Metabolomic analysis of pathways related to rice grain chalkiness by a notched-belly mutant with high occurrence of white-belly grains
© The Author(s). 2017
Received: 24 August 2016
Accepted: 25 January 2017
Published: 7 February 2017
Grain chalkiness is a highly undesirable trait deleterious to rice appearance and milling quality. The physiological and molecular foundation of chalkiness formation is still partially understood, because of the complex interactions between multiple genes and growing environments.
We report the untargeted metabolomic analysis of grains from a notched-belly mutant (DY1102) with high percentage of white-belly, which predominantly occurs in the bottom part proximal to the embryo. Metabolites in developing grains were profiled on the composite platforms of UPLC/MS/MS and GC/MS. Sampling times were 5, 10, 15, and 20 days after anthesis, the critical time points for chalkiness formation. A total of 214 metabolites were identified, covering most of the central metabolic pathways and partial secondary pathways including amino acids, carbohydrates, lipids, cofactors, peptides, nucleotides, phytohormones, and secondary metabolites. A comparison of the bottom chalky part and the upper translucent part of developing grains of DY1102 resulted in 180 metabolites related to chalkiness formation.
Generally, in comparison to the translucent upper part, the chalky endosperm had lower levels of metabolites regarding carbon and nitrogen metabolism for synthesis of storage starch and protein, which was accompanied by perturbation of pathways participating in scavenging of reactive oxygen species, osmorugulation, cell wall synthesis, and mineral ion homeostasis. Based on these results, metabolic mechanism of chalkiness formation is discussed, with the role of embryo highlighted.
KeywordsGrain chalkiness Metabolomic analysis Notched-belly mutants Rice White-belly
Chalkiness is the opaque portion in rice endosperm due to loose packing of starch granules and protein bodies [1, 2]. Depending on its location on or within the endosperm, chalkiness is grouped into white-belly, white-core, white-back, etc. Grain chalkiness is a highly undesirable trait, with detrimental effects on rice appearance quality as well as milling, eating, and cooking quality. High occurrence of grain chalkiness has been a major problem facing many rice-producing areas throughout the world, in particular in a warming climate . Reducing chalky grain rate has been one of the primary objectives for rice breeders and producers.
Chalkiness is a complex quantitative trait, governed by polygenes and subjected to large variations in environmental factors like temperature and humidity and cultural practices such as fertilization and irrigation . By molecular marker-based QTL analyses of numerous mapping populations, more than 140 QTLs were identified across all 12 chromosomes for the chalkiness trait, mostly among Asian cultivars . Of these, several major QTLs such as on chromosomes 1, 5, 7, 8, and 9 are stably expressed across multiple environments or various populations . In addition, several possible candidate genes responsible for grain chalkiness have been cloned by fine mapping, including pyruvate orthophosphate dikinase, starch synthase IIIa, UDP-glucose pyrophosphorylase, cell wall invertase, and H + -translocating pyrophosphatase (V-PPase), as reviewed by Sreenivasulu et al. . Based on this, the molecular basis of chalkiness formation is partly understood. Li et al.  studied the function of Chalk5, which encodes V-PPase with inorganic pyrophosphate hydrolysis and H+-translocation activity. They found that Chalk5 could disturb pH homeostasis in the endomembrane trafficking system, resulting in an abnormal decrease in protein body number and size, causing aberrant shape and spatial arrangements of starch granules and protein bodies.
Rice quality is essentially determined by chemical composition of the storages of starch and protein that derive from carbon (C) and nitrogen (N) metabolism during grain filling. Metabolomics is one of the “omics” studies for integrative analysis of metabolites in organisms . Together with genomics, transcriptomics, and proteomics, it is a powerful tool to provide a link between genotype and phenotype and to generate new knowledge of systems biology. Besides providing an understanding of the metabolic state in plants under various circumstances, it can be applied to clarify the functions of unknown genes by using natural variants or mutants of the target plants, and might be useful in breeding programs because of the close relation between quality traits like taste and metabolic conditions . In rice, metabolomic approaches have yielded new insights into the composition and regulation of seed metabolism , correlations between the metabolic phenotype and geographic origin of japonica and indica rice , grain filling-related metabolism under high temperature , food quality prediction , and the bioactive compounds in cooked rice . However, there is little information concerning the metabolomic analysis of grain chalkiness formation, and the metabolic foundation of it remains unclear. In addition, few studies were conducted to investigate the metabolic changes during the early stage of grain filling, which is the critical time point for chalkiness formation.
Using chemical mutagen of EMS, we indentified a notched-belly mutant with high ratio of white-belly grains . The notched-line divided grains into two parts, with the upper part nearly all translucent whereas the bottom part, proximal to the embryo, having high occurrence of white-belly. Using the matured grains of this mutant, Lin et al.  developed a novel comparison system that can minimize the influence of genetic background and growing environment, and conducted a comprehensive survey of endosperm proteomics, unraveling diverse but well regulated pathways responsible for grain chalkiness. In this study, we profiled and compared the metabolomes between the upper translucent part and the bottom chalky part of developing grains from the notched-belly mutant. In total, 214 metabolites were identified, among them the abundances of 180 metabolites were significantly altered in the chalky part. These altered metabolites indicated that chalkiness formation was associated with a metabolic shift from C and N metabolism to reactive oxygen species (ROS) scavenging, osmorugulation, cell wall synthesis, and iron homeostasis. Levels of metabolites related to secondary metabolism and plant hormone biosynthesis were also remarkably altered in the chalky part. These results extend our understanding of metabolomic mechanism underlying grain chalkiness.
Metabolite profiling of rice endosperm
Positional variation of white-belly occurrence within panicle under different growing conditions for the notched-belly mutant
Notched-belly grain/total grain (%)
White-belly grain/notched-belly grain (%)
In previous proteomic survey of the mature endosperm of the mutant, Lin et al.  found that several key proteins involved in carbohydrate metabolism (especially cell wall synthesis) and protein synthesis, folding and degradation were responsible for rice grain chalkiness. These proteomic data prompted us to investigate the metabolic changes related to chalky tissue formation during grain development.
Metabolic changes regarding developmental stages
During rice seed development, the embryo and the endosperm develop synchronously in terms of cell division and differentiation, and complete their morphological differentiation and development on about 10 DAA . After this, embryo tissues ripen physiologically, and the endosperm accumulated the majority of its storage substance on 25 to 30 DAA. Notably, the formation of chalkiness starts on about 5 DAA, as reported by Li et al.  who showed that Chalk5 expression levels were much higher on 5 and 10 DAA.
Number of the significantly altered metabolites in relation to developmental and chalkiness effects (p ≤ 0.05)
Total number of the significantly changed
Number of increase
Number of decrease
Altered amino acid metabolism in the chalky endosperm
Glutamate-derived amino acids
Amino acids Glu, Gln, His, and Pro were decreased in the chalky endosperm, while Arg was up-regulated (Fig. 4). The three common polyamines in plants, putrescine (Put), spermidine (Spd), and spermine (Spm) were all elevated since 10 DAA. Stachydrine (proline betaine) was increased during all the four sampling times. The acetylation of glutamate into N-acetylglutamate (NAG) is the first step of conversion of Glu to Arg. The higher level of NAG was accompanied by higher levels of citrulline and Arg on 15 DAA and 20 DAA, despite of reduction in ornithine on 5 DAA. The chalky endosperm also had lower levels of γ-aminobutyric acid (GABA), but this compound declined in both parts over time, reaching its minimum on 15 DAA. In addition, chalky endosperm showed lower levels of oxidized glutathione (GSSG) in the mid-development time points, suggesting a less oxidative environment. Consistent with this, there were lower levels of reduced glutathione (GSH) and 5-oxoproline (5-OP).
Aspartate-derived amino acids
Generally, amino acids Asn, Ile, Thr, and Lys showed a decreasing trend in the chalky endosperm, while Asp was increased on 15 DAA and Asn slightly increased on 10 DAA as well (Fig. 4). In addition, 2-aminoadipate, derived from Lys, was decreased over time. N-6-trimethyllysine, a protein-lysine degradation product, was significantly higher in chalky endosperm at the late stages of sampling (15 and 20 DAA).
Serine-derived amino acids
Aromatic amino acids
The aromatic amino acids, Phe, Tyr, and Trp, are central molecules in plant metabolism, serving as precursors for a wide range of secondary metabolites. Their precursor, shikimate, was decreased on 5 DAA and 10 DAA, but increased on 15 DAA (Fig. 4). Trp and Phe showed a decreasing trend, while Tyr had no clear trend. By contrast, the derivative of Tyr, tyramine, was reduced. Serotonin is thought to serve antioxidant and defense functions in plants. It was detected to decrease in the chalky endosperm from to 5 DAA to 15 DAA. In addition, quinate, a metabolite synthesized in a lateral branch of the shikimate biosynthesis pathway, was increased on 10 DAA and 15 DAA. Although its physiological role has not been completely elucidated, quinate accumulation has been detected in plants during fungal invasion  and after treatment with herbicides that inhibit amino acid biosynthesis .
Sulfur-containing amino acids
Sulfur is an essential macronutrient, and is primarily used to synthesize Cys, Met, and numerous essential and secondary metabolites, such as 5′-methylthioadenosine (MTA), S-methylmethionine (SMM), and S-adenosylmethionine (SAM). Cys, Met, and SMM were reduced in the chalky endosperm, while MTA showed an increasing trend irrespective of a decrease on 5 DAA (Fig. 4). Methionine sulfoxide (MetO), the product of Met oxidation by reactive oxygen species, was reduced. Similarly, S-adenosyl-homocysteine (SAH), the intermediate involving in the metabolic regeneration of SAM through the SAM cycle, was also declined. In addition, O-acetylserine (OAS), the intermediate of the pathway of Ser to Cys, was decreased on 10 DAA.
Altered C metabolism in the chalky endosperm
One important metabolite pool in plants is the hexose phosphates, and another is composed of pentose phosphate pathway intermediates and the triose phosphates glyceraldehydes 3-phosphate and dihydroxyacetone phosphate . The hexose phosphate pool consists of three interconvertible metabolic intermediates: glucose 6-phosphate (G6P), glucose 1-phosphate (G1P), and fructose 6-phosphate (F6P), deriving from phosphorylation of free hexose or degradation of starch and sucrose. The hexose pool contributes intermediates to glycolysis as well as many biosynthetic processes including starch and sucrose synthesis, cell wall formation, and the oxidative reactions of the pentose.
F6P and its derivatives
F6P and intermediates in glycolysis pathway and tricarboxylic acid (TCA) cycle were altered in the chalky endosperm. F6P and 3-PGA in the chalky endosperm showed a similar trend, being not significantly different on 5 DAA and 10 DAA, but higher on 15 DAA and 20 DAA in comparison with that in the translucent endosperm (Fig. 5). Pyruvate increased slightly only on 10 DAA in the chalky endosperm. In TCA cycle, four metabolites, malate, fumarate, succinate, and isocitrate, were elevated at early maturation stage of 15 DAA or 20 DAA, whereas α-ketoglutarate was reduced on 5 DAA. Generally, these intermediates demonstrated an increasing pattern, indicated that in the chalky endosperm, glycolysis and TCA cycle was shifted to mainly produce selective intermediates involved in metabolism of amino acids, fatty acids, secondary metabolites, and cell wall.
G6P and its derivatives
In the chalky endosperm, G6P showed no significant variation from 5 DAA to 15 DAA, but increased significantly on 20 DAA (Fig. 5). It was worth noting that significantly increased G6P was accompanied by significantly decreased glucose. Glucose, the precursor of G6P and cell wall biosynthesis, was reduced during all the four sampling times. In addition, abundances of other plant cell wall precursors, including gluconate and xylose, showed the similar decreasing trend. On the other hand, arabitol did not change markedly in the chalky endosperm. In the pentose phosphate pathway, ribose 5-phosphate was significantly decreased on 15 DAA, while sedoheptulose-7-phosphate was not significant altered. Glycerate involving in photorespiration tended to be increased, slightly on 10 DAA and 20 DAA and markedly on 15 DAA. Furthermore, ribose and ribulose, the crucial components of nucleic acids, were decreased in the chalky endosperm.
G1P and its derivatives
G1P in the chalky endosperm was reduced on the beginning of cell division (5 DAA), and then kept at a comparable level after that (Fig. 5). Sucrose, the main C-transport substance from leaf to grain, was decreased on 10 DAA and 15 DAA, indicating the insufficient supply of photosynthetic assimilates of the source organs. Albeit there were significant increases of maltose, it was coupled with decrease in glucose and fructose, implying an accelerated breakdown of starch or a retarded conversion of maltose to glucose and fructose.
Chalky endosperm accumulated significantly higher levels of raffinose than translucent endosperm (Fig. 5). UDP-galactose, one of the precursors of raffinose, was also higher at most time points, although galactinol, its direct precursor, was similar in the two parts at most time points. Consistent with the increase in raffinose, two sugar alcohols, erythritol and threitol, also showed this pattern (Additional file 1).
In addition to higher raffinose, chalky endosperm also showed higher starch intermediates and hexose phosphates than translucent endosperm in the later time points (Additional file 1). The translucent part had higher levels of sucrose and hexoses in all time points except for the final one. Whether this pattern suggests different qualitative mechanisms for regulating carbon distribution in these lines, or if it only reflects slightly shifted kinetics, is not known.
Both kinds of endosperms accumulated phytic acid (inositol 6-phosphate, I6P) and its precursor I5P over the time course, with the trend accelerated on 20 DAA (Fig. 5). However, chalky endosperm showed slightly more phytic acid, and significantly more I5P in the last two time points. I3P was somewhat higher in chalky tissue only at the intermediate time points, while I1P was similar to I5P in its pattern, except that these compounds did not change in translucent endosperm over development. It is possible that early steps in phytate synthesis are accelerated in chalky endosperm, thus pushing the reactions; alternatively, slower conversion to the final product phytic acid in chalky endosperm might account for higher levels of intermediates.
For precursors of cell wall, rhamnose was decreased on 5 DAA, while arabinose did not varied significantly (Fig. 5). Furthermore, trehalose, an alpha,alpha-1,1-linked glucose disaccharide which is indispensable for regulating sugar utilization, was elevated on 20 DAA. It is worth noting that there was a co-occurring decrease of glycerol and glycerol-3-phophosphate (G3P), both of which are associated with lipid metabolism.
Other metabolites related to chalky endosperm formation
In general, most of lipids showed a decreasing trend in the chalky tissue (Additional file 1). However, it was noted a somewhat different kinetic pattern for developmental-related changes in several lipids. Four free fatty acids including eicosenoate, linoleate, linolenate, and oleate were all reduced in the chalky part. Similarly, glycerolipids like 1-linoleoylglycerol and 2-linoleoylglycerol, phospholipids such as 1-oleoylgly- cerophosphocholine and glycerophosphorylcholine, and phytosterols of beta-sitosterol and campesterol as well as their precursor squalene, had reduced level in the chalky endosperm. Phytosphingosine was low in chalky endosperm throughout development, but was high in translucent part early, followed by a decline. Its precursor sphinganine was similar for both parts. Notably, in choline pathway, choline was decreased, while its precursors choline phosphate and phosphoethanolamine were increased (Fig. 4).
In general, CPGECs increased as seed development progressed (Additional file 1). Pantothenate in CoA metabolism was increased for both the chalky and translucent parts, with the former having relatively higher level. In nicotinate and nicotinamide metabolism, nicotinamide adenine dinucleotide (NAD+) was slightly decreased in the chalky part on 5 DAA, while nicotinate and trigonelline (N’-methylnicotinate) were elevated during seed development. Phosphate decreased for both parts, and it was lower on 5 DAA and 10 DAA whereas slightly higher on 20 DAA in the chalky endosperm.
There were hints of differences in the oxidative stress environments for the two parts. Chalky endosperm showed higher levels of ascorbate, suggesting a less oxidative environment. Consistent with this, there were lower levels of two ascorbate catabolites, dehydroascorbate and threonate. In addition, amphiphilic antioxidants alpha-tocopherol and gamma-tocopherol were up-regulated on 20 DAA, but they were lower on 5 DAA and 10 DAA.
Purine and pyrimidine metabolism
Except adenine, adenosine, allantoin, guanosine-2′,3′-cyclic monophosphate, and uridine, the 17 detected metabolites in nucleic acid metabolism were all reduced for both parts during the four sampling times (Additional file 1). Compared with the translucent endosperm, chalky part had higher level of adenosine-2′,3′-cyclic monophosphate and 5′-GMP but lower level of adenosine, allantoin, and inosine.
The majority of peptides were decreased for both parts during the four sampling times, and the chalky endosperm showed lower levels (Additional file 1). Interestingly, the dipaptide threonylphenylalanine was increased markedly in the chalky parts, in particular on 15 DAA and 20 DAA.
Two phytohormones, ABA and IAA, as well as two compounds associated with physiological signaling in plants, serotonin and GABA, were detected (Fig. 4; Additional file 1). While the changes were not dramatic, chalky endosperm had lower ABA (although variable) early in seed development, while translucent part started higher and decreased over time. At the last time point ABA levels were essentially the same in both lines. IAA was similar for both lines, being slightly higher in chalky endosperm on 15 DAA, and showed an increase over time.
Benzenoid 4-hydroxybenzoate decreased in the chalky and translucent part as seed development progressed, and it was relatively higher in the chalky endosperm on 20 DAA (Additional file 1). Phenylpropanoids indentified were decreased during seed development, with the exception that coumarate did changed significantly. In addition, ferulate, sinapate, and coumaroylquinate tended to be higher in the chalky endosperm on 15 DAA and 20 DAA. Terpenoid squalene was increased for both parts, and the chalky part was lower on 10 DAA but higher on 15 DAA.
The notched-belly mutant
Grain chalkiness is a complex trait, controlled by multiple genes and varying with growing environments. So far, omics approaches have been employed to explore genes or biochemical pathways responsible for chalkiness [2, 14, 21], and our knowledge concerning chalkiness has been enlarged. However, only few QTLs or genes have been identified and functionally analyzed, and the metabolic mechanisms underlying chalkiness are still imperfectly understood.
We obtained a notched-belly mutant with high ratio of white-belly grains. The notched line can be seen on 5 DAA, possibly due to the restriction of the hulls on caryopsis elongation (Fig. 1). The constriction by bending alters the morphology of the caryposis, resulting in different shape of the upper and lower part, as shown in Fig. 1. It is worth studying if this change in morphology should cause disruptions in starch metabolism, granule formation and thereby chalkiness. Future studies on the loading routes of sucrose along the vertical axis of the notched belly seed versus the non-notched parent line are indispensible for clarifying the mechanism of this particular mutant.
The notched-line divided the endosperm into two separate compartments, with the upper part being translucent whereas the bottom being mainly opaque on about 20 DAA. These two parts of grain develop to mature independently, as they are separately supplied with assimilates from source organs through vascular bundles on the dorsal side. Thus the formation of the notched-line makes possible the chasing of the process of chalky tissue formation. Using the upper part as control, we compared the changes in metabolic profiles between the translucent and chalky part. Intriguingly, this comparison was made within the same genetic background, and is different from those performed either between varieties or between chalky grains and perfect grains for a given variety. Therefore, the findings of biochemical pathways such as carbohydrate metabolism, S-containing amino acid metabolism, and ROS scavenging are of relevance to chalkiness formation.
It is obvious that the position within grains contributes to difference in the occurrence of chalkiness between the two parts. Because of the notched-line, the upper endosperm is entirely isolated from the embryo, and thus is hardly affected by it. On the contrary, the bottom part is attached to the embryo, and is subjected to be influenced by the embryo. Thus the difference between the upper and lower parts was mainly associated with the compound effect of the chalkiness and the location within the grain (the effect of the embryo), as also discussed in our previous works on matured grains [14, 22]. Because it is impossible to recognize perfect grains or chalky grains at early stage, the compound effect of chalkiness and embryo can not be dissected at present time. Therefore the findings of this study have limitations when explaining the mechanisms underlying grain chalkiness, and the role of embryo in the formation of grain chalkiness needs to be thorough examined.
Sensing and regulating C and N status
Starch and protein, the products of C and N metabolism, are major factors affecting rice grain quality. Grain chalkiness is the result of poor filling of starch granules or imbalances in the subtle adjustment to starch degradation pathway triggered under stress . In addition, decrease in biosynthesis of proteins also plays crucial roles, as supported by the characterization of the recently cloned Chalk5  and the microscopic observation of chalky tissue . C and N metabolism are therefore the key pathways toward chalkiness formation, and a comprehensive understanding of them is central to the study on grain chalkiness. However, the mechanism by which plants sense signals relating to C and N status and subsequently regulating them still remains open.
In the present study, we detected several key metabolites involved in sensing and regulating carbon and nitrogen status. The chalky part contained lower sucrose content, as was accompanied by decrease in Glu and Gln, while Asp and Asn, carrying more N atoms per C, were increased on 15 DAA and 5 DAA, respectively. The inverse changing patterns of Glu/Gln and Asp/Asn indicate that the C skeletons for biosynthesis of amino acids may be limited in the chalky part so that it uses a more economical compound to transport N. Supporting the argument above, Arg, with a high N:C ratio (4:6) and serving as a main N storage compound, was up-regulated in the chalky endosperm. This was partially due to higher levels of its precursors, N-acetylglutamate and citulline. In addition, Arg plays a major metabolic role in seed maturation and germination, phloem and xylem transport, and accumulates under stress conditions .
It has been suggested that GABA, a four-C nonprotein amino acid, constitutes a readily accessible nontoxic reserve of C and N for amino acid metabolism and tricarboxylic acid (TCA) cycle activity, which is of particular relevance during stress events. In Arabidopsis, GABA shunt metabolic pathway that converts Glu to GABA has been shown to have a profound effect on the C-N balance of the seed . Our results showed that chalky endosperm contained significantly lower GABA during seed development, and its role in chalkiness formation merits further investigation.
Pathways involved in chalkiness formation
S-containing amino acids
The accumulation of osmolytes is the common strategy adopted by plants to combat the environmental stresses. Among them, betaine has strong osmoprotectant properties and confers tolerance to salinity and drought . Trehalose has also been proven useful to improve abiotic stress tolerance . The potential role of raffinose in stress tolerance has been intensively studied in seeds, particularly with respect to desiccation tolerance and longevity in the dehydrated state . Three polyamines namely Put, Spm, and Spd are considered as involved in variety of physiological response to abiotic stresses like salt and drought . High temperature during grain filing stage causes high occurrence of chalkiness, and recent studies indicated a link between osmoregualation and chalkiness formation under heat stress. Wada et al.  assessed the effects of dry wind on chalky ring formation in a growth chamber. They concluded that osmotic adjustment plays a crucial role in the down-regulating of starch biosynthesis, which leads to chalky ring formation under short-term hot and dry wind conditions. In addition, magnetic resonance imaging of the early stage caryopses in the high-temperature condition revealed lower water content around the centre of the endosperm, where the chalky tissues occurs . Osmolytes varied significantly between the chalky part and translucent part in this study. Although Pro was reduced, most of the osmolytes identified including betaines (glycine betaine, trigonelline, and stachydrine), trehalose, raffinose, and PAs were all increased in the chalky endosperm, suggesting that osmoregulation be triggered in the chalky endosperm.
Cell wall synthesis
Plant cells are surrounded by a cell wall that plays a key role in plant growth, structural integrity, and defense. Liu et al.  showed that down-regulation of genes responsible for cell wall synthesis (two cellulose synthase genes) and up-regulation of genes for cell wall hydrolysis (α-L-arabinofuranosidase and α-D-xylosidase) were linked with chalkiness occurrence. By contrast, previous proteomic survey of Lin et al.  demonstrated that five enzymes responsible for cell wall synthesis, including rhamnose biosynthetic enzyme, UDP-glucose 4-epimerase, UDP-glucose 6-dehydrogenase, UDP-arabinopyranose mutase, and UDP-glucuronic acid decarboxylase, were all up-regulated in the chalky endosperm of grains with white-belly. The cell wall is a complex and diverse structure that is mainly composed of polysaccharides. In the current study, we identified five precursors of cell wall (rhamnose, arabinose, glucose, gluconate, and xylose) that changed markedly between the chalky part and translucent part. Notably, they showed similar decreasing trend in the chalky endosperm, although arabinose did not varied significantly. Altogether, these studies indicate the relation of the cell wall synthesis with chalkiness formation.
Mineral ion homeostasis
Plants acquire minerals for sustaining growth and development. We identified several metabolites related to translocation and storage of minerals such as phytic acid, nicotianamine, and phytosphingolipids, which changed significantly between the chalky and translucent endosperms. Phytic acid is a nearly ubiquitous component of plant seeds and is usually the most abundant form of phosphate . It has negatively charged phosphates, making it an efficient chelating agent of mineral cations like K, Mg, Fe, Ca and Zn. Nicotianamine is responsible for the translocation of Fe and Mn from leaf to the developing seeds . The sphingolipids are a diverse class of membrane lipids that constitute up to 20 to 30% of the tonoplast and plasma membrane lipid in plants . Although their roles are not well defined, recent studies indicate that they have critical functions such as playing an important role in mineral ion homoeostasis . Phytic acid and its precursors were higher in the chalky endosperm, and this was accompanied by a simultaneous increase in raffinose. In soybean, a single, recessive mutation also confers a synchronous change of phytic acid and raffinose, which may be due to the fact that they share at least myo-inositol-1-phosphate and possibly free myo-inositol as a common intermediate . In addition, nicotianamine and phytosphingosine were both down-regluated in the chalky endosperm. The results indicate different ability of mobilizing or storing the minerals between the chalky and translucent endosperm, which needs verification by analysis of the mineral composition.
In summary, the notched-belly mutant is applicable to explore the physiological and molecular foundation of grain chalkiness, for it provides a high efficient comparison under the same genetic background that can minimize the influence of genotype and environment. Using this mutant, we observed a reduction in activity of C and N metabolism used for starch, protein, and cell wall synthesis in the chalky endosperm, which coincided with the disturbance of other pathways like osmoregulation, ROS scavenging, and ion homeostasis, implying a stressful condition (oxidative or osmostic) in the chalky part. It is still uncertain what triggers this metabolic shift. In addition, we cannot rule out the possibility that the differences in metabolite levels between the two sections are caused by some constriction between them due to the notched line; however, there is currently no evidence that this is the case. Clearly the development of chalkiness is a complex trait and this is corroborated by the metabolomic analysis in this study. An integrated approach using a combination of omics platforms and an examination of the physiology of the grain will be needed to understand this phenomenon fully.
Wuyujing3 had become a popular japonica rice cultivar in southern China with high eating quality since it was released in 1992 . It is an inbred cultivar and not a hydrid or transgenetic cultivar. The seeds of Wuyujing3 were provided by the breeder of Wujin Academy of Agricultural Sciences, Jiangsu China. In 2007, about 2 kg seeds of Wuyujing3 were treated by 0.5% ethyl methane sulfonate, and then were sown in paddy field and mixed harvested as a mutant population. A notched-belly mutant with white-belly (DY1102) was identified from the M2 population . It had high ratio of notched-belly grains, ranging from 66.58% (greenhouse) to 75.67% (ambient temperature) in 2014. Further, it had high occurrence of white-belly mainly on the bottom part, being 85.12% under greenhouse and 93.13% under high night temperature treatments (Table 1). Notably, grains on the middle primary rachis has higher ratio of white-belly, and were sampled for metabolic analysis.
In 2014, five plants of DY1102 were hand transplanted in a plastic pot filled with 15 kg clay soil, 30 cm in height and 34 cm in diameter. The soil contained 0.83 g/kg total N, 10.72 mg/kg available P, and 69.15 mg/kg exchangeable K. The basal fertilizers before transplanting were 1.0 g N, 1.2 g P2O5, and 0.9 g K2O/pot, and those for topdressing at the panicle initiation stage were 1.0 g N, 0.6 g P2O5, and 0.9 g K2O/pot. All the plants were grown under natural environment condition at seedling stage. Approximately 2–3 days before anthesis, 45 pots were transferred to a 28/23 °C plant chamber (12 h day–12 h night cycles; Model FYS-10; Hengyu, Nanjing, China) until maturity. This plant chamber has the light intensity above 75% of ambient light condition, and the relative humidity was maintained 70 ± 5% by a humidity regulator . Flowering dates of caryopsis on middle primary rachis were marked. Sampling time points were 5, 10, 15, and 20 days after antheisis (DAA), conducted at 9:00 am. Grains were frozen by liquid nitrogen and then stored at −80 °C until analysis. Optical photos of developing grains were taken by stereomicroscope (Stemi 2000-C, Zeiss Jena GmbH, Germany) with digital camera (PC1145, Canon, Japan). For metabolomic analysis, the developing grains were dehusked, and the embryos were removed. Subsequently, the remaining endosperms were cut into upper tranlucent parts (Tr) and bottom chalky parts (Ch) along the notched line. Each sample was measured with six biological replicates.
Metabolite profiling by multiple analytical platforms
Metabolite profiling was performed by SJTU-Metabolon Joint Metabolomics Laboratory (http://www.metabolon.com/) by the composite platforms: UPLC/MS/MS positive ion mode, UPLC/MS/MS negative ion mode, and GC/MS . Briefly, the upper and lower parts of the endosperm were grounded into fine powders under liquid nitrogen condition. Metabolites in lyophilized powder (about 40 mg) were extracted using 400 μL of methanol with internal standards. The resulting extraction solution was divided into 3 parts and then freeze-dried. UPLC-MS/MS analysis included both positive and negative ion modes, the products of which were redissolved and then carried out on Waters Acquity UPLC (Milford, USA) coupled with a Thermo LTQ XL MS (Thermo Inc., Massachusetts, USA). For GC-MS analysis, samples were derivatized by bistrimethyl-silyltrifluoroacetamide (BSTFA) and analyzed on Thermo Ultra GC-ISQ platform equipped by an electron impact ionization system (Thermo Inc., Massachusetts, USA).
Data analysis and construction of the metabolic atlas
Raw data of three analytical platforms were extracted, peak-identified and quality control processed using Metabolon Laboratory Information Management System. Metabolites were identified by automated comparison to reference library that was created using approximately 1500 authentic standards in Metabolon. Metabolites quantification, data normalization, and significant changed metabolites analysis were performed according to Lawton et al.  and Rao et al. . Briefly, a data normalization step was conducted to correct variations due to instrument inter-day tuning differences. Metabolites quantitative values were derived from integrated raw detector counts of the mass spectrometers. To preserve all of the variation, all compounds of widely different raw data were normalized directly on a similar graphical scale, with the normalized intensities scaled by their median values for each compound (Additional file 1). Significance tests were performed by Welch’s two-sample t-test using the programs R. Metabolic pathways were determined by reference to databases Plant Metabolic Network (PMN; http://www.plantcyc.org/) and Kyoto Encyclopedia of Genes and Genomes pathway (KEGG; http://www.genome.jp/kegg/pathway.html) to construct the metabolic atlas.
Cofactors, prosthetic groups, electron carriers
Day after anthesis
Reactive oxygen species
We thank Fang Cheng and Sheng Quan from SJTU-Metabolom Joint Metabolomics Laboratory, Shanghai Jiao Tong University for their work in the UPLC/MS/MS and GC/MS analysis.
This work was supported in part by grants from the National Natural Science Foundation of China (31171485 and 31470086), the National High Technology Research and Development Program of China (2014AA10A605), and the National Science and Technology Supporting Program of China (2012BAD04B08 and 2013BAD07B09).
Availability of data and materials
The data sets supporting the conclusions of this article are included within the article and its additional files. Of note that parameters of MS/MS spectra data process are commercial secrets of Metabolon (http://www.metabolon.com/); however, the raw data of MS/MS spectra (about 1 G) is available by contacting with the correspondent author (firstname.lastname@example.org).
Z Lin, XZ, and Z Liu conceived and designed the experiments; ZW, YJ, AD, GL, and SW performed part of the experiments; Z Lin, YD, and Z Liu analyzed the data and prepared the manuscript. All authors have read and approved the final manuscript.
All the authors declare that they have no conflict of interest.
Consent for publication
Ethics approval and consent to participate
Open AccessThis article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.
- Xi M, Lin ZM, Zhang XC, Liu ZH, Li GH, Wang QS, Wang SH, Ding YF. Endosperm structure of white-belly and white-core rice grains shown by scanning electron microscopy. Plant Prod Sci. 2014;17(4):285–90.View ArticleGoogle Scholar
- Sreenivasulu N, Butardo Jr VM, Misra G, Cuevas RP, Anacleto R, Kavi Kishor PB. Designing climate-resilient rice with ideal grain quality suited for high-temperature stress. J Exp Bot. 2015;66(7):1737–48.View ArticlePubMedPubMed CentralGoogle Scholar
- Fitzgerald MA, McCouch SR, Hall RD. Not just a grain of rice: the quest for quality. Trends Plant Sci. 2009;14(3):133–9.View ArticlePubMedGoogle Scholar
- Zhou L, Liang S, Ponce K, Marundon S, Ye G, Zhao X. Factors affecting head rice yield and chalkiness in indica rice. Field Crops Res. 2015;172:1–10.View ArticleGoogle Scholar
- Sun W, Zhou Q, Yao Y, Qiu X, Xie K, Yu S. Identification of genomic regions and the isoamylase gene for reduced grain chalkiness in rice. PLoS One. 2015;10(3):e0122013.View ArticlePubMedPubMed CentralGoogle Scholar
- Li Y, Fan C, Xing Y, Yun P, Luo L, Yan B, Peng B, Xie W, Wang G, Li X, Xiao J, Xu C, He Y. Chalk5 encodes a vacuolar H(+)-translocating pyrophosphatase influencing grain chalkiness in rice. Nat Genet. 2014;46(4):398–404.View ArticlePubMedGoogle Scholar
- Qu G, Quan S, Mondol P, Xu J, Zhang D, Shi J. Comparative metabolomic analysis of wild type and mads3 mutant rice anthers. J Integr Plant Biol. 2014;56(9):849–63.View ArticlePubMedGoogle Scholar
- Oikawa A, Matsuda F, Kusano M, Okazaki Y, Saito K. Rice metabolomics. Rice. 2008;1(1):63–71.View ArticleGoogle Scholar
- Chen W, Gong L, Guo Z, Wang W, Zhang H, Liu X, Yu S, Xiong L, Luo J. A novel integrated method for large-scale detection, identification, and quantification of widely targeted metabolites: application in the study of rice metabolomics. Mol Plant. 2013;6(6):1769–80.View ArticlePubMedGoogle Scholar
- Hu C, Shi J, Quan S, Cui B, Kleessen S, Nikoloski Z, Tohge T, Alexander D, Guo L, Lin H, Wang J, Cui X, Rao J, Luo Q, Zhao X, Fernie AR, Zhang D. Metabolic variation between japonica and indica rice cultivars as revealed by non-targeted metabolomics. Sci Rep. 2014;4:5067.PubMedGoogle Scholar
- Yamakawa H, Hakata M. Atlas of rice grain filling-related metabolism under high temperature: joint analysis of metabolome and transcriptome demonstrated inhibition of starch accumulation and induction of amino acid accumulation. Plant Cell Physiol. 2010;51(5):795–809.View ArticlePubMedPubMed CentralGoogle Scholar
- Kim JK, Park SY, Lim SH, Yeo Y, Cho HS, Ha SH. Comparative metabolic profiling of pigmented rice (Oryza sativa L.) cultivars reveals primary metabolites are correlated with secondary metabolites. J Cereal Sci. 2013;57(1):14–20.View ArticleGoogle Scholar
- Heuberger AL, Lewis MR, Chen M-H, Brick MA, Leach JE, Ryan EP. Metabolomic and functional genomic analyses reveal varietal differences in bioactive compounds of cooked rice. PLoS One. 2010;5(9):e12915.View ArticlePubMedPubMed CentralGoogle Scholar
- Lin ZM, Zhang XC, Yang XY, Li GH, Tang S, Wang SH, Ding YF, Liu ZH. Proteomic analysis of proteins related to rice grain chalkiness using iTRAQ and a novel comparison system based on a notched-belly mutant with white-belly. BMC Plant Biol. 2014;14:163.View ArticlePubMedPubMed CentralGoogle Scholar
- Wang Z, Gu Y, Hirasawa T, Ookawa T, Yanahara S. Comparison of caryopsis development between two rice varieties with remarkable difference in grain weights. Acta Bot Sin. 2004;46(6):698–710.Google Scholar
- Rao J, Cheng F, Hu C, Quan S, Lin H, Wang J, Chen G, Zhao X, Alexander D, Guo L, Wang G, Lai J, Zhang D, Shi J. Metabolic map of mature maize kernels. Metabolomics. 2014;10(5):775–87.View ArticleGoogle Scholar
- Hoshikawa K. Anthesis, fertilization and development of caryopsis. In: Matsuo T, Hoshikawa K, editors. Science of the rice plant, vol 1, morphology. Food and Agriculture Policy Research Center, Tokyo. 1993. p. 339–76.Google Scholar
- Parker D, Beckmann M, Zubair H, Enot DP, Caracuel-Rios Z, Overy DP, Snowdon S, Talbot NJ, Draper J. Metabolomic analysis reveals a common pattern of metabolic re-programming during invasion of three host plant species by Magnaporthe grisea. Plant J. 2009;59(5):723–37.View ArticlePubMedGoogle Scholar
- Orcaray L, Zulet A, Zabalza A, Royuela M. Impairment of carbon metabolism induced by the herbicide glyphosate. J Plant Physiol. 2012;169(1):27–33.View ArticlePubMedGoogle Scholar
- Zeeman S. Carbohydrate metabolism. In: Buchanan B, Wilhelm Gruissem W, Jones R, editors. Biochemistry and molecular biology of plants. 2nd ed. Hoboken: John Wiley & Sons; 2015. p. 567–609.Google Scholar
- Liu X, Guo T, Wan X, Wang H, Zhu M, Li A, Su N, Shen Y, Mao B, Zhai H, Mao L, Wan J. Transcriptome analysis of grain-filling caryopses reveals involvement of multiple regulatory pathways in chalky grain formation in rice. BMC Genomics. 2010;11(1):730.View ArticlePubMedPubMed CentralGoogle Scholar
- Lin Z, Zheng D, Zhang X, Wang Z, Lei J, Liu Z, Li G, Wang S, Ding Y. Chalky part differs in chemical composition from translucent part of japonica rice grains as revealed by a notched-belly mutant with white-belly. J Sci Food Agr. 2016;96:3937–43.View ArticleGoogle Scholar
- Lea PJ, Sodek L, Parry MA, Shewry PR, Halford NG. Asparagine in plants. Ann Appl Biol. 2007;150(1):1–26.View ArticleGoogle Scholar
- Renault H, El Amrani A, Palanivelu R, Updegraff EP, Yu A, Renou J-P, Preuss D, Bouchereau A, Deleu C. GABA accumulation causes cell elongation defects and a decrease in expression of genes encoding secreted and cell wall-related proteins in Arabidopsis thaliana. Plant Cell Physiol. 2011;52(5):894–908.View ArticlePubMedPubMed CentralGoogle Scholar
- Hell R, Wirtz M. Molecular biology, biochemistry and cellular physiology of cysteine metabolism in Arabidopsis thaliana. The Arabidopsis Book. 2011;9:e0154.View ArticlePubMedPubMed CentralGoogle Scholar
- Waduwara-Jayabahu I, Oppermann Y, Wirtz M, Hull ZT, Schoor S, Plotnikov AN, Hell R, Sauter M, Moffatt BA. Recycling of methylthioadenosine is essential for normal vascular development and reproduction in Arabidopsis. Plant Physiol. 2012;158(4):1728–44.View ArticlePubMedPubMed CentralGoogle Scholar
- Ravanel S, Gakière B, Job D, Douce R. The specific features of methionine biosynthesis and metabolism in plants. Proc Natl Acad Sci U S A. 1998;95(13):7805–12.View ArticlePubMedPubMed CentralGoogle Scholar
- Tan Q, Zhang L, Grant J, Cooper P, Tegeder M. Increased phloem transport of S-methylmethionine positively affects sulfur and nitrogen metabolism and seed development in pea plants. Plant Physiol. 2010;154(4):1886–96.View ArticlePubMedPubMed CentralGoogle Scholar
- Szarka A, Tomasskovics B, Bánhegyi G. The ascorbate-glutathione-α-tocopherol triad in abiotic stress response. Int J Mol Sci. 2012;13(4):4458–83.View ArticlePubMedPubMed CentralGoogle Scholar
- Voll LM, Abbasi A-R. Are there specific in vivo roles for α-and γ-tocopherol in plants? Plant Signal Behav. 2007;2(6):486–8.View ArticlePubMedPubMed CentralGoogle Scholar
- Giri J. Glycinebetaine and abiotic stress tolerance in plants. Plant Signal Behav. 2011;6(11):1746–51.View ArticlePubMedPubMed CentralGoogle Scholar
- Crowe JH. Trehalose as a “chemical chaperone”: fact and fantasy. Adv Exp Med Biol. 2007;594:143–58.View ArticlePubMedGoogle Scholar
- Hitz WD, Carlson TJ, Kerr PS, Sebastian SA. Biochemical and molecular characterization of a mutation that confers a decreased raffinosaccharide and phytic acid phenotype on soybean seeds. Plant Physiol. 2002;128(2):650–60.View ArticlePubMedPubMed CentralGoogle Scholar
- Takahashi T, Kakehi J-I. Polyamines: ubiquitous polycations with unique roles in growth and stress responses. Ann Bot. 2010;105(1):1–6.View ArticlePubMedGoogle Scholar
- Wada H, Masumoto-Kubo C, Gholipour Y, Nonami H, Tanaka F, Erra-Balsells R, Tsutsumi K, Hiraoka K, Morita S. Rice chalky ring formation caused by temporal reduction in starch biosynthesis during osmotic adjustment under foehn-induced dry wind. PLoS One. 2014;9(10):e110374.View ArticlePubMedPubMed CentralGoogle Scholar
- Ishimaru T, Horigane AK, Ida M, Iwasawa N, San-oh YA, Nakazono M, Nishizawa NK, Masumura T, Kondo M, Yoshida M. Formation of grain chalkiness and changes in water distribution in developing rice caryopses grown under high-temperature stress. J Cereal Sci. 2009;50(2):166–74.View ArticleGoogle Scholar
- Raboy V. Approaches and challenges to engineering seed phytate and total phosphorus. Plant Sci. 2009;177(4):281–96.View ArticleGoogle Scholar
- Schuler M, Rellán-Álvarez R, Fink-Straube C, Abadía J, Bauer P. Nicotianamine functions in the phloem-based transport of iron to sink organs, in pollen development and pollen tube growth in Arabidopsis. Plant Cell. 2012;24(6):2380–400.View ArticlePubMedPubMed CentralGoogle Scholar
- Lynch DV, Dunn TM. An introduction to plant sphingolipids and a review of recent advances in understanding their metabolism and function. New Phytol. 2004;161(3):677–702.View ArticleGoogle Scholar
- Chao DY, Gable K, Chen M, Baxter I, Dietrich CR, Cahoon EB, Guerinot ML, Lahner B, Lü S, Markham JE. Sphingolipids in the root play an important role in regulating the leaf ionome in Arabidopsis thaliana. Plant Cell. 2011;23(3):1061–81.View ArticlePubMedPubMed CentralGoogle Scholar
- Yang XY, Lin ZM, Liu ZH, Alim MA, Bi JG, Li GH, Wang QS, Wang SH, Ding YF. Physicochemical and sensory properties of japonica rice varied with production areas in China. J Integr Agr. 2013;12(10):1748–56.View ArticleGoogle Scholar
- Shi P, Tang L, Lin C, Liu L, Wang H, Cao W, Zhu Y. Modeling the effects of post-anthesis heat stress on rice phenology. Field Crops Res. 2015;177:26–36.View ArticleGoogle Scholar
- Lawton KA, Berger A, Mitchell M, Milgram KE, Evans AM, Guo L, Hanson RW, Kalhan SC, Ryals JA, Milburn MV. Analysis of the adult human plasma metabolome. Pharmacogenomics. 2008;9(4):383–97.View ArticlePubMedGoogle Scholar