, as compared to other substrates (6e16%). In general, a very high share of 14C from substrates (except from acetate) was found in the water soluble fraction (up to 88%) and 14C found in the starch/protein/cellulose fraction was low (6e45%), compared to when feeding substrates through detached panicles (compare Fig. 2(A) and (B)). 3.3. Spatial distribution of acetate in grains Despite that 14C-acetate fed through detached panicles was incorporated to a surprisingly low extent into lipids of the grain as

Fig. 3. Distribution of total grain 14C accumulation from acetate between lipids, starch/ protein/cellulose, and water soluble compounds in embryo þ scutellum (e þ sc) and endosperm (es) in medium-oil (MO) and high-oil (HO) cultivars after 48 h of feeding substrate through detached oat panicles. Prot; protein, Cellu; cellulose. Results are the mean ± standard deviation for two samples.

mentioned above (Fig. 2(A)), the resolution of 14C distribution in different parts of the grain was still high enough to see clear cultivar differences (Fig. 3). The proportion of net accumulated 14C from acetate in the grain that was found in lipids was the same for the embryo and scutellum of both cultivars, whereas, for the endosperm there was a significant difference with 4% and 12% for the MO and HO cultivars, respectively (Fig. 3). This data indicates that when 14C-acetate is supplied to the grains through detached panicles, cultivars show a difference in allocation of 14C into lipids in the endosperm, but not in the embryo. 3.4. Influence of light on carbon accumulation Detached panicles of oat fed with 14C-sucrose were covered with aluminum foil to see how light influenced total carbon accumulation as well as distribution of incorporated carbon (Fig. 4). It was shown that net accumulation of 14C from sucrose per grain was drastically reduced (63%) when panicles of the HO

Fig. 2. Distribution of total 14C accumulation in grains between lipids, starch/protein/cellulose, and water soluble compounds from five different substrates labelled with 14C in medium-oil (MO) and high-oil (HO) cultivars of oat. Substrates were supplied to grains either through (A) detached oat panicles in liquid culture (48 h of incubation) or by (B) direct injection into grain endosperm using a syringe (24 h of incubation). Sucr; sucrose, Gluc; glucose, Mal; malate, Pyr; pyruvate, Acet; acetate, Prot; protein, Cellu; cellulose. Results are the mean ± standard deviation for two (A) or three (B) samples.

350

Å. Grimberg / Plant Physiology and Biochemistry 83 (2014) 346e355

Fig. 4. Effect of light cycle regime on total (A) and distribution of (B) 14C accumulation from sucrose in grains of medium-oil (MO) and high-oil (HO) oat cultivars after 48 h of feeding substrate through detached panicles. Light; 16 h light/8 h dark cycle, dark; 24 h darkness, Prot; protein, Cellu; cellulose. Results are the mean ± standard deviation for two samples.

cultivar did not have access to light, as compared to exposed to a normal day/night cycle (Fig. 4(A)). The MO cultivar also seemed to show a decrease (13%) in net accumulated 14C per grain when exposed to darkness, but this effect was not statistically significant. The distribution of net accumulated 14C from sucrose in the grain between different storage compounds was not drastically changed by light regimes even though the proportions of 14C recovered in lipids were 3% higher for both cultivars when grains developed in light as compared to in darkness (Fig. 4(B)). Proportions of net accumulated 14C in the grain that were recovered in lipids were found to be 6% higher in the HO as compared to the MO cultivar, in both light regimes. The data show that light is a factor that drastically influences carbon accumulation in the HO but only marginal in the MO cultivar, but also that light affects the allocation of carbon between different storage compounds in the grain.

3.5. Temporal distribution of

14

C from acetate in oat grains

Since acetate was the only substrate shown to label lipids to a high extent (46% of total 14C net accumulation in the grain) 24 h after injection into the seed (Fig. 2(B)), the metabolism of this substrate was subjected to a more detailed time-course analysis (Fig. 5). At the time of injection, nearly 100% of the injected 14C from acetate was found in the water soluble phase (Fig. 5(B)). Twenty minutes after injection, the 14C found in the water soluble phase was decreased to approximately 60% of total grain 14C and that found in lipids was 30%, and with the remaining 10% in the starch/ protein/cellulose fraction (Fig. 5). The proportion of 14C from acetate that accumulated in lipids then increased up to three days after injection (53% of total grain 14C), where after it stayed more or less constant up to seven days. During this time, the proportion of 14C from acetate in the water soluble phase decreased down to 11% of

Fig. 5. Time course of (A) net accumulation of 14C from acetate in oat grains and (B) distribution of 14C from acetate between lipids, starch/protein/cellulose and water soluble compounds after injection of substrate into oat grain endosperm of medium-oil (MO) and high-oil (HO) cultivars. Results are the mean ± standard deviation for three samples.

Å. Grimberg / Plant Physiology and Biochemistry 83 (2014) 346e355

total grain 14C whereas that in the starch/protein/cellulose fraction increased up to approximately 35%. No cultivar differences in the distribution of 14C from acetate between different storage compounds in the grain were seen. Large cultivar differences were instead observed in the distribution of 14C from injected acetate between different lipid classes over time (3.6). Total absolute amounts of 14C from acetate injected into each grain varied considerably between each injection due to some practical difficulties as mentioned above (3.1, Fig. 5(A)), therefore not allowing for determining cultivar differences of this parameter. However, it was roughly estimated from data that the major part of injected 14C seemed to stay in the grain (i.e. not lost as respired 14 CO2) of both cultivars, at least for the time period studied.

351

Of total 14C from acetate incorporated into lipids, a significantly higher proportion was accumulated in TAG in the HO (21%) compared to in the MO (7%) cultivar 2 h after injection (Fig. 6). This proportion increased much faster in the HO as compared to the MO cultivar, reaching the maximum levels in TAG after 24 h and three days, respectively, where after the proportion stayed more or less constant. At three and seven days after injection, no significant cultivar difference in the proportion of 14C accumulating in TAG out of total lipids could be seen (37%). In reverse to the differences observed for TAG, the proportion of 14C accumulating in PL out of total lipids started at a higher level in the MO cultivar (69%) as compared to the HO cultivar (54%). For the HO cultivar this proportion had decreased at 24 h after injection, but was delayed until three days after injection for the MO cultivar. Similar to the observations for TAG, the proportions of 14C in PL out of total lipids showed no significant cultivar difference at three and seven days after injection.

increased the understanding of oil metabolism in cereals. However, the knowledge of metabolic and genetic factors that explain why oat is unusual among the cereals in having high amounts of endosperm oil, as well as why some oat varieties have high oil content as compared to others, remains incomplete. In order to further biochemically characterize the oat endosperm and compare lipid synthesis in two genetically very close oat cultivars with different endosperm oil concentration, 14C-labelled substrates from both upstream and downstream parts of glycolysis were supplied to the oat grain and their metabolic fate determined. Substrates were supplied to the grain endosperm either through the stems of detached oat panicles in liquid nutrient medium in vitro (a method previously shown to mimic grain composition during development in planta (Ekman et al., 2008)), or by direct injection in vivo using a syringe (Vigeolas and Geigenberger, 2004; Rolletschek et al., 2005). These two methods can both be regarded to be more close to in vivo conditions of developing seeds as compared to using isolated plastids. Liquid cultures of embryos from different species, even with attached endosperm, have been developed that closely mimic the development in planta (Chen and Shachar-Hill, 2012; Alonso et al., 2011). However, no such cultures have been developed for oat grains and such a method would any how not be able to address some of the issues here dealt with, such as carbon loading into the grain from the crease phloem and the influence of pericarp photosynthesis on carbon allocation. It can be pointed out that different methods available for substrate feeding into seed tissues, each with different advantages and drawbacks, are valuable and can complement each other. A high-oil phenotype could be the result from a high FA and oil synthesis and/or low oil turnover, and/or from a longer FA and oil accumulation period. FA and oil synthesis could also indirectly be influenced by the upstream competition of sugars for different storage compound pathways. This study aimed to dissect cultivar differences in some of these aspects.

4. Discussion

4.2. Preferred carbon sources for lipid labelling in the oat grain

4.1. Oat as a model species for oil accumulation in the cereal endosperm

Feeding studies on isolated plastids from developing seeds have shown that a range of glycolytic intermediates support FA synthesis but to different extents in different species. In oilseed rape embryos, mustard cotyledons and castor bean endosperms, pyruvate was superior to acetate as the more efficient substrate for FA synthesis, while malate was shown to be more effective than pyruvate or acetate in castor bean endosperms and sunflower embryos (Smith et al., 1992; Kang and Rawsthorne, 1996; Pleite et al., 2005; €uerle, 1986). In oilseed rape embryos, glucose-6Liedvogel and Ba phosphate and dihydroxyacetone phosphate supported higher

3.6. Temporal distribution of lipid classes

14

C from acetate between different

Among the cereals, significant oil accumulation in the endosperm is a unique feature of oat (Banas et al., 2007; Liu, 2011; Price and Parsons, 1975; Barthole et al., 2012). Previous studies on quantification and localization of oil in the oat grain both during seed filling and germination (Banas et al., 2007; Leonova et al., 2010; Heneen et al., 2008, 2009), as well as transcriptome and metabolite analyses of developing grains (Hayden et al., 2011), have

Fig. 6. Time course of distribution of 14C from acetate between different lipid classes 0 min up to seven days after direct injection into oat grain endosperm of medium-oil (MO) and high-oil (HO) cultivars. Results are the mean ± standard deviation for three samples.

352

Å. Grimberg / Plant Physiology and Biochemistry 83 (2014) 346e355

rates of FA synthesis than either malate or acetate (Kang and Rawsthorne, 1994). Most substrates tested in this study (sucrose, glucose, pyruvate, malate, and acetate) supplied to the oat grain were only incorporated into lipids to a relatively low extent. Acetate is a direct precursor of acetyl-CoA which is the starting metabolite in FA synthesis. The proportion of accumulated 14C from acetate in oat grain lipids (up to 15% of total grain 14C accumulation) by supply through detached panicles was unexpectedly low compared to in detached oat leaves (approximately 50% of total leaf 14C accumulation, data not shown) and acetate is known to be highly incorporated into lipids of isolated spinach leaf chloroplasts (Roughan et al., 1976). However, potential metabolic effects on longdistance transported substrates into developing grains on detached panicles must be considered, especially for substrates that are not well known transport metabolites. In general, the phloem is a conduit transport system for a variety of different molecules (sugars, peptides, proteins, nucleic acids, and lipids) for nutritional, defense, and developmental purposes (Benning et al., 2012). Sucrose is regarded to be the major transport form of photosynthetically fixed carbon even though hexoses were suggested to also be important carbon transport molecules (van Bel and Hess, 2008). Since pyruvate, acetate, and malate can cross the plastid membrane (Neuhaus and Emes, 2000), it is possible that these metabolites can be transported intact from the stem into the endosperm, even though this has not been studied. By changing supply method of substrates to direct injection in grains developing in planta, a method previously used on rape seeds and maize kernels (Vigeolas and Geigenberger, 2004; Rolletschek et al., 2005), the accumulation of 14C from acetate into lipids was increased to 46% of total accumulation in grain, which was much higher compared to other injected substrates (up to 16%). Having in mind that the oil concentration of oat grains is relatively low (6e10%) as compared to true oil seeds, acetate can be regarded to be a good marker for lipids in the oat grain endosperm, if supplied by injection. The contrasting proportions of acetate incorporation into lipids between the two different supply methods used in this study indicates that acetate might be metabolized somewhere along the transport route or during the unloading mechanism into the oat grain endosperm when supplied through detached panicles. Even though 14C from the different substrates that were supplied through the detached panicle system accumulated in grain lipids to a proportionally low extent (6e20% of total recovered 14C in the grain), significant cultivar differences were still observed with higher proportions (3e8%) of total recovered 14C found in the lipid fraction of the HO as compared to the MO cultivar, for all substrates tested. This indicates a higher partitioning of carbons into lipids in the HO as compared to the MO cultivar, which might be caused by a higher rate of FA synthesis. If grain lipids are not turned over during development in any of the studied cultivars, a higher rate of FA synthesis would be in agreement with the higher rate of total lipid accumulation observed in grains developing on detached panicles of the HO as compared to MO cultivar at the earlier stages of development (Banas et al., 2007; Ekman et al., 2008). It can also be speculated that, at least for the sugars, the higher partitioning of carbons into lipids could be the result from a lower upstream competition of carbons for starch synthesis in the HO cultivar (meaning a lower starch synthesis capacity), which would be in agreement with the decreasing starch content associated with increased oil in oat (Peterson and Wood, 1997). 4.3. Cultivar differences in carbon transport and unloading When substrates were instead directly injected into the grain, the proportions of net accumulated 14C in the grain that were found

in lipids did not show any cultivar difference. This might be explained by cultivar differences in the efficiency of carbon transport and unloading mechanisms into the grain which can only be revealed when feeding developing oat grains through the detached panicle close to in vivo conditions. The method of direct injection bypasses this natural route for carbon uptake and instead forces a predefined small amount of 14C into the endosperm from where it enters metabolism. Sugar ratios (sucrose/hexose) are thought to regulate carbon channelling into, and also carbon partitioning within, sink organs and are governed by the sucrose cleaving enzymes sucrose synthases (EC 2.4.1.13) and invertases (EC 3.2.1.26) (Ruan, 2014; Weschke et al., 2003; Sturm and Tang, 1999). Interestingly, in a transcriptome study it was shown that the expression of genes encoding sucrose transporters and a sucrose synthase were similar, whereas a gene encoding an invertase was lower, in the HO as compared to the MO cultivar, and other transcripts encoding hexose metabolizing enzymes were also shown to be differentially expressed between cultivars (Hayden et al., 2011). In the same study the levels of both sucrose and hexoses were shown to be higher in grains of the HO as compared to the MO cultivar (Hayden et al., 2011) which is in agreement with our and previous study (Ekman et al., 2008) of oat grains where the net accumulation of total 14C label per grain from sucrose (supplied through detached panicles) was significantly higher in the HO compared to the MO cultivar. This indicates that mechanisms involved in the transport and unloading of sucrose can influence the allocation of carbon between different storage compounds in the grain and can possibly be one reason to the high-oil phenotype of the HO cultivar. 4.4. Cultivar differences in lipid synthesis was localized to endosperm The HO cultivar allocated a higher share of 14C from acetate fed through detached panicles into lipids (8% higher of total recovered radioactivity in grain) as compared to the MO cultivar but it was not known to which part of the grain this difference was localized to. One benefit with feeding carbon substrates by the natural route through the stem (even though potentially metabolized prior to reaching the grain endosperm as discussed in 4.2) is that it will be distributed in a natural way into different parts of the grain, as compared to injecting substrates directly into the endosperm where it will probably be metabolized more local at the injection site. Therefore, the spatial distribution of acetate label was only analysed for grains fed through the detached panicles. The results clearly showed that cultivar differences in carbon partitioning into lipid synthesis was localized to the oat endosperm, and not to the small but lipid (oil) dense embryo, which might indicate a higher FA synthesis in the endosperm of the HO as compared to the MO cultivar. This is well in agreement with previous determined spatial distribution of 14C from sucrose fed to grains through detached panicles (Ekman et al., 2008) as well as with analytical data and microscope analysis on lipid accumulation in grains of these oat cultivars (Banas et al., 2007; Heneen et al., 2009). This localization of cultivar differences in lipid metabolism to the endosperm is of great importance since this tissue makes out the largest part of the grain and should therefore be the target for the potential development of high-oil yielding cereals. 4.5. Cultivar differences in triacylglycerol synthesis Since 14C from acetate was shown to label the lipid fraction of oat grains to a very high extent (up to 46% of net accumulated 14C in the grain) when supplied by direct injection as compared to all other substrates tested in our study, the fate of 14C from acetate was

Å. Grimberg / Plant Physiology and Biochemistry 83 (2014) 346e355

followed from the injection moment up to seven days after injection. The variability in net accumulated 14C in grains from injected acetate was too high to resolve potential cultivar differences in absolute amounts of carbon uptake into FA synthesis, but the variability in distribution of this 14C between different compounds in the grain was low. The relative partitioning of net accumulated 14 C from acetate into lipids did not show any cultivar differences despite the known cultivar difference in oil content, which was at first surprising. On the other hand, the distribution of 14C between different lipid classes revealed large cultivar differences. To explain this it should first be noted that free acetate is not regarded to be the main substrate for FA synthesis in oil accumulating cells in vivo (Li-Beisson et al., 2010; Oliver et al., 2009, and see discussion in 4.6), and requires conversion to acetyl-CoA by acetyl-CoA synthetase (EC 6.2.1.1) before it is available for FA synthesis by acetyl-CoA carboxylase (ACC, EC 6.4.1.2). Acetyl-CoA synthetase was actually suggested to be important for the detoxification of fermentation metabolites in vegetative tissues (Li-Beisson et al., 2010; Oliver et al., 2009), while ACC has been linked to a QTL identified to have a major effect on oil content in oat grains (Kianian et al., 1999). It is therefore possible that the HO cultivar has a higher rate of FA synthesis compared to the MO related to ACC. However, this cultivar difference cannot be revealed in our study when using directly injected acetate since the activity of acetyl-CoA synthetase will probably be minor as compared to the activity of pyruvate dehydrogenase, for the total production of acetyl-CoA used for FA synthesis. Further in vitro studies will be needed to confirm this hypothesis. It is possible that substrates fed through the detached panicle were metabolized prior to reaching the grain (see 4.2). This could mean that carbons from acetate entered the grain as another metabolite that did not enter FA synthesis through acetyl-CoA synthetase, which could possibly explain the cultivar difference observed for the partitioning of 14C from acetate into lipids, when using the detached panicle method, in contrast to when using the injection method. This means that we cannot conclude from our data using directly injected 14C-acetate if the oat cultivars have different rates of FA synthesis, but after acetate has been taken up in FA synthesis our data show that the formed FAs are channelled into TAG synthesis faster and also reach the maximum level in TAG earlier, in the HO as compared to the MO cultivar. Moreover, the proportion of 14C from acetate found in the PL fraction out of 14C in total lipids showed a reverse pattern to that of 14C in TAG, starting at a much higher level and with a later start of decrease in the MO as compared to the HO cultivar, which coincided with the maximum levels in TAG. Therefore, our data suggest that one reason to the higher oil content in the HO cultivar is a more efficient transfer of formed FAs from PL into TAG (Stymne and Stobart, 1987). To further elucidate which specific enzymes in acyl lipid metabolism that exert a strong influence in the flow of acyl groups from PL to TAG, one approach could be to use metabolic control analysis (Harwood et al., 2013), but also classical biochemistry. There is a number of enzymes involved in this process, such as phospholipid:diacylglycerol acyltransferase (PDAT), acyl-CoA:lysophosphatidylcholine acyltransferase (LPCAT), and phospholipases (Bates et al., 2013). An important finding from the time-course analysis of injected 14 C from acetate during seven days was that the proportions of total 14 C accumulation in lipids that were found in TAG increased and did not decrease after reaching the maximum levels in any of the cultivars. Together with the estimation that the amounts of 14C from acetate accumulating in TAG did not seem to decrease after injection, this indicated that turnover of TAG is probably not a major factor involved in determining the final oil concentration in the oat grain, at least not during the developmental period studied. Oat endosperm has very low amount of oleosins, causing the lipid

353

droplets to fuse (Heneen et al., 2008). Oleosins have been shown to prevent TAG mobilization in leaf cells (Fan et al., 2013) which otherwise are rapidly turned over in these cells. Therefore it can be assumed that the oat endosperm lack active TAG degrading enzymes during grain development. However, it cannot be ruled out though that turnover of TAG in the oat grain occurs at later developmental stages as suggested in a previous study of fatty acid profile changes of oat grain lipids of the MO cultivar (Banas et al., 2007). Knowledge about potential TAG degradation during grain development is important when developing a strategy to increase oil content of cereal endosperms since it could significantly influence the final oil concentration of the mature grain. It should be pointed out that rapid degradation of oat endosperm TAG occurs upon germination, probably due to secretion of lipases from the scutellum (Leonova et al., 2010; Jensen and Heltved, 1982). 4.6. The role of light for carbon accumulation Even though acetate was shown to label lipids to a high extent in our data on oat grains and for example on spinach leaf plastids and developing oilseed rape embryos (Roughan et al., 1976; Perry and Harwood, 1993), 13C metabolic flux analyses on several oilaccumulating plant storage tissues showed that acetate is probably not a major entry point of carbon flux into the plastid for FA synthesis in vivo. From available 13C metabolic flux analysis data of intact embryos of oil seeds from rape, soybean, sunflower and maize it appears that two general patterns for the origin of carbons precursors for FA synthesis can be drawn: In green photoheterotrophic seeds i) the major entry point of carbon flux into the plastid is at the level of trios-P coming from both cytosolic glycolysis and the RuBisCO bypass, ii) NAD(P)H and ATP is mainly supplied by light-driven photosynthesis. In non-green heterotrophic seeds i) the major entry point of carbon flux into plastid is more upstream in glycolysis as hexose-P with both plastidic glycolysis and OPPP contributing to pyruvate formation ii) the major provider of NAD(P)H is the OPPP and with ATP coming from mitochondrial respiration (Baud and Lepiniec, 2010; Alonso et al., 2007; Schwender et al., 2004; Goffman et al., 2005; Alonso et al., 2010; Allen et al., 2009). The only cereal endosperm characterized using 13 C metabolic flux analyses up to date is that of maize, which has basically no oil in the endosperm in contrary to oat (Alonso et al., 2011). The oat grain endosperm is a heterotrophic tissue during seed filling, but the pericarp is green. Our data showed that availability of light can dramatically affect the metabolic status of the oat grain observed as a severe reduction (63%) of net carbon accumulation from sucrose when detached panicles of the HO cultivar were exposed to darkness indicating either a decrease in uptake, or an increase in respiration, of carbons. The distribution of 14C from supplied sucrose between different storage compounds also changed slightly in both cultivars when oat grains on detached panicles developed in darkness with a lower proportion (3% less of recovered radioactivity) going into lipids which probably reflects a lower energy state resulting in a decreased investment in the energetically expensive oil synthesis. Since the effect of darkness on total carbon incorporation to the grain was much smaller (13% decrease) for the MO cultivar and not statistically significant, it can be speculated that the HO cultivar is much more dependent on photosynthetic contribution, as compared to the MO cultivar, to keep its normal carbon metabolism. It has been suggested that the role of pericarp photosynthesis for storage accumulation of cereal grains is as O2 provider through the light reactions which stimulates mitochondrial respiration and therefore ATP supply, and not for CO2 fixation since only 2% of final starch in the mature grain can be explained by grain photosynthesis (Watson and Duffus, 1988). In

354

Å. Grimberg / Plant Physiology and Biochemistry 83 (2014) 346e355

agreement with this, oxygen levels in developing grains of barley were shown to be highest close to the pericarp photosynthetic cell layers, with steep gradients within the endosperm (Rolletschek et al., 2004). In fact, it was suggested that oxygen availability plays a general role for the balance of lipid/starch storage in seeds (Rolletschek et al., 2005) which maybe explains why oil in most cereal grains is mainly located in the aleurone layer and nearby cells. It is possible that oat grains developing in darkness have a reduced energy state caused by decreased oxygen levels resulting in reduced carbon import into the oat grain. Previous measurements actually indicated that grains of the HO cultivar had higher respiration rate implicating a higher reliance on mitochondrial oxidative phosphorylation as supplier of ATP (and therefore a higher O2 consumption), compared to the MO cultivar (Hayden et al., 2011) which might explain why the HO cultivar was more severely affected by exposure to darkness. Interestingly, metabolite analyses showed that ATP and NAD(P)H levels in the HO cultivar was more or less exhausted as compared to the MO cultivar which probably indicates a higher consumption of these co-factors due to a higher lipid synthesis in the HO cultivar (Hayden et al., 2011).

injection using a syringe, but also highlighted the importance of choosing the right substrate feeding method depending on what should be studied. To determine the relative fluxes of carbon into core metabolic pathways using 13C flux analyses in the oat grain endosperm and compare with the study previously made for maize (Alonso et al., 2011) could significantly expand the understanding of the underlying mechanisms explaining the differences in oil accumulating capacities between cereal endosperms with large differences in oil content. Acknowledgements This work was supported by the Swedish Research Council for Environment, Agricultural Sciences and Spatial Planning (Formas) and the Swedish Governmental Agency for Innovation System (VINNOVA). I would also like to thank Prof. Sten Stymne for valuable discussions and for reading and commenting on the manuscript. Contributions to the manuscript The sole author was responsible for planning and performing the experiments as well as writing the manuscript.

4.7. Comparison of substrate supply methods References Supplying substrates through the stem of detached oat panicles very well mimics the natural conditions of how photosynthetically fixed carbon from leaves are long-distance transported through the crease of the grain further passing the transfer region of cells to finally enter the endosperm and embryo (Thorne, 1985). Even though, most importantly, cultivar differences in grain composition is contained using this method, a limitation is a lower seed filling compared to in planta conditions (Ekman et al., 2008), and also that fed substrates might be metabolized on the way through the stem meaning that substrates should be chosen with care. By directly injecting the substrate to the grain endosperm using a syringe, the problem of not knowing in what form the substrate enters the endosperm is circumvented and since grains are still on the intact plant during the experiment it very well mimics natural conditions for seed filling in vivo. A limitation to this method is that the syringe is leaving a small wound on the seed that might affect results, even though data on rape seeds indicated that it did not (Vigeolas et al., 2007). Another limitation is that there is a high variability in the total amount of injected substrate due to practical difficulties at the injection moment (discussed in 3.5), even though the variability of the proportional distribution of substrate between different grain fractions or lipid species was low. Important to take into account is also that injected substrate is probably mainly localized to the endosperm, not the embryo, which might be a benefit if endosperm metabolism should be studied. It was also suggested from this study that even though labelling the lipid fractions to a very high extent, injected acetate could not be used to study carbon partitioning into FA synthesis, but instead to study the redistribution of carbon between different lipid species after it has been incorporated into FA synthesis. 5. Conclusions The understanding of oil synthesis in the cereal endosperm and the reasons to why oat is unique among the cereals in storing relatively high amounts of endosperm oil is still partial. Knowledge from studies on oat can be crucial for developing a strategy for developing high-oil cereals by traditional breeding and/or genetic engineering. This study identified 14C-acetate to label lipids in the cereal grain endosperm to a very high extent if supplied by direct

Allen, D.K., Ohlrogge, J.B., Shachar-Hill, Y., 2009. The role of light in soybean seed filling metabolism. Plant J. 58, 220e234. Alonso, A.P., Goffman, F.D., Ohlrogge, J.B., Shachar-Hill, Y., 2007. Carbon conversion efficiency and central metabolic fluxes in developing sunflower (Helianthus annuus L.) embryos. Plant J. 52, 296e308. Alonso, A.P., Dale, V.L., Shachar-Hill, Y., 2010. Understanding fatty acid synthesis in developing maize embryos using metabolic flux analysis. Metab. Eng. 12, 488e497. Alonso, A.P., Val, D.L., Shachar-Hill, Y., 2011. Central metabolic fluxes in the endosperm of developing maize seeds and their implications for metabolic engineering. Metab. Eng. 13, 96e107. Banas, A., Debski, H., Banas, W., Heneen, W.K., Dahlqvist, A., Bafor, M., Gummeson, P.O., Martilla, S., Ekman, Å., Carlsson, A.S., Stymne, S., 2007. Lipids in grain tissues of oat (Avena sativa): differences in content, time of deposition, and fatty acid composition. J. Exp. Bot. 58, 2463e2470. Barthole, G., Lepiniec, L., Rogowsky, P.M., Baud, S., 2012. Controlling lipid accumulation in cereal grains. Plant Sci. 185e186, 33e39. Bates, P.D., Stymne, S., Ohlrogge, J., 2013. Biochemical pathways in seed oil synthesis. Curr. Opin. Plant Biol. 16, 358e364. Baud, S., Lepiniec, L., 2010. Physiological and developmental regulation of seed oil production. Prog. Lipid Res. 49, 235e249. Baud, S., Mendoza, M.S., To, A., Harscoet, E., Lepiniec, L., Dubreucq, B., 2007. WRINKLED1 specifies the regulatory action of LEAFY COTYLEDON2 towards fatty acid metabolism during seed maturation in Arabidopsis. Plant J. 50, 825e838. Benning, U.F., Tamot, B., Guelette, B.S., Hoffmann-Benning, S., 2012. New aspects of phloem-mediated long-distance lipid signaling in plants. Front. Plant Sci. 3, 00053. Bligh, E.G., Dyer, W.J., 1959. A rapid method of total lipid extraction and purification. Can. J. Biochem. Physiol. 37, 911e917. Carlsson, A.S., Yilmaz, J.L., Green, A.G., Stymne, S., Hofvander, P., 2011. Replacing fossil oil with fresh oil e with what and for what? Eur. J. Lipid Sci. Technol. 113, 812e831. Chen, X., Shachar-Hill, Y., 2012. Insights into metabolic efficiency from flux analysis. J. Exp. Bot. 63, 2343e2351. De Vries, F.W.T.P., Brunsting, A.H.M., Van Laar, H.H., 1974. Products, requirements and efficiency of biosynthesis a quantitative approach. J. Theor. Biol. 45, 339e377. Dudley, J.W., Lambert, R.J., 2010. 100 generations of selection for oil and protein in corn. In: Plant Breeding Reviews. John Wiley & Sons, Inc., pp. 79e110. Ekman, Å., Hayden, D.M., Dehesh, K., Bulow, L., Stymne, S., 2008. Carbon partitioning between oil and carbohydrates in developing oat (Avena sativa L.) seeds. J. Exp. Bot. 59, 4247e4257. Fan, J., Yan, C., Zhang, X., Xu, C., 2013. Dual role for phospholipid:diacylglycerol acyltransferase: enhancing fatty acid synthesis and diverting fatty acids from membrane lipids to triacylglycerol in Arabidopsis leaves. Plant Cell. 25, 3506e3518. FAOSTAT, 2012. Food and Agriculture Organization of the United Nations. 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, 91e101.

Å. Grimberg / Plant Physiology and Biochemistry 83 (2014) 346e355 Goffman, F.D., Alonso, A.P., Schwender, J., Shachar-Hill, Y., Ohlrogge, J.B., 2005. Light enables a very high efficiency of carbon storage in developing embryos of rapeseed. Plant Physiol. 138, 2269e2279. Harwood, J.L., Ramli, U.S., Tang, M., Quant, P.A., Weselake, R.J., Fawcett, T., Guschina, I.A., 2013. Regulation and enhancement of lipid accumulation in oil crops: the use of metabolic control analysis for informed genetic manipulation. Eur. J. Lipid Sci. Technol. 115, 1239e1246. Hayden, D.M., Rolletschek, H., Borisjuk, L., Corwin, J., Kliebenstein, D.J., Grimberg, A., Stymne, S., Dehesh, K., 2011. Cofactome analyses reveal enhanced flux of carbon into oil for potential biofuel production. Plant J. 67, 1018e1028. Heneen, K.W., Karlsson, G., Brismar, K., Gummeson, P.O., Marttila, S., Leonova, S., Carlsson, A.S., Bafor, M., Banas, A., Mattsson, B., Debski, H., Stymne, S., 2008. Fusion of oil bodies in endosperm of oat grains. Planta 228, 589e599. Heneen, W.K., Banas, A., Leonova, S., Carlsson, A.S., Marttila, S., Debski, H., Stymne, S., 2009. The distribution of oil in the oat grain. Plant Signal Behav. 4, 55e56. Jensen, S., Heltved, F., 1982. Visualization of enzyme activity in germinating cereal seeds using a lipase sensitive fluorochrome. Carlsb. Res. Commun. 47, 297e303. Kang, F., Rawsthorne, S., 1994. Starch and fatty acid synthesis in plastids from developing embryos of oilseed rape (Brassica napus L.). Plant J. 6, 795e805. Kang, F., Rawsthorne, S., 1996. Metabolism of glucose-6-phosphate and utilization of multiple metabolites for fatty acid synthesis by plastids from developing oilseed rape embryos. Planta 199, 321e327. Kianian, S.F., Egli, M.A., Phillips, R.L., Rines, H.W., Somers, D.A., Gengenbach, B.G., Webster, F.H., Livingston, S.M., Groh, S., O'Donoughue, L.S., Sorrells, M.E., Wesenberg, D.M., Stuthman, D.D., Fulcher, R.G., 1999. Association of a major groat oil content QTL and an acetyl-CoA carboxylase gene in oat. Theor. Appl. Genet. 98, 884e894. Leonova, S., Grimberg, Å., Marttila, S., Stymne, S., Carlsson, A.S., 2010. Mobilization of lipid reserves during germination of oat (Avena sativa L.), a cereal rich in endosperm oil. J. Exp. Bot. 61, 3089e3099. Li-Beisson, Y., Shorrosh, B., Beisson, F., Andersson, M.X., Arondel, V., Bates, P.D., Baud, S., Bird, D., Debono, A., Durrett, T.P., Franke, R.B., Graham, I.A., Katayama, K., Kelly, A.A., Larson, T., Markham, J.E., Miquel, M., Molina, I., Nishida, I., Rowland, O., Samuels, L., Schmid, K.M., Wada, H., Welti, R., Xu, C., Zallot, R., Ohlrogge, J., 2010. Acyl-lipid metabolism. Arab. Book 8, 11. Liedvogel, B., B€ auerle, R., 1986. Fatty-acid synthesis in chloroplasts from mustard (Sinapis alba L.) cotyledons: formation of acetyl coenzyme A by intraplastid glycolytic enzymes and a pyruvate dehydrogenase complex. Planta 169, 481e489. Liu, K., 2011. Comparison of lipid content and fatty acid composition and their distribution within seeds of 5 small grain species. J. Food Sci. 76, C334eC342. Moose, S.P., Dudley, J.W., Rocheford, T.R., 2004. Maize selection passes the century mark: a unique resource for 21st century genomics. Trends Plant Sci. 9, 358e364. Neuhaus, H.E., Emes, M.J., 2000. Nonphotosynthetic metabolism in plastids. Annu Rev. Plant Physiol. Plant Mol. Biol. 51, 111e140. Oliver, D.J., Nikolau, B.J., Wurtele, E.S., 2009. Acetyl-CoAdlife at the metabolic nexus. Plant Sci. 176, 597e601. Perry, H.J., Harwood, J.L., 1993. Radiolabelling studies of acyl lipids in developing seeds of Brassica napus: use of [1-14C]acetate precursor. Phytochemistry 33, 329e333. Peterson, D.M., Wood, D.F., 1997. Composition and structure of high-oil oat. J. Cereal Sci. 26, 121e128.

355

s, R., Martínez-Force, E., Rawsthorne, S., 2005. The sources Pleite, R., Pike, M.J., Garce of carbon and reducing power for fatty acid synthesis in the heterotrophic plastids of developing sunflower (Helianthus annuus L.) embryos. J. Exp. Bot. 56, 1297e1303. Price, P.B., Parsons, J.G., 1975. Lipids of seven cereal grains. J. Am. Oil Chem. Soc. 52, 490e493. Rolletschek, H., Weschke, W., Weber, H., Wobus, U., Borisjuk, L., 2004. Energy state and its control on seed development: starch accumulation is associated with high ATP and steep oxygen gradients within barley grains. J. Exp. Bot. 55, 1351e1359. Rolletschek, H., Koch, K., Wobus, U., Borisjuk, L., 2005. Positional cues for the starch/ lipid balance in maize kernels and resource partitioning to the embryo. Plant J. 42, 69e83. Roughan, P.G., Slack, C.R., Holland, R., 1976. High rates of [1-14C]acetate incorporation into the lipid of isolated spinach chloroplasts. Biochem. J. 158, 593e601. Ruan, Yong-Ling, 2014. Sucrose metabolism: gateway to diverse carbon use and sugar signaling. Annu. Rev. Plant Biol. 65, 33e67. Schwender, J., Ohlrogge, J.B., 2002. Probing in vivo metabolism by stable isotope labeling of storage lipids and proteins in developing Brassica napus embryos. Plant Physiol. 130, 347e361. Schwender, J., Goffman, F., Ohlrogge, J.B., Shachar-Hill, Y., 2004. Rubisco without the Calvin cycle improves the carbon efficiency of developing green seeds. Nature 432, 779e782. Shen, B., Allen, W.B., Zheng, P., Li, C., Glassman, K., Ranch, J., Nubel, D., Tarczynski, M.C., 2010. Expression of ZmLEC1 and ZmWRI1 increases seed oil production in Maize. Plant Physiol. 153, 980e987. Smith, R.G., Gauthier, D.A., Dennis, D.T., Turpin, D.H., 1992. Malate- and pyruvatedependent fatty acid synthesis in leucoplasts from developing castor endosperm. Plant Physiol. 98, 1233e1238. Sturm, A., Tang, G.Q., 1999. The sucrose-cleaving enzymes of plants are crucial for development, growth and carbon partitioning. Trends Plant Sci. 4, 401e407. Stymne, S., Stobart, A.K., 1987. Triacylglycerol biosynthesis. Biochem. Plants 9, 175e214. Thorne, J.H., 1985. Phloem Unloading of C and N assimilates in developing seeds. Annu. Rev. Plant Physiol. 36, 317e343. van Bel, A.J.E., Hess, P.H., 2008. Hexoses as phloem transport sugars: the end of a dogma? J. Exp. Bot. 59, 261e272. Vanhercke, T., Wood, C.C., Stymne, S., Singh, S.P., Green, A.G., 2013. Metabolic engineering of plant oils and waxes for use as industrial feedstocks. Plant Biotechnol. J. 11, 197e210. Vigeolas, H., Geigenberger, P., 2004. Increased levels of glycerol-3-phosphate lead to a stimulation of flux into triacylglycerol synthesis after supplying glycerol to developing seeds of Brassica napus L. in planta. Planta 219, 827e835. Vigeolas, H., Waldeck, P., Zank, T., Geigenberger, P., 2007. Increasing seed oil content in oil-seed rape (Brassica napus L.) by over-expression of a yeast glycerol-3phosphate dehydrogenase under the control of a seed-specific promoter. Plant Biotechnol. J. 5, 431e441. Watson, P.A., Duffus, C.M., 1988. Carbon dioxide fixation by detached cereal caryopses. Plant Physiol. 87, 504e509. Weschke, W., Panitz, R., Gubatz, S., Wang, Q., Radchuk, R., Weber, H., Wobus, U., 2003. The role of invertases and hexose transporters in controlling sugar ratios in maternal and filial tissues of barley caryopses during early development. Plant J. 33, 395e411.