- Split View
-
Views
-
Cite
Cite
Takashi Ohdan, Perigio B. Francisco, Takayuki Sawada, Tatsuro Hirose, Tomio Terao, Hikaru Satoh, Yasunori Nakamura, Expression profiling of genes involved in starch synthesis in sink and source organs of rice, Journal of Experimental Botany, Volume 56, Issue 422, December 2005, Pages 3229–3244, https://doi.org/10.1093/jxb/eri292
- Share Icon Share
Abstract
A comprehensive analysis of the transcript levels of genes which encode starch-synthesis enzymes is fundamental for the assessment of the function of each enzyme and the regulatory mechanism for starch biosynthesis in source and sink organs. Using quantitative real-time RT-PCR, an examination was made of the expression profiles of 27 rice genes encoding six classes of enzymes, i.e. ADPglucose pyrophosphorylase (AGPase), starch synthase, starch branching enzyme, starch debranching enzyme, starch phosphorylase, and disproportionating enzyme in developing seeds and leaves. The modes of gene expression were tissue- and developmental stage-specific. Four patterns of expression in the seed were identified: group 1 genes, which are expressed very early in grain formation and are presumed to be involved in the construction of fundamental cell machineries, de novo synthesis of glucan primers, and initiation of starch granules; group 2 genes, which are highly expressed throughout endosperm development; group 3 genes, which have transcripts that are low at the onset but which rise steeply at the start of starch synthesis in the endosperm and are thought to play essential roles in endosperm starch synthesis; and group 4 genes, which are expressed scantly, mainly at the onset of grain development, and might be involved in synthesis of starch in the pericarp. The methodology also revealed that the defect in the cytosolic AGPase small subunit2b (AGPS2b) transcription from the AGPS2 gene in endosperm sharply enhanced the expressions of endosperm and leaf plastidial AGPS1, the endosperm cytosolic AGPase large subunit2 (AGPL2), and the leaf plastidial AGPL1.
Introduction
Starch is the end-product of photosynthesis in source tissues and is stored as energy reserves in sink tissues. Starch has two major components, the basically linear α-polyglucan amylose and the branched α-polyglucan amylopectin. The α-1,4-glucosidic link chains of both amylose and amylopectin are elongated by the addition of the glucose moiety from ADPglucose, which is synthesized by ADPglucose pyrophosphorylase (AGPase) from glucose-1-P, to the non-reducing end of the α-glucan acceptor molecule. The elongation reactions for the α-1,4-chains of amylose and amylopectin are distinctively catalysed by a starch granule-bound form of starch synthase (GBSS) and a soluble form of starch synthase (SS), respectively. Amylopectin has a much more defined structure called ‘tandem-cluster structure’ than glycogen, because it is composed of tandem-linked clusters (approximately 9–10 nm each in length) where linear α-1,4-glucan chains are regularly branched via α-1,6-glucosidic linkages, as opposed to the glycogens of bacteria and animals which have a more randomly branched structure (Thompson, 2000). The distinct structure of amylopectin contributes to the crystalline organization of the starch granule (Gallant et al., 1997). Variation of the cluster fine structure is considered to cause variations in starch functional properties between species, tissues, and genetic backgrounds.
Amylose is synthesized by AGPase and GBSS while amylopectin is synthesized by the co-ordinated actions of AGPase, SS, starch branching enzyme (BE), and starch debranching enzyme (DBE). Disproportionating enzyme (DPE) and phosphorylase (PHO) are generally considered to be involved in starch degradation, but recent studies suggest that both DPE and PHO may play some parts in starch biosynthesis, although the precise mechanisms of their roles are unclear (Colleoni et al., 1999; Ball and Morell, 2003; Tetlow et al., 2004). A greater complexity in plant starch biosynthesis is that, although the α-1,4- and α-1,6-glucosidic linkages of glycogen in bacteria and animals can possibly be synthesized by a single form of glycogen synthase and glycogen branching enzyme, those of amylopectin are formed by multiple types of SS (SSI, SSII, SSIII, and SSIV), BE (BEI and BEII), and DBE [isoamylase (ISA) and pullulanase (PUL)]. In addition, several types of these enzymes have multiple isoforms, the number of which is plant species specific. Of these examined so far, each isoform of SS, BE, and DBE plays a distinct role in amylopectin biosynthesis (Nakamura, 2002). Therefore, analysis of the expression patterns of individual genes is important to understand the features of tissue-specific and developmental stage-specific starch biosynthesis.
Metabolic pathways involved in starch biosynthesis are different between source and sink tissues. In addition to the structural difference in amylopectin and starch granules, the carbon of starch is derived from fructose-6-P in the Calvin–Benson cycle in photosynthetic tissues, while in sink tissues it is derived from sucrose, which is translocated from source tissues through the phloem. Therefore, sucrose must be metabolized to sugar-P or ADPglucose in the cytoplasm, and either or both of these compounds are translocated into the amyloplasts via the compound-specific hexose monophosphate translocator or the ADGglucose translocator, respectively (James et al., 2003).
The completed genome sequences of the model plants Arabidopsis and rice make it possible to predict the total number of rice genes involved in starch biosynthesis. Recently, Smith et al. (2004) analysed the diurnal changes in the transcriptome of genes encoding the starch-metabolizing enzyme in the leaves of the dicot Arabidopsis to reveal both transcriptional and post-transcriptional regulation of starch metabolism. However, their analyses were confined to leaves. As far as is known, no similar study has been performed in monocots such as cereals, although transcript levels of some classes of enzyme in rice plants such as SS (Hirose and Terao, 2004; Jiang et al., 2004; Dian et al., 2005; Suzuki et al., 2005) and BE genes (Mizuno et al., 2001; Suzuki et al., 2005) have been reported.
In the present study, changes in transcripts of all the known genes encoding six classes of enzymes presumably involved in starch synthesis, including AGPase, SS plus GBSS, BE, DBE, PHO, and DPE in developing rice endosperm were measured by quantitative real-time PCR, and the results were compared with their transcript levels in the leaves. The present observations revealed that the pattern of gene expression is gene specific, depending on each gene of each class during endosperm development, and that the genes which are highly expressed in the endosperm are quite different from genes that are markedly expressed in the leaf. The dramatic effects of the rice shrunken mutation defect in the cytosolic AGPase small subunit2b (AGPS2b) transcription from the AGPS2 gene on the expressions of other AGPase genes are also reported.
Materials and methods
Plant material
Rice plants (Oryza sativa ssp. japonica cv. Nipponbare) were grown in the field in plastic pots filled with paddy field soil. After heading, rice plants were transferred to a greenhouse. For seed sampling, spikelets were marked on the flowering day and harvested at 1, 2, 3, 5, 7, 10, 15, and 20 d after flowering (DAF). For leaf analysis, rice plants (cv. Nipponbare) were grown in a paddy field and 1 month thereafter, three leaf blades were taken from individual plants: the uppermost expanding leaf (young) and the two below (the upper one considered as a fully developed leaf and the lower one as an old leaf). Once severed from the plant, all seed and leaf samples were immediately frozen in liquid nitrogen and stored at −80 °C until use.
For mutant analysis, Oryza sativa ssp. japonica cv. Kinmaze (wild type) and its shrunken2 mutant EM22 were grown in a paddy field. For total RNA extractions, the fully developed leaf and developing seeds were harvested when heads hung down, immediately frozen in liquid nitrogen, and stored at −80 °C until use.
RNA extraction and cDNA preparation
Total RNAs were isolated from seed endosperms using the method of Chang et al. (1993). First-strand cDNA was synthesized with 5 μg total RNA using SuperScript II RNase H− Reverse Transcriptase (Invitrogen) and oligo-dT13 primer.
Total RNA from leaves of cv. Nipponbare, and seeds and leaves of cv. Kinmaze and its shrunken mutant EM22 were extracted with the RNeasy Plant Mini Kit (Qiagen) according to the manufacturer's instructions. The extracts were treated with RNase-free DNase (Qiagen) to completely remove contaminating genomic DNA. Two micrograms of leaf or seed total RNA were used for first-strand cDNA synthesis with the iScript™ cDNA Synthesis Kit (Bio-Rad).
Quantitative real-time RT-PCR
Aliquots of the first-strand cDNA mixtures corresponding to 5 ng of total RNA served as the templates for quantitative real-time RT-PCR analysis with the Quantitect™ SYBR Green PCR Kit (Qiagen). Reactions were carried out on an iCycler (Bio-Rad) according to the manufacturers' protocols. The gene-specific primers used for quantitative PCR are listed in Table 1. To optimize PCR conditions for each primer set, annealing temperature, PCR efficiency, and standard curve were examined. The specificity of the PCR amplification was checked with a melt curve analysis (from 55 °C to 94 °C) following the final cycle of the PCR. To verify the specificity of each primer set, their amplification products were cloned in the pGEM T-Easy vector (Promega) and sequenced in an ABI 3100 Automated DNA Sequencer (Applied Biosystems).
Enzyme . | Gene name . | EC no. . | Acc. no. . | Chr. . | Primer sequence . | Amplicon size (bp) . | Other name (reference) . |
---|---|---|---|---|---|---|---|
ADP-glucose pyrophosphorylase | OsAGPS1 | 2.7.7.27 | AK073146 | 9 | [F] GTGCCACTTAAAGGCACCATT | 97 | |
small subunit 1 | [R] CCCACATTTCAGACACGGTTT | ||||||
ADP-glucose pyrophosphorylase | OsAGPS2a | 2.7.7.27 | AK071826 | 8 | [F] ACTCCAAGAGCTCGCAGACC | 147 | |
small subunit 2a | [R] GCCTGTAGTTGGCACCCAGA | ||||||
ADP-glucose pyrophosphorylase | OsAGPS2b | 2.7.7.27 | AK103906 | 8 | [F] AACAATCGAAGCGCGAGAAA | 186 | Shrunken |
small subunit 2ba | [R] GCCTGTAGTTGGCACCCAGA | (Kawagoe et al., 2005) | |||||
ADP-glucose pyrophosphorylase | OsAGPL1 | 2.7.7.27 | D50317 | 5 | [F] GGAAGACGGATGATCGAGAAAG | 140 | |
large subunit 1 | [R] CACATGAGATGCACCAACGA | ||||||
ADP-glucose pyrophosphorylase | OsAGPL2 | 2.7.7.27 | U66041 | 1 | [F] AGTTCGATTCAAGACGGATAGC | 96 | |
large subunit 2a | [R] CGACTTCCACAGGCAGCTTATT | ||||||
ADP-glucose pyrophosphorylase | OsAGPL3 | 2.7.7.27 | AK069296 | 3 | [F] AAGCCAGCCATGACCATTTG | 131 | |
large subunit 3 | [R] CACACGGTAGATTCACGAGACAA | ||||||
ADP-glucose pyrophosphorylase | OsAGPL4 | 2.7.7.27 | AK121036 | 7 | [F] TCAACGTCGATGCAGCAAAT | 77 | |
large subunit 4 | [R] ATCCCTCAGTTCCTAGCCTCATT | ||||||
Starch synthase I | OsSSI | 2.4.1.21 | D16202 | 6 | [F] GGGCCTTCATGGATCAACC | 279 | |
[R] CCGCTTCAAGCATCCTCATC | |||||||
Starch synthase IIa | OsSSIIa | 2.4.1.21 | AF419099 | 6 | [F] GCTTCCGGTTTGTGTGTTCA | 54 | SSII-3 |
[R] CTTAATACTCCCTCAACTCCACCAT | (Hirose et al., 2004; Jiang et al., 2004) | ||||||
Starch synthase IIb | OsSSIIb | 2.4.1.21 | AF395537 | 2 | [F] TAGGAGCAACGGTGGAAGTGA | 89 | SSII-2 |
[R] GTGAACGTGAGTACGTGACCAAT | (Hirose et al., 2004; Jiang et al., 2004) | ||||||
Starch synthase IIc | OsSSIIc | 2.4.1.21 | AF383878 | 10 | [F] GACCGAAATGCCTTTTTCTCG | 256 | SSII-1 |
[R] GGGCTTGGAGCCTCTCCTTA | (Hirose et al., 2004) | ||||||
Starch synthase IIIa | OsSSIIIa | 2.4.1.21 | AY100469 | 8 | [F] GCCTGCCCTGGACTACATTG | 334 | SSIII-2 |
[R] GCAAACATATGTACACGGTTCTGG | (Hirose et al., 2004) | ||||||
Starch synthase IIIb | OsSSIIIb | 2.4.1.21 | AF432915 | 4 | [F] ATTCCGCTCGCAAGAACTGA | 224 | SSIII-1 |
[R] CAACCGCAGGATAACGGAAA | (Hirose et al., 2004) | ||||||
Starch synthase IVa | OsSSIVa | 2.4.1.21 | AY100470 | 1 | [F] GGGAGCGGCTCAAACATAAA | 237 | SSIV-1 |
[R] CCGTGCACTGACTGCAAAAT | (Hirose et al., 2004) | ||||||
Starch synthase IVb | OsSSIVb | 2.4.1.21 | AY373258 | 5 | [F] ATGCAGGAAGCCGAGATGTT | 75 | SSIV-2 |
[R] ACGACAATGGGTGCCAAGAT | (Hirose et al., 2004; Jiang et al., 2004) | ||||||
Granule-bound starch synthase I | OsGBSSI | 2.4.1.21 | X62134 | 6 | [F] AACGTGGCTGCTCCTTGAA | 218 | Waxy |
[R] TTGGCAATAAGCCACACACA | (Okagaki, 1992) | ||||||
Granule-bound starch synthase II | OsGBSSII | 2.4.1.21 | AY069940 | 7 | [F] AGGCATCGAGGGTGAGGAG | 246 | |
[R] CCATCTGGCCCACATCTCTA | |||||||
Starch branching enzyme I | OsBEI | 2.4.1.18 | D11082 | 6 | [F] TGGCCATGGAAGAGTTGGC | 191 | |
[R] CAGAAGCAACTGCTCCACC | |||||||
Starch branching enzyme IIa | OsBEIIa | 2.4.1.18 | AB023498 | 4? | [F] GCCAATGCCAGGAAGATGA | 128 | |
[R] GCGCAACATAGGATGGGTTT | |||||||
Starch branching enzyme IIb | OsBEIIb | 2.4.1.18 | D16201 | 2 | [F] ATGCTAGAGTTTGACCGC | 261 | Amylose-extender |
[R] AGTGTGATGGATCCTGCC | (Nishi et al., 2001) | ||||||
Starch debranching enzyme: | OsISA1 | 3.2.1.68 | AB093426 | 8 | [F] TGCTCAGCTACTCCTCCATCATC | 132 | Sugary-1 |
Isoamylase I | [R] AGGACCGCACAACTTCAACATA | (Kubo et al., 1999) | |||||
Starch debranching enzyme: | OsISA2 | 3.2.1.68 | AC132483 | 5 | [F] TAGAGGTCCTCTTGGAGG | 170 | |
Isoamylase II | [R] AATCAGCTTCTGAGTCACCG | ||||||
Starch debranching enzyme: | OsISA3 | 3.2.1.68 | AP005574 | 9 | [F] ACAGCTTGAGACACTGGGTTGAG | 100 | |
Isoamylase III | [R] GCATCAAGAGGACAACCATCTG | ||||||
Starch debranching enzyme: | OsPUL | 3.2.1.41 | AB012915 | 4 | [F] ACCTTTCTTCCATGCTGG | 202 | |
Pullulanase | [R] CAAAGGTCTGAAAGATGGG | ||||||
Starch phosphorylase L | OsPHOL | 2.4.1.1 | AK063766 | 3 | [F] TTGGCAGGAAGGTTTCGCT | 66 | |
[R] CGAAGCCTGAAGTGAACTTGCT | |||||||
Starch phosphorylase Ha | OsPHOH | 2.4.1.1 | AK103367 | 1 | [F] CACCAAGACGAAGCTCATCAAG | 126 | |
[R] TTCACTCGTTGCTGGGTTCTC | |||||||
Disproportionating enzyme I | OsDPE1 | 2.4.1.25 | AP004306 | 7 | [F] TCTGTCGCTTGAACTTACAGAA | 77 | |
[R] TTTGGCTGCAAGTATTGTTCT | |||||||
Disproportionating enzyme IIa | OsDPE2 | 2.4.1.25 | AK067082 | 7 | [F] CAAGTACACCACAAGACCAGCAA | 107 | |
[R] CGTCCAACAGCGAATCCAAT |
Enzyme . | Gene name . | EC no. . | Acc. no. . | Chr. . | Primer sequence . | Amplicon size (bp) . | Other name (reference) . |
---|---|---|---|---|---|---|---|
ADP-glucose pyrophosphorylase | OsAGPS1 | 2.7.7.27 | AK073146 | 9 | [F] GTGCCACTTAAAGGCACCATT | 97 | |
small subunit 1 | [R] CCCACATTTCAGACACGGTTT | ||||||
ADP-glucose pyrophosphorylase | OsAGPS2a | 2.7.7.27 | AK071826 | 8 | [F] ACTCCAAGAGCTCGCAGACC | 147 | |
small subunit 2a | [R] GCCTGTAGTTGGCACCCAGA | ||||||
ADP-glucose pyrophosphorylase | OsAGPS2b | 2.7.7.27 | AK103906 | 8 | [F] AACAATCGAAGCGCGAGAAA | 186 | Shrunken |
small subunit 2ba | [R] GCCTGTAGTTGGCACCCAGA | (Kawagoe et al., 2005) | |||||
ADP-glucose pyrophosphorylase | OsAGPL1 | 2.7.7.27 | D50317 | 5 | [F] GGAAGACGGATGATCGAGAAAG | 140 | |
large subunit 1 | [R] CACATGAGATGCACCAACGA | ||||||
ADP-glucose pyrophosphorylase | OsAGPL2 | 2.7.7.27 | U66041 | 1 | [F] AGTTCGATTCAAGACGGATAGC | 96 | |
large subunit 2a | [R] CGACTTCCACAGGCAGCTTATT | ||||||
ADP-glucose pyrophosphorylase | OsAGPL3 | 2.7.7.27 | AK069296 | 3 | [F] AAGCCAGCCATGACCATTTG | 131 | |
large subunit 3 | [R] CACACGGTAGATTCACGAGACAA | ||||||
ADP-glucose pyrophosphorylase | OsAGPL4 | 2.7.7.27 | AK121036 | 7 | [F] TCAACGTCGATGCAGCAAAT | 77 | |
large subunit 4 | [R] ATCCCTCAGTTCCTAGCCTCATT | ||||||
Starch synthase I | OsSSI | 2.4.1.21 | D16202 | 6 | [F] GGGCCTTCATGGATCAACC | 279 | |
[R] CCGCTTCAAGCATCCTCATC | |||||||
Starch synthase IIa | OsSSIIa | 2.4.1.21 | AF419099 | 6 | [F] GCTTCCGGTTTGTGTGTTCA | 54 | SSII-3 |
[R] CTTAATACTCCCTCAACTCCACCAT | (Hirose et al., 2004; Jiang et al., 2004) | ||||||
Starch synthase IIb | OsSSIIb | 2.4.1.21 | AF395537 | 2 | [F] TAGGAGCAACGGTGGAAGTGA | 89 | SSII-2 |
[R] GTGAACGTGAGTACGTGACCAAT | (Hirose et al., 2004; Jiang et al., 2004) | ||||||
Starch synthase IIc | OsSSIIc | 2.4.1.21 | AF383878 | 10 | [F] GACCGAAATGCCTTTTTCTCG | 256 | SSII-1 |
[R] GGGCTTGGAGCCTCTCCTTA | (Hirose et al., 2004) | ||||||
Starch synthase IIIa | OsSSIIIa | 2.4.1.21 | AY100469 | 8 | [F] GCCTGCCCTGGACTACATTG | 334 | SSIII-2 |
[R] GCAAACATATGTACACGGTTCTGG | (Hirose et al., 2004) | ||||||
Starch synthase IIIb | OsSSIIIb | 2.4.1.21 | AF432915 | 4 | [F] ATTCCGCTCGCAAGAACTGA | 224 | SSIII-1 |
[R] CAACCGCAGGATAACGGAAA | (Hirose et al., 2004) | ||||||
Starch synthase IVa | OsSSIVa | 2.4.1.21 | AY100470 | 1 | [F] GGGAGCGGCTCAAACATAAA | 237 | SSIV-1 |
[R] CCGTGCACTGACTGCAAAAT | (Hirose et al., 2004) | ||||||
Starch synthase IVb | OsSSIVb | 2.4.1.21 | AY373258 | 5 | [F] ATGCAGGAAGCCGAGATGTT | 75 | SSIV-2 |
[R] ACGACAATGGGTGCCAAGAT | (Hirose et al., 2004; Jiang et al., 2004) | ||||||
Granule-bound starch synthase I | OsGBSSI | 2.4.1.21 | X62134 | 6 | [F] AACGTGGCTGCTCCTTGAA | 218 | Waxy |
[R] TTGGCAATAAGCCACACACA | (Okagaki, 1992) | ||||||
Granule-bound starch synthase II | OsGBSSII | 2.4.1.21 | AY069940 | 7 | [F] AGGCATCGAGGGTGAGGAG | 246 | |
[R] CCATCTGGCCCACATCTCTA | |||||||
Starch branching enzyme I | OsBEI | 2.4.1.18 | D11082 | 6 | [F] TGGCCATGGAAGAGTTGGC | 191 | |
[R] CAGAAGCAACTGCTCCACC | |||||||
Starch branching enzyme IIa | OsBEIIa | 2.4.1.18 | AB023498 | 4? | [F] GCCAATGCCAGGAAGATGA | 128 | |
[R] GCGCAACATAGGATGGGTTT | |||||||
Starch branching enzyme IIb | OsBEIIb | 2.4.1.18 | D16201 | 2 | [F] ATGCTAGAGTTTGACCGC | 261 | Amylose-extender |
[R] AGTGTGATGGATCCTGCC | (Nishi et al., 2001) | ||||||
Starch debranching enzyme: | OsISA1 | 3.2.1.68 | AB093426 | 8 | [F] TGCTCAGCTACTCCTCCATCATC | 132 | Sugary-1 |
Isoamylase I | [R] AGGACCGCACAACTTCAACATA | (Kubo et al., 1999) | |||||
Starch debranching enzyme: | OsISA2 | 3.2.1.68 | AC132483 | 5 | [F] TAGAGGTCCTCTTGGAGG | 170 | |
Isoamylase II | [R] AATCAGCTTCTGAGTCACCG | ||||||
Starch debranching enzyme: | OsISA3 | 3.2.1.68 | AP005574 | 9 | [F] ACAGCTTGAGACACTGGGTTGAG | 100 | |
Isoamylase III | [R] GCATCAAGAGGACAACCATCTG | ||||||
Starch debranching enzyme: | OsPUL | 3.2.1.41 | AB012915 | 4 | [F] ACCTTTCTTCCATGCTGG | 202 | |
Pullulanase | [R] CAAAGGTCTGAAAGATGGG | ||||||
Starch phosphorylase L | OsPHOL | 2.4.1.1 | AK063766 | 3 | [F] TTGGCAGGAAGGTTTCGCT | 66 | |
[R] CGAAGCCTGAAGTGAACTTGCT | |||||||
Starch phosphorylase Ha | OsPHOH | 2.4.1.1 | AK103367 | 1 | [F] CACCAAGACGAAGCTCATCAAG | 126 | |
[R] TTCACTCGTTGCTGGGTTCTC | |||||||
Disproportionating enzyme I | OsDPE1 | 2.4.1.25 | AP004306 | 7 | [F] TCTGTCGCTTGAACTTACAGAA | 77 | |
[R] TTTGGCTGCAAGTATTGTTCT | |||||||
Disproportionating enzyme IIa | OsDPE2 | 2.4.1.25 | AK067082 | 7 | [F] CAAGTACACCACAAGACCAGCAA | 107 | |
[R] CGTCCAACAGCGAATCCAAT |
Cytosolic type.
Enzyme . | Gene name . | EC no. . | Acc. no. . | Chr. . | Primer sequence . | Amplicon size (bp) . | Other name (reference) . |
---|---|---|---|---|---|---|---|
ADP-glucose pyrophosphorylase | OsAGPS1 | 2.7.7.27 | AK073146 | 9 | [F] GTGCCACTTAAAGGCACCATT | 97 | |
small subunit 1 | [R] CCCACATTTCAGACACGGTTT | ||||||
ADP-glucose pyrophosphorylase | OsAGPS2a | 2.7.7.27 | AK071826 | 8 | [F] ACTCCAAGAGCTCGCAGACC | 147 | |
small subunit 2a | [R] GCCTGTAGTTGGCACCCAGA | ||||||
ADP-glucose pyrophosphorylase | OsAGPS2b | 2.7.7.27 | AK103906 | 8 | [F] AACAATCGAAGCGCGAGAAA | 186 | Shrunken |
small subunit 2ba | [R] GCCTGTAGTTGGCACCCAGA | (Kawagoe et al., 2005) | |||||
ADP-glucose pyrophosphorylase | OsAGPL1 | 2.7.7.27 | D50317 | 5 | [F] GGAAGACGGATGATCGAGAAAG | 140 | |
large subunit 1 | [R] CACATGAGATGCACCAACGA | ||||||
ADP-glucose pyrophosphorylase | OsAGPL2 | 2.7.7.27 | U66041 | 1 | [F] AGTTCGATTCAAGACGGATAGC | 96 | |
large subunit 2a | [R] CGACTTCCACAGGCAGCTTATT | ||||||
ADP-glucose pyrophosphorylase | OsAGPL3 | 2.7.7.27 | AK069296 | 3 | [F] AAGCCAGCCATGACCATTTG | 131 | |
large subunit 3 | [R] CACACGGTAGATTCACGAGACAA | ||||||
ADP-glucose pyrophosphorylase | OsAGPL4 | 2.7.7.27 | AK121036 | 7 | [F] TCAACGTCGATGCAGCAAAT | 77 | |
large subunit 4 | [R] ATCCCTCAGTTCCTAGCCTCATT | ||||||
Starch synthase I | OsSSI | 2.4.1.21 | D16202 | 6 | [F] GGGCCTTCATGGATCAACC | 279 | |
[R] CCGCTTCAAGCATCCTCATC | |||||||
Starch synthase IIa | OsSSIIa | 2.4.1.21 | AF419099 | 6 | [F] GCTTCCGGTTTGTGTGTTCA | 54 | SSII-3 |
[R] CTTAATACTCCCTCAACTCCACCAT | (Hirose et al., 2004; Jiang et al., 2004) | ||||||
Starch synthase IIb | OsSSIIb | 2.4.1.21 | AF395537 | 2 | [F] TAGGAGCAACGGTGGAAGTGA | 89 | SSII-2 |
[R] GTGAACGTGAGTACGTGACCAAT | (Hirose et al., 2004; Jiang et al., 2004) | ||||||
Starch synthase IIc | OsSSIIc | 2.4.1.21 | AF383878 | 10 | [F] GACCGAAATGCCTTTTTCTCG | 256 | SSII-1 |
[R] GGGCTTGGAGCCTCTCCTTA | (Hirose et al., 2004) | ||||||
Starch synthase IIIa | OsSSIIIa | 2.4.1.21 | AY100469 | 8 | [F] GCCTGCCCTGGACTACATTG | 334 | SSIII-2 |
[R] GCAAACATATGTACACGGTTCTGG | (Hirose et al., 2004) | ||||||
Starch synthase IIIb | OsSSIIIb | 2.4.1.21 | AF432915 | 4 | [F] ATTCCGCTCGCAAGAACTGA | 224 | SSIII-1 |
[R] CAACCGCAGGATAACGGAAA | (Hirose et al., 2004) | ||||||
Starch synthase IVa | OsSSIVa | 2.4.1.21 | AY100470 | 1 | [F] GGGAGCGGCTCAAACATAAA | 237 | SSIV-1 |
[R] CCGTGCACTGACTGCAAAAT | (Hirose et al., 2004) | ||||||
Starch synthase IVb | OsSSIVb | 2.4.1.21 | AY373258 | 5 | [F] ATGCAGGAAGCCGAGATGTT | 75 | SSIV-2 |
[R] ACGACAATGGGTGCCAAGAT | (Hirose et al., 2004; Jiang et al., 2004) | ||||||
Granule-bound starch synthase I | OsGBSSI | 2.4.1.21 | X62134 | 6 | [F] AACGTGGCTGCTCCTTGAA | 218 | Waxy |
[R] TTGGCAATAAGCCACACACA | (Okagaki, 1992) | ||||||
Granule-bound starch synthase II | OsGBSSII | 2.4.1.21 | AY069940 | 7 | [F] AGGCATCGAGGGTGAGGAG | 246 | |
[R] CCATCTGGCCCACATCTCTA | |||||||
Starch branching enzyme I | OsBEI | 2.4.1.18 | D11082 | 6 | [F] TGGCCATGGAAGAGTTGGC | 191 | |
[R] CAGAAGCAACTGCTCCACC | |||||||
Starch branching enzyme IIa | OsBEIIa | 2.4.1.18 | AB023498 | 4? | [F] GCCAATGCCAGGAAGATGA | 128 | |
[R] GCGCAACATAGGATGGGTTT | |||||||
Starch branching enzyme IIb | OsBEIIb | 2.4.1.18 | D16201 | 2 | [F] ATGCTAGAGTTTGACCGC | 261 | Amylose-extender |
[R] AGTGTGATGGATCCTGCC | (Nishi et al., 2001) | ||||||
Starch debranching enzyme: | OsISA1 | 3.2.1.68 | AB093426 | 8 | [F] TGCTCAGCTACTCCTCCATCATC | 132 | Sugary-1 |
Isoamylase I | [R] AGGACCGCACAACTTCAACATA | (Kubo et al., 1999) | |||||
Starch debranching enzyme: | OsISA2 | 3.2.1.68 | AC132483 | 5 | [F] TAGAGGTCCTCTTGGAGG | 170 | |
Isoamylase II | [R] AATCAGCTTCTGAGTCACCG | ||||||
Starch debranching enzyme: | OsISA3 | 3.2.1.68 | AP005574 | 9 | [F] ACAGCTTGAGACACTGGGTTGAG | 100 | |
Isoamylase III | [R] GCATCAAGAGGACAACCATCTG | ||||||
Starch debranching enzyme: | OsPUL | 3.2.1.41 | AB012915 | 4 | [F] ACCTTTCTTCCATGCTGG | 202 | |
Pullulanase | [R] CAAAGGTCTGAAAGATGGG | ||||||
Starch phosphorylase L | OsPHOL | 2.4.1.1 | AK063766 | 3 | [F] TTGGCAGGAAGGTTTCGCT | 66 | |
[R] CGAAGCCTGAAGTGAACTTGCT | |||||||
Starch phosphorylase Ha | OsPHOH | 2.4.1.1 | AK103367 | 1 | [F] CACCAAGACGAAGCTCATCAAG | 126 | |
[R] TTCACTCGTTGCTGGGTTCTC | |||||||
Disproportionating enzyme I | OsDPE1 | 2.4.1.25 | AP004306 | 7 | [F] TCTGTCGCTTGAACTTACAGAA | 77 | |
[R] TTTGGCTGCAAGTATTGTTCT | |||||||
Disproportionating enzyme IIa | OsDPE2 | 2.4.1.25 | AK067082 | 7 | [F] CAAGTACACCACAAGACCAGCAA | 107 | |
[R] CGTCCAACAGCGAATCCAAT |
Enzyme . | Gene name . | EC no. . | Acc. no. . | Chr. . | Primer sequence . | Amplicon size (bp) . | Other name (reference) . |
---|---|---|---|---|---|---|---|
ADP-glucose pyrophosphorylase | OsAGPS1 | 2.7.7.27 | AK073146 | 9 | [F] GTGCCACTTAAAGGCACCATT | 97 | |
small subunit 1 | [R] CCCACATTTCAGACACGGTTT | ||||||
ADP-glucose pyrophosphorylase | OsAGPS2a | 2.7.7.27 | AK071826 | 8 | [F] ACTCCAAGAGCTCGCAGACC | 147 | |
small subunit 2a | [R] GCCTGTAGTTGGCACCCAGA | ||||||
ADP-glucose pyrophosphorylase | OsAGPS2b | 2.7.7.27 | AK103906 | 8 | [F] AACAATCGAAGCGCGAGAAA | 186 | Shrunken |
small subunit 2ba | [R] GCCTGTAGTTGGCACCCAGA | (Kawagoe et al., 2005) | |||||
ADP-glucose pyrophosphorylase | OsAGPL1 | 2.7.7.27 | D50317 | 5 | [F] GGAAGACGGATGATCGAGAAAG | 140 | |
large subunit 1 | [R] CACATGAGATGCACCAACGA | ||||||
ADP-glucose pyrophosphorylase | OsAGPL2 | 2.7.7.27 | U66041 | 1 | [F] AGTTCGATTCAAGACGGATAGC | 96 | |
large subunit 2a | [R] CGACTTCCACAGGCAGCTTATT | ||||||
ADP-glucose pyrophosphorylase | OsAGPL3 | 2.7.7.27 | AK069296 | 3 | [F] AAGCCAGCCATGACCATTTG | 131 | |
large subunit 3 | [R] CACACGGTAGATTCACGAGACAA | ||||||
ADP-glucose pyrophosphorylase | OsAGPL4 | 2.7.7.27 | AK121036 | 7 | [F] TCAACGTCGATGCAGCAAAT | 77 | |
large subunit 4 | [R] ATCCCTCAGTTCCTAGCCTCATT | ||||||
Starch synthase I | OsSSI | 2.4.1.21 | D16202 | 6 | [F] GGGCCTTCATGGATCAACC | 279 | |
[R] CCGCTTCAAGCATCCTCATC | |||||||
Starch synthase IIa | OsSSIIa | 2.4.1.21 | AF419099 | 6 | [F] GCTTCCGGTTTGTGTGTTCA | 54 | SSII-3 |
[R] CTTAATACTCCCTCAACTCCACCAT | (Hirose et al., 2004; Jiang et al., 2004) | ||||||
Starch synthase IIb | OsSSIIb | 2.4.1.21 | AF395537 | 2 | [F] TAGGAGCAACGGTGGAAGTGA | 89 | SSII-2 |
[R] GTGAACGTGAGTACGTGACCAAT | (Hirose et al., 2004; Jiang et al., 2004) | ||||||
Starch synthase IIc | OsSSIIc | 2.4.1.21 | AF383878 | 10 | [F] GACCGAAATGCCTTTTTCTCG | 256 | SSII-1 |
[R] GGGCTTGGAGCCTCTCCTTA | (Hirose et al., 2004) | ||||||
Starch synthase IIIa | OsSSIIIa | 2.4.1.21 | AY100469 | 8 | [F] GCCTGCCCTGGACTACATTG | 334 | SSIII-2 |
[R] GCAAACATATGTACACGGTTCTGG | (Hirose et al., 2004) | ||||||
Starch synthase IIIb | OsSSIIIb | 2.4.1.21 | AF432915 | 4 | [F] ATTCCGCTCGCAAGAACTGA | 224 | SSIII-1 |
[R] CAACCGCAGGATAACGGAAA | (Hirose et al., 2004) | ||||||
Starch synthase IVa | OsSSIVa | 2.4.1.21 | AY100470 | 1 | [F] GGGAGCGGCTCAAACATAAA | 237 | SSIV-1 |
[R] CCGTGCACTGACTGCAAAAT | (Hirose et al., 2004) | ||||||
Starch synthase IVb | OsSSIVb | 2.4.1.21 | AY373258 | 5 | [F] ATGCAGGAAGCCGAGATGTT | 75 | SSIV-2 |
[R] ACGACAATGGGTGCCAAGAT | (Hirose et al., 2004; Jiang et al., 2004) | ||||||
Granule-bound starch synthase I | OsGBSSI | 2.4.1.21 | X62134 | 6 | [F] AACGTGGCTGCTCCTTGAA | 218 | Waxy |
[R] TTGGCAATAAGCCACACACA | (Okagaki, 1992) | ||||||
Granule-bound starch synthase II | OsGBSSII | 2.4.1.21 | AY069940 | 7 | [F] AGGCATCGAGGGTGAGGAG | 246 | |
[R] CCATCTGGCCCACATCTCTA | |||||||
Starch branching enzyme I | OsBEI | 2.4.1.18 | D11082 | 6 | [F] TGGCCATGGAAGAGTTGGC | 191 | |
[R] CAGAAGCAACTGCTCCACC | |||||||
Starch branching enzyme IIa | OsBEIIa | 2.4.1.18 | AB023498 | 4? | [F] GCCAATGCCAGGAAGATGA | 128 | |
[R] GCGCAACATAGGATGGGTTT | |||||||
Starch branching enzyme IIb | OsBEIIb | 2.4.1.18 | D16201 | 2 | [F] ATGCTAGAGTTTGACCGC | 261 | Amylose-extender |
[R] AGTGTGATGGATCCTGCC | (Nishi et al., 2001) | ||||||
Starch debranching enzyme: | OsISA1 | 3.2.1.68 | AB093426 | 8 | [F] TGCTCAGCTACTCCTCCATCATC | 132 | Sugary-1 |
Isoamylase I | [R] AGGACCGCACAACTTCAACATA | (Kubo et al., 1999) | |||||
Starch debranching enzyme: | OsISA2 | 3.2.1.68 | AC132483 | 5 | [F] TAGAGGTCCTCTTGGAGG | 170 | |
Isoamylase II | [R] AATCAGCTTCTGAGTCACCG | ||||||
Starch debranching enzyme: | OsISA3 | 3.2.1.68 | AP005574 | 9 | [F] ACAGCTTGAGACACTGGGTTGAG | 100 | |
Isoamylase III | [R] GCATCAAGAGGACAACCATCTG | ||||||
Starch debranching enzyme: | OsPUL | 3.2.1.41 | AB012915 | 4 | [F] ACCTTTCTTCCATGCTGG | 202 | |
Pullulanase | [R] CAAAGGTCTGAAAGATGGG | ||||||
Starch phosphorylase L | OsPHOL | 2.4.1.1 | AK063766 | 3 | [F] TTGGCAGGAAGGTTTCGCT | 66 | |
[R] CGAAGCCTGAAGTGAACTTGCT | |||||||
Starch phosphorylase Ha | OsPHOH | 2.4.1.1 | AK103367 | 1 | [F] CACCAAGACGAAGCTCATCAAG | 126 | |
[R] TTCACTCGTTGCTGGGTTCTC | |||||||
Disproportionating enzyme I | OsDPE1 | 2.4.1.25 | AP004306 | 7 | [F] TCTGTCGCTTGAACTTACAGAA | 77 | |
[R] TTTGGCTGCAAGTATTGTTCT | |||||||
Disproportionating enzyme IIa | OsDPE2 | 2.4.1.25 | AK067082 | 7 | [F] CAAGTACACCACAAGACCAGCAA | 107 | |
[R] CGTCCAACAGCGAATCCAAT |
Cytosolic type.
Determination of mRNA copy number
The copy number of mRNA was determined by a procedure based on that recommended by Applied Biosystems (http://www.appliedbiosystems.com/support/apptech/). In this procedure, the mass of a single plasmid template containing the target sequence was calculated and this mass was equated to one copy of the target gene sequence. The purified plasmid template was then quantified and serially diluted in TE to obtain plasmid solutions that differed by several orders of magnitude. Aliquots were used as templates for quantitative real-time PCR. Data were plotted to generate the standard curve. Plotting the values obtained by quantitative real-time PCR from any sample against this standard curve yields the approximate copy number of the target gene in the sample. In this study, the gene sequence copy number was equated to the mRNA copy number. The amount of mRNA synthesized by each gene was normalized as copy number per nanogram of total RNA according to Bustin (2000) and Gachon et al. (2004).
Identification of the members of starch-metabolizing enzyme gene families with DNA database search and computer analysis
To identify the members of each of the gene classes involved in starch biosynthesis, the databases of the National Center for Biotechnology Information (http://www.ncbi.nlm.nih.gov/) were searched for general entries of nucleotide sequences, and the KOME (Knowledge-based Oryza Molecular biological Encyclopedia) database of the National Institute of Agrobiological Sciences (http://cdna01.dna.affrc.go.jp/cDNA/) for full-length cDNAs. Multiple sequence alignment analyses for the deduced amino acid sequences were carried out using the CLUSTAL W program (http://clustalw.genome.jp/).
Results
Phylogenetic relationships among the genes for starch synthesis in rice, maize, Arabidopsis, and potato
The present study focused on the genes encoding six classes of enzymes, namely AGPase, SS (and GBSS), BE, DBE (ISA and PUL), PHO, and DPE (Table 1). For each enzyme class, the phylogenetic relationships among the rice genes and their corresponding homologues in another widely studied monocot maize (Zea mays) and the model dicot Arabidopsis thaliana were examined. In cases when genes corresponding to those of rice were absent in maize, the homologous genes from potato (Solanum tuberosum) were included in the analysis. The results of the comparisons are presented as dendrograms in Fig. 1. In all cases, the rice genes were phylogenetically more related to those of maize than Arabidopsis. These data support the current view that the evolution of the starch biosynthetic genes in monocots occurred after the divergence of the monocots from the dicots.
The dendrograms indicate that rice AGPS1 and AGPS2 were more recently evolved than the older AGPL3 and AGPL4, both of which have homologues reported in Arabidopsis (AtAPL1 and AtAPL2, respectively) but not in maize (Fig. 1A). The biggest class of starch-synthesizing genes, the starch synthase (SS), was phylogenetically separated into five subgroups corresponding to the five subclasses of SS genes: SSI, SSII, SSIII, SSIV, and GBSS (Fig. 1B). The SSII subclass appeared to be the most ancient. SSIIc, which had a homologue in Arabidopsis but not in maize, emerged much earlier than SSIIa and SSIIb. The rest of the SS genes arose from a common lineage from which SSI emerged first. Later coming out of the line were ones for the SSIII and SSIV genes and another for the GBSS genes. All SS genes had homologues in maize except for SSIIc, SSIIIa, SSIVa, and SSIVb, and GBSSII which had counterparts in Arabidopsis. All rice BE genes had corresponding genes in maize. OsBEI was likely to be more recent than OsBEIIa and OsBEIIb (Fig. 1C). Among the rice debranching enzymes, ISA3 may have evolved earliest, followed by ISA1, ISA2, and then PUL (Fig. 1D). All these DBEs had their homologues in both maize and Arabidopsis. OsPHOL, the more ancient PHO, had a homologue in maize, and OsPHOH in Arabidopsis (Fig. 1E). Similarly, maize has no known counterpart to both the rice DPE genes, but Arabidopsis and potato did (Fig. 1F). However, because the maize genome has not yet been fully determined, it is probable that maize, also a monocot, has a counterpart for all the rice genes involved in starch synthesis. If these maize genes are finally identified, it is predicted that they will be proven to be phylogenetically closer to their homologues in rice than in Arabidopsis.
Description of the gene classes
AGPase:
Plant AGPase is a heterotetrameric enzyme composed of two each of the larger regulatory subunits (AGPL) and the smaller catalytic subunits (AGPS) (Smith-White and Preiss, 1992; Preiss and Sivak, 1996; Greene and Hannah, 1998). Cereals have both plastidial and cytosolic AGPases (Denyer et al., 1996; Beckles et al., 2001; Hannah et al., 2001). Rice has a total of six genes encoding AGPase: two for AGPS (OsAGPS1 and OsAGPS2) and four for AGPL (OsAGPL1, OsAGPL2, OsAGPL3, and OsAGPL4). The AGPS2 gene apparently encodes the transcripts for AGPS2a and AGPS2b, which differ only in their first exons (AGPS2a exon 1 is encoded by a portion of the AGPS2 gene that serves as intron 1 of AGPS2b), and are either processed from the common pre-mRNA by alternative splicing mechanisms or produced by different promoters. AGPS2a is a 1930 bp transcript and carries a putative transit peptide sequence, thus probably plastidial in location, while AGPS2b is 1749 bp and has no transit peptide making it probably a cytosolic form. Similarly, the barley AGPase highly homologous small subunit transcripts bepsF1 (expressed in endosperm) and blps14 (expressed in leaves) are transcripts of a single gene for the AGPase small subunit (Thorbjønsen et al., 1996b), and the wheat small subunit transcripts AGP.S.1a (260 bp, probably cytosolic) and AGP.S.1b (353 bp, probably plastidial) are derived from the T.a.AGP.S.1 gene (Burton et al., 2002). In maize, however, the transcripts for the cytosolic and plastidial small subunits of AGPase were verified as products of two different genes (Hannah et al., 2001).
Judging from the presence of putative transit peptides and sequence homologies with corresponding AGPases from other species for which localization has been characterized (data not shown), OsAGPS1, OsAGPS2a, OsAGPL1, OsAGPL3, and OsAGPL4 are considered to be localized in plastids, whereas OsAGPS2b and OsAGPL2, which correspond to a maize Brittle-2 protein and a Shrunken-2 protein, respectively, are present in the cytosol (Bhave et al., 1990; Hannah et al., 2001).
SS and GBSS:
It is generally accepted that soluble starch synthase (SS) and starch granule-bound starch synthase (GBSS) are involved in amylopectin and amylose synthesis, respectively, although the possibility that GBSS plays a role in forming long chains of amylopectin cannot be ruled out (Delrue et al., 1992).
There are four types of SS: SSI, SSII, SSIII, and SSIV, while GBSS has two types: GBSSI and GBSSII. Rice has a total of eight genes for SS: a single gene for SSI (OsSSI), three genes for SSII (OsSSIIa, OsSSIIb, and OsSSIIc), two genes for SSIII (OsSSIIIa and OsSSIIIb), and two genes for SSIV (OsSSIVa and OsSSIVb) (Hirose and Terao, 2004). A single gene each for GBSSI (OsGBSSI) and GBSSII (OsGBSSII) are present in rice (Hirose and Terao, 2004).
BE:
Plants have two BE types, BEI and BEII. Rice contains a single gene for BEI (OsBEI) and two genes for BEII (OsBEIIa and OsBEIIb). It has been reported that OsBEIIb is specifically expressed in the endosperm whereas both BEI and BEIIa are expressed in all tissues examined (Yamanouchi and Nakamura, 1992; Mizuno et al., 1993).
DBE:
Plants have two types of DBE, ISA and PUL, which differ in properties such as substrate specificity (Nakamura, 1996). Rice has three ISA genes (Kubo et al., 2005) and a single PUL gene (Nakamura et al., 1996).
PHO:
Plant PHO is composed of two types, plastidial PHOL and cytosolic PHOH (Shimomura et al., 1982; Steup, 1988). A single gene each for PHOH and PHOL are present in rice.
DPE:
Plants have the plastidial type DPE1 and the cytosolic type DPE2 (Lu and Sharkey, 2004). Rice has a single gene for each.
Measurement of transcript levels by RT-PCR analysis
Expression profiles of these rice genes were determined by quantitative real-time PCR analysis. The oligonucleotide primer pairs used for quantitative measurement of the specific amounts of transcript for individual genes of interest are listed in Table 1. To discriminate the two transcript types of the AGPS2 gene, i.e. the AGPS2a and AGPS2b transcripts, the primer pairs were designed from their variable first exons. To verify the specificity of each primer set, their amplification products were cloned and sequenced. Plasmid DNAs containing these genes were used to establish standard curves, and the amount of mRNA synthesized by each gene was normalized as copy number per nanogram of total RNA according to Bustin (2000) and Gachon et al. (2004).
Expression patterns of each class of genes during endosperm development
Rice grain development in this study was divided into the following stages: initiation stage corresponding to 1–3 DAF when starch is synthesized exclusively in the pericarp and seed dry weight is relatively constant; early developmental stage (3–5 DAF) when seed total dry matter starts to increase and endosperm starch begins to accumulate; middle stage (5–10 DAF) when endosperm starch and grain dry weight rapidly increase; and late stage (10 DAF onward) when both endosperm starch and seed weight continue to increase, then maximum values are reached and they become constant. The expression profiles throughout seed development for each class of genes are described below.
AGPase:
AGPL1 and AGPS1 were vigorously expressed at the early phase of grain development, their transcripts being abundant from the onset of seed development, sharply rose to peak at 3 or 5 DAF then declined abruptly thereafter. By contrast, those of the two cytosolic forms, AGPS2b and AGPL2, were low until 3 DAF, then increased dramatically to peak at 5 DAF (Fig. 2A). All throughout seed development, the transcript level of AGPL3 was relatively low while AGPL4 and AGPS2a were scarcely expressed. These results suggest that AGPS1 combines with AGPL1 to form the plastidial AGPase, which is probably important before or/and at the early stage of endosperm/embryo development and/or pericarp. On the other hand, AGPS2b and AGPL2 possibly associate to form a heterotetrameric cytosolic AGPase, which is markedly expressed after 5 DAF to play an important role in starch accumulation via AGPglucose synthesis in endosperm.
SS and GBSS:
The expression profiles of the SS and GBSS genes were highly variable (Fig. 2B). The SSI transcript level was already high at the earliest phase of seed formation (1–3 DAF), slightly increased to peak at 5 DAF when starch synthesis in the endosperm begins, then remained almost constant through the late-milking stage of endosperm development (15 DAF), suggesting that SSI is important at all stages of seed development. This is consistent with the fact that SSI is the major SS form in cereal endosperm (Cao et al., 1999).
The transcripts of SSIIa and SSIIIa rapidly increased from low levels at the onset of seed development to peak at 5–7 DAF and then continued to be significantly high. It is interesting that, among the SS genes, SSIIa appeared to be the most vigorously expressed, more than SSI or SSIIIa, whereas SSI and SSIIIa are known to account for the major SS activities in cereal endosperm (Cao et al., 1999).
By contrast, transcripts for SSIIIb were abundant at the start of grain formation until the early stage of seed development, but dropped precipitously at the onset of endosperm starch synthesis. On the other hand, SSIVb transcripts were low at the start then rose briefly at 5 DAF to the maximum level attained by SSIIIb and gradually decreased. Across all sampling periods, the transcripts of SSIIIb and SSIVb were consistently lower than those of SSIIa, SSIIIa, and SSI. The rest of the SS genes, i.e. SSIIb, SSIIc, and SSIVa were barely expressed in the seed, all having their maximum transcripts at the earliest period of grain formation and decreasing to basal levels at 5 DAF, the start of starch accumulation in the endosperm (Fig. 2B2).
While the GBSSI transcripts were low at the beginning of seed development, they dramatically increased at 5 DAF and remained abundant until the end of endosperm development (Fig. 2B1). Clearly, the amount of GBSSI transcript was markedly higher than GBSSII or any of the SS genes, its maximum level being more than 3-fold higher than that of the most vigorously expressed SS gene (SSIIa) and approximately 55-fold of the least-expressed SS gene (SSIVa). By stark contrast, the expression of GBSSII was very low, and limited only to the early stage of seed development (Fig. 2B2).
BE:
Compared with the number of transcripts of the other two BEs, those of BEIIa were at least one order higher at the onset of the grain-development stage (1–3 DAF), as shown in Fig. 2C. This relatively higher transcript level, however, was transient because it started to dwindle at 5 DAF when endosperm starch started to accumulate. By contrast, the relatively low transcripts of BEIIb and BEI rapidly increased from 3 DAF to peak at 5 and 7 DAF, respectively, and despite some fluctuations remained high until 15 DAF. While the expression profiles of BEIIb and BEI were similar, BEIIb transcripts from 7–20 DAF were only about half of those for BEI. These patterns of changes in the transcript levels of the BEII isoforms are consistent with the findings of Mizuno et al. (2001). These observations might suggest that BEIIb and BEI are needed in great amounts during endosperm starch synthesis, particularly straight after the initiation process, whereas BEIIa plays an important role in processes occurring before efficient starch synthesis in the endosperm. As with other genes, however, the transcript levels of BEII isoforms are not necessarily the measures of their enzymatic activities. Illustrative of this reality in rice are the observations that, while the transcript level of BEIIa was much lower than that of BEIIb after 7 DAF (Fig. 2C), the activities of BEIIa and BEIIb are actually maintained at the same levels throughout endosperm development (data not shown).
It is interesting that BEI and BEIIb are expressed early in and throughout endosperm development in rice (Mizuno et al., 1993), maize (Gao et al., 1996), and sorghum (Mutisya et al., 2003), whereas in wheat (Morell et al., 1997) and barley (Sun et al., 1998; Mutisya et al., 2003) the BEI gene is not expressed until relatively late in endosperm development as compared with the BEIIb gene.
DBE:
All the DBEs had comparatively low transcript levels at the onset of seed development (1–3 DAF), which were maintained by ISA2 and ISA3 all throughout seed development. By contrast, the transcripts of ISA1 and PUL rose steeply from 3 DAF to reach peak levels at 7 DAF of about 12- and 56-fold higher, respectively, then gradually decreased to lower but still significant levels at 20 DAF (Fig. 2D), suggesting that ISA1 and PUL play important roles in the whole endosperm starch accumulation.
PHO:
PHOL transcripts were numerous at the onset of seed development, rapidly increased to peak at 5 DAF, and diminished to a low but significant level until 15 DAF, suggesting that PHOL plays an important part in starch biosynthesis throughout endosperm development. The transcripts of PHOH, which never exceeded those of PHOL were scarce from the start and dwindled from 5 DAF (Fig. 2E).
DPE:
Although the absolute number of transcripts of DPE2 was consistently lower than that for DPE1, the expression patterns of the two genes were similar (Fig. 2F). Their transcripts were already plentiful at the onset of seed development, further increased to peak by 5 DAF, then abruptly decreased, suggesting a possibility that the involvement of both DPEs in seed development is confined to processes occurring before and/or at the initial stage of starch biosynthesis in rice endosperm.
Comparison of the expression patterns of each class of genes between leaf and endosperm
The expression patterns of the genes involved in starch metabolism were similar among young, fully developed, and old leaves, although in general young leaves had more transcripts than older ones (Fig. 3A). Because this trend across leaf age was apparent for the members of each gene class, only the data for the young leaf are reported.
AGPase:
In young leaf, the transcript level of AGPS2a was about 11-fold higher than that of AGPS1 (Fig. 3A) but the AGPS2b transcript was not detected (Fig. 3A). In seed, the transcript level for AGPS2a was consistently low while that for AGPS2b, although lower than AGPS1 at the initial developmental stage, became higher at stages after 7 DAF (Fig. 2A). These data suggest that AGPS2a and AGPS2b play important roles in the synthesis of leaf transitory starch and endosperm storage starch, respectively. In green leaves, AGPL3 transcripts were predominant, those of AGPL1, AGPL2, and AGPL4 being scarce (Fig. 3A), while in developing seed, AGPL4 transcripts were very low as compared with those of the other three AGPL genes (Fig. 2A).
SS and GBSS:
SSI was markedly expressed in developing endosperm (Fig. 2B) as well as in the leaf (Fig. 3B). Among the SSII genes, SSIIb was the major form expressed in the leaf while SSIIa, which was markedly expressed in the endosperm was scarcely expressed. The SSIIc transcript level was very low in both leaf and seed (Fig. 3B). SSIIIb was the only SSIII isoform expressed in the leaf (Fig. 3B) but, in seed during vigorous starch biosynthesis, SSIIIa expression was higher than SSIIIb (Fig. 2B). The transcript levels of SSIVb were higher than SSIVa in both leaf and endosperm (Figs 2B, 3B). The expression of GBSSII was leaf specific while GBSSI was endosperm specific (Figs 2B, 3B). These different expression patterns of SS and GBSS genes are consistent with previous reports (Hirose and Terao, 2004; Dian et al., 2005).
BE:
The major BE forms expressed in the leaf were BEIIa and BEI (Fig. 3C), while BEI and BEIIb were predominant in the endosperm at the middle and late developmental stages (Fig. 2C). BEIIb expression was specific to endosperm, its transcript being very low in the leaf (Figs 2C, 3C). These results are in agreement with previous studies (Yamanouchi and Nakamura, 1992; Mizuno et al., 1993).
DBE:
ISA1 and ISA2 were expressed only slightly in leaves but significantly in seeds (Figs 2D, 3D). ISA3 was expressed markedly in the leaf but scarcely in the endosperm (Figs 2D, 3D). The PUL transcript level was significantly lower in the leaf than in the seed (Figs 2D, 3D).
PHO:
PHOH expression was about 10-fold higher than PHOL in the leaf whereas in the endosperm, PHOL transcripts were about 6-fold higher than PHOH (Figs 2E, 3E).
DPE:
DPE2 was the predominant form in the leaf, whereas both DPE1 and DPE2 were significantly expressed in the endosperm but they were almost restricted to the early developmental stage (Figs 2F, 3F).
Effects of shrunken mutation on expression of AGPase genes
Results of recent studies indicate that the defective AGPS2 in the shrunken mutant reduces the total AGPase activity to about 20% of the wild type (Kawagoe et al., 2005). Yano et al. (1984) reported previously that the AGPS2 protein is almost completely lacking in the endosperm, and that reduced AGPase activity causes a decline in starch accumulation in the endosperm, leading to shrivelling of the seed at maturity.
The transcripts of the cytosolic AGPS2b, which were absent from the leaves of the wild type and its shrunken2 mutant EM22, were practically non-existent in the seeds of the mutant (Fig. 4). AGPS2a transcript abundance was comparable in the leaves of the mutant and the wild type but was more than 4-fold higher in the seeds of the mutant (Fig. 4). These results strongly suggest that the shrunken seed phenotype of EM22 is caused by the absence of transcripts for AGPS2b, probably the result of either mutation in the AGPS2 gene itself or anomalies in the splicing of the pre-mRNA transcript. It is striking to note that the expressions of AGPS1, AGPL1, AGPL3, and especially AGPL2 as well as AGPS2a, were dramatically enhanced in the seed (Fig. 4A). On the other hand, the most dramatic effect of the shrunken mutation in the leaf was the remarkable increase in transcript levels of both AGPS1 and AGPL1, while that of AGPL3, the major AGPL isoform in the leaf, was unaffected by the mutation (Fig. 4B). These results indicate that the reduction of AGPS2b expression caused by the shrunken mutation greatly elevates the expression of the AGPL and other AGPS genes in the seed and leaf.
Discussion
Amylopectin structure is influenced by starch-synthesizing enzymes
The fine structures as well as the relative amounts of amylopectin and amylose are specific for plant species and plant tissues where they are synthesized, and these features influence the physicochemical properties of starch granules. Several enzymes influence the fine structure of amylopectin, mainly by altering its cluster structure, although the modes of changes in amylopectin cluster structure strongly depend on the individual enzyme affected. In rice endosperm, several lines of evidence for this are available. First, SSIIa is responsible for the elongation of short-chains with DP5-10 to form longer A and B1 chains with DP11-24 within the cluster of amylopectin, and hence the difference in capacity for SSIIa discriminates the structural alteration of amylopectin between indica-type and japonica-type rice varieties having active SSIIa and inactive SSIIa, respectively (Umemoto et al., 2002; Nakamura et al., 2005). Secondly, BEIIb plays a specific role in the formation of A-chains of the amylopectin cluster (Nishi et al., 2001) and the levels of BEIIb activity affect the extent of structural changes in the cluster (Tanaka et al., 2004). Thirdly, ISA1 is essential for the formation of the highly organized cluster structure of amylopectin (Kubo et al., 1999) and the level of ISA1 expression influences both the fine structure of amylopectin and starch thermal properties (Fujita et al., 2003; Kubo et al., 2005). These observations indicate that, in rice, the expression levels of sets of genes are programmed depending on the tissues and their developmental stages, otherwise each plant tissue cannot have a fixed structure of amylopectin, which is basically maintained throughout the life cycle. The tissue- and developmental stage-specific expression patterns of all genes for starch biosynthesis determine the fine structure of amylopectin in a particular tissue at a particular developmental stage, and hence the starch physicochemical characteristics, which influence the physiological status of cells. The physiological significance of the morphological and physicochemical properties of starch granules, however, is poorly understood.
The composition and the relative activities of isozymes of each class of the starch-synthesizing enzymes are known to differ between monocots and dicots. In most cereals (monocot), for instance, the BE class has several isozymes. In rice, the expression of the BEIIb gene is endosperm-specific while the BEIIa gene is ubiquitous (Yamanouchi and Nakamura, 1992; Mizuno et al., 1993, 2001). By contrast, dicots have either only a single BEII-type gene or multiple BEII genes, if any, that belong to the same gene subfamily (Sun et al., 1998). These differences in the relative activities and site of expressions of BE isoforms give rise to different X-ray diffraction patterns of starch granules, i.e. A-type in cereal grains, B-type or C-type in dicot organs (Hizukuri, 1996), and B-type in BEIIb-suppressed rice endosperm (Tanaka et al., 2004).
Expression profiling separates the genes involved in endosperm starch synthesis into four groups
The rice genome has a total of 27 candidate genes encoding the starch biosynthesizing enzymes and these are distributed in six classes as follows: six for AGPase, 10 for SS and GBSS, three for BE, four for DBE, two for PHO, and two for DPE. The present study analysed the transcriptome of these genes expressed in rice leaves and seeds as representative photosynthetic and non-photosynthetic organs, respectively. Detailed transcriptome analysis during seed endosperm development was also examined.
Results of the present study revealed four major patterns of expression of the rice starch-synthesizing genes during seed development, separating the enzymes into four groups (Fig. 2, Table 2; Supplemental Fig. 1). The first group is characterized by a high level of expression at a very early stage (initiation) of seed development, reaching a peak at 3 or 5 DAF followed by an abrupt decline, as exhibited by OsAGPS1, OsAGPL1, OsSSIIIb, OsBEIIa, OsPHOH, OsDPE1, and OsDPE2.
Gene group . | Expression level at grain developmental stages . | . | . | . | |||
---|---|---|---|---|---|---|---|
. | Initiationa . | Earlyb . | Middlec . | Lated . | |||
. | (1–3 DAF) . | (3–5 DAF) . | (5–10 DAF) . | (10–20 DAF) . | |||
Group 1 | |||||||
OsAGPS1, OsAGPL1, | High level | Increases to maximum level | Abruptly declines to low level | Stable low level | |||
OsSSIIIb, OsBEIIa, | at 3 to/or 5 DAF | through 7 DAF then stabilizes | |||||
OsPHOH, OsDPE1, | |||||||
OsDPE2 | |||||||
Group 2 | |||||||
OsAGPL3, OsSSI, | Medium level | Rapidly increases to reach | Gradually declines to low level | Continues to decline or | |||
OsSSIVb, OsISA2, | peak at 5 DAF | at 10 DAF | briefly rises at 15 DAF | ||||
OsPHOL | |||||||
Group 3 | |||||||
OsAGPS2b, OsAGPL2, | Low level | Rapidly increases, some | Slightly decreases or some | Slightly increases or | |||
OsSSIIa, OsSSIIIa, | attaining peak level | continue to rise to reach peak | gradually declines | ||||
OsGBSSI, OsBEI, | at 5 DAF | level at 7 DAF | |||||
OsBEIIb, OsISA1, | |||||||
OsPUL | |||||||
Group 4 | |||||||
OsAGPS2a, OsAGPL4, | Low level | Low level | Decreases to basal level | Constantly at basal level | |||
OsSSIIb, OsSSIIc, | |||||||
OsSSIVa, OsGBSSII, | |||||||
OsISA3 |
Gene group . | Expression level at grain developmental stages . | . | . | . | |||
---|---|---|---|---|---|---|---|
. | Initiationa . | Earlyb . | Middlec . | Lated . | |||
. | (1–3 DAF) . | (3–5 DAF) . | (5–10 DAF) . | (10–20 DAF) . | |||
Group 1 | |||||||
OsAGPS1, OsAGPL1, | High level | Increases to maximum level | Abruptly declines to low level | Stable low level | |||
OsSSIIIb, OsBEIIa, | at 3 to/or 5 DAF | through 7 DAF then stabilizes | |||||
OsPHOH, OsDPE1, | |||||||
OsDPE2 | |||||||
Group 2 | |||||||
OsAGPL3, OsSSI, | Medium level | Rapidly increases to reach | Gradually declines to low level | Continues to decline or | |||
OsSSIVb, OsISA2, | peak at 5 DAF | at 10 DAF | briefly rises at 15 DAF | ||||
OsPHOL | |||||||
Group 3 | |||||||
OsAGPS2b, OsAGPL2, | Low level | Rapidly increases, some | Slightly decreases or some | Slightly increases or | |||
OsSSIIa, OsSSIIIa, | attaining peak level | continue to rise to reach peak | gradually declines | ||||
OsGBSSI, OsBEI, | at 5 DAF | level at 7 DAF | |||||
OsBEIIb, OsISA1, | |||||||
OsPUL | |||||||
Group 4 | |||||||
OsAGPS2a, OsAGPL4, | Low level | Low level | Decreases to basal level | Constantly at basal level | |||
OsSSIIb, OsSSIIc, | |||||||
OsSSIVa, OsGBSSII, | |||||||
OsISA3 |
Dehulled developing seeds (including the endosperm and the pericarp) were analysed.
Grain starch not detected, and no increase in grain dry weight.
Grain dry weight increases from 3 DAF, grain starch accumulates from 5 DAF.
Rapid increases in grain starch content and in grain dry weight.
Continued rapid increases in grain starch content and in grain dry weight.
Gene group . | Expression level at grain developmental stages . | . | . | . | |||
---|---|---|---|---|---|---|---|
. | Initiationa . | Earlyb . | Middlec . | Lated . | |||
. | (1–3 DAF) . | (3–5 DAF) . | (5–10 DAF) . | (10–20 DAF) . | |||
Group 1 | |||||||
OsAGPS1, OsAGPL1, | High level | Increases to maximum level | Abruptly declines to low level | Stable low level | |||
OsSSIIIb, OsBEIIa, | at 3 to/or 5 DAF | through 7 DAF then stabilizes | |||||
OsPHOH, OsDPE1, | |||||||
OsDPE2 | |||||||
Group 2 | |||||||
OsAGPL3, OsSSI, | Medium level | Rapidly increases to reach | Gradually declines to low level | Continues to decline or | |||
OsSSIVb, OsISA2, | peak at 5 DAF | at 10 DAF | briefly rises at 15 DAF | ||||
OsPHOL | |||||||
Group 3 | |||||||
OsAGPS2b, OsAGPL2, | Low level | Rapidly increases, some | Slightly decreases or some | Slightly increases or | |||
OsSSIIa, OsSSIIIa, | attaining peak level | continue to rise to reach peak | gradually declines | ||||
OsGBSSI, OsBEI, | at 5 DAF | level at 7 DAF | |||||
OsBEIIb, OsISA1, | |||||||
OsPUL | |||||||
Group 4 | |||||||
OsAGPS2a, OsAGPL4, | Low level | Low level | Decreases to basal level | Constantly at basal level | |||
OsSSIIb, OsSSIIc, | |||||||
OsSSIVa, OsGBSSII, | |||||||
OsISA3 |
Gene group . | Expression level at grain developmental stages . | . | . | . | |||
---|---|---|---|---|---|---|---|
. | Initiationa . | Earlyb . | Middlec . | Lated . | |||
. | (1–3 DAF) . | (3–5 DAF) . | (5–10 DAF) . | (10–20 DAF) . | |||
Group 1 | |||||||
OsAGPS1, OsAGPL1, | High level | Increases to maximum level | Abruptly declines to low level | Stable low level | |||
OsSSIIIb, OsBEIIa, | at 3 to/or 5 DAF | through 7 DAF then stabilizes | |||||
OsPHOH, OsDPE1, | |||||||
OsDPE2 | |||||||
Group 2 | |||||||
OsAGPL3, OsSSI, | Medium level | Rapidly increases to reach | Gradually declines to low level | Continues to decline or | |||
OsSSIVb, OsISA2, | peak at 5 DAF | at 10 DAF | briefly rises at 15 DAF | ||||
OsPHOL | |||||||
Group 3 | |||||||
OsAGPS2b, OsAGPL2, | Low level | Rapidly increases, some | Slightly decreases or some | Slightly increases or | |||
OsSSIIa, OsSSIIIa, | attaining peak level | continue to rise to reach peak | gradually declines | ||||
OsGBSSI, OsBEI, | at 5 DAF | level at 7 DAF | |||||
OsBEIIb, OsISA1, | |||||||
OsPUL | |||||||
Group 4 | |||||||
OsAGPS2a, OsAGPL4, | Low level | Low level | Decreases to basal level | Constantly at basal level | |||
OsSSIIb, OsSSIIc, | |||||||
OsSSIVa, OsGBSSII, | |||||||
OsISA3 |
Dehulled developing seeds (including the endosperm and the pericarp) were analysed.
Grain starch not detected, and no increase in grain dry weight.
Grain dry weight increases from 3 DAF, grain starch accumulates from 5 DAF.
Rapid increases in grain starch content and in grain dry weight.
Continued rapid increases in grain starch content and in grain dry weight.
The second group, composed of OsAGPL3, OsSSI, OsSSIVb, OsISA2, and OsPHOL, has a medium expression level at initiation of grain formation, rapidly rises to peak at 5 DAF, gradually declines through 10 DAF, then either rises slightly at 15 DAF (OsSSI and OsISA2), or continues to decline. Because preliminary results show that ISA2 plays an important role in amylopectin biosynthesis by forming a functional heteromeric debranching enzyme protein with OsISA1, as found in potato tubers (Hussain et al., 2003), OsISA2 has been classified into this group, although its transcript level is low.
The third group is characterized by a basal or a very low level of expression at the initial stage of seed development followed by a rapid increase in transcripts from about 5 or 7 DAF, which is maintained at significantly high levels until seed maturation. Possessing this pattern are the genes for OsAGPS2b, OsAGPL2, OsSSIIa, OsSSIIIa, OsGBSSI, OsBEI, OsBEIIb, OsISA1, and OsPUL. Since both OsSSIIIa and OsISA1 are also expressed to some extent during the initial stage of the seed development, these enzymes may function in amylopectin synthesis from the very early stage of endosperm development.
The fourth group is characterized by low transcript level at the start (1–5 DAF) and further decreases to a basal, barely detectable level throughout the rest of seed development as exhibited by OsAGPS2a, OsAGPL4, OsSSIIb, OsSSIIc, OsSSIVa, OsGBSSII, and OsISA3.
Altogether, the above observations show that the timing of gene expression greatly differs among gene classes and among members of the same class, consistent with a previous report of different patterns of changes in enzyme activities during rice endosperm development (Nakamura and Yuki, 1992). The apparent grouping of many of the genes into the four expression patterns suggests that the physiological significance of starch synthesis dramatically changes at around 5 DAF during seed development. This particular period seems to be the exact timing when the seed starts to accumulate a huge amount of starch in the endosperm. It is assumed that the enzymes exhibiting the first pattern (the first group) play crucial roles in starch biosynthesis for the construction of the fundamental cell machineries in the embryo, aleurone layer, pericarp, and endosperm until about 3 DAF. It is also possible that the first-group enzymes play important roles in the de novo synthesis of glucan primers and the initiation of starch granules in the endosperm. The group 1 genes, together with the early but scarcely expressed group 4 genes, may also be involved in the synthesis of transitory starch in the seed pericarp. Recent work on SS genes in rice grains (Hirose and Terao, 2004) showed that the transcripts of SSIIb, SSIIIb, and GBSSII are mainly localized in the pericarp. These SS genes were designated as ‘early expressers’ because their transcripts lasted only until 5 DAF when endosperm starch evidently starts to accumulate.
It is assumed that PHO and DPE, which are usually considered to be involved in the process of starch degradation, may also play a role in the initial processes of starch biosynthesis in seeds, although their functions in starch biosynthesis have not been proven in plants, except in Chlamydomonas (Colleoni et al., 1999; Critchley et al., 2001).
At about 7 DAF or later, the endosperm cells are specialized to produce starch from sucrose and are destined to store the starch granules until amyloplasts enlarge to fill the whole cell space. At this time, the high expression of the enzymes belonging to the second and third groups could be closely related to the capacity of endosperm to synthesize starch efficiently. In this sense, it is reasonable to assume that all the enzymes that play specific roles in starch production in the endosperm are included in this category (second and third groups), because rice mutants lacking these enzymes such as SSIIa (Umemoto et al., 2002), BEI (Satoh et al., 2003b), BEIIb (Nishi et al., 2001), and ISA1 (Kubo et al., 1999) exhibit modified starch phenotypes in the endosperm (Nakamura, 2002; Satoh et al., 2003a).
The expression of the starch-synthesizing genes radically differs between leaves and seeds
The patterns of gene expression in the leaf sharply contrast with those in the seed (compare Fig. 3 with Fig. 2), as indicated by previous data on the activities of enzymes involved in carbohydrate metabolism in leaves and seeds of rice (Nakamura et al., 1989). Some genes such as OsAGPS2 (for the AGPS2a transcript), OsAGPL3, OsSSIIb, OsSSIIIb, OsGBSSII, OsBEIIa, OsISA3, OsPHOH, and OsDPE2 are specifically or markedly expressed in the leaf and none or only slightly in the seed, while OsAGPS2 (for the AGPS2b transcript), OsAGPL1, OsAGPL2, OsSSIIa, OsSSIIIa, OsGBSSI, OsBEIIb, and OsISA1 are preferably expressed in the seed (Table 3; Supplemental Fig. 2). The expressions of OsAGPS2a, OsSSIIb, OsSSIIIb, OsGBSSII, and OsDPE2 are considered leaf specific while those for OsAGPS2b, OsSSIIa, OsGBSSI, and OsBEIIb are regarded as seed specific because their transcripts are not detected in the other organ (Table 3; Supplemental Fig. 2). OsSSI is highly expressed in both organs. There is no correlation between the organ-specific gene expression patterns and the developmental stage-specific expression patterns in the endosperm, as can be gleaned from Tables 2, 3; Supplemental Figs 1, 2).
Tissue . | Genes . |
---|---|
Leaf | OsAGPS2a, OsAGPL3, OsSSIIb, OsSSIIIb, OsGBSSII, OsBEIIa, OsISA3, OsPHOH, OsDPE2 |
Endsperm | OsAGPS2b, OsAGPL1, OsAGPL2, OsSSIIa, OsSSIIIa, OsGBSSI, OsBEIIb, OsISA1 |
Tissue . | Genes . |
---|---|
Leaf | OsAGPS2a, OsAGPL3, OsSSIIb, OsSSIIIb, OsGBSSII, OsBEIIa, OsISA3, OsPHOH, OsDPE2 |
Endsperm | OsAGPS2b, OsAGPL1, OsAGPL2, OsSSIIa, OsSSIIIa, OsGBSSI, OsBEIIb, OsISA1 |
Expression in the endosperm is defined as ‘specific’ (underlined) and ‘preferred’ when the transcript level is at least 100-fold and 10-fold, respectively, of that in the leaf. Corresponding threshold values in the leaf are less than 0.1-fold and between 0.1–0.25-fold, respectively, of that in the endosperm.
Tissue . | Genes . |
---|---|
Leaf | OsAGPS2a, OsAGPL3, OsSSIIb, OsSSIIIb, OsGBSSII, OsBEIIa, OsISA3, OsPHOH, OsDPE2 |
Endsperm | OsAGPS2b, OsAGPL1, OsAGPL2, OsSSIIa, OsSSIIIa, OsGBSSI, OsBEIIb, OsISA1 |
Tissue . | Genes . |
---|---|
Leaf | OsAGPS2a, OsAGPL3, OsSSIIb, OsSSIIIb, OsGBSSII, OsBEIIa, OsISA3, OsPHOH, OsDPE2 |
Endsperm | OsAGPS2b, OsAGPL1, OsAGPL2, OsSSIIa, OsSSIIIa, OsGBSSI, OsBEIIb, OsISA1 |
Expression in the endosperm is defined as ‘specific’ (underlined) and ‘preferred’ when the transcript level is at least 100-fold and 10-fold, respectively, of that in the leaf. Corresponding threshold values in the leaf are less than 0.1-fold and between 0.1–0.25-fold, respectively, of that in the endosperm.
The mechanism for organ-specific gene expression might reflect differences in the pathway for signal transduction of gene expression between the leaf as a photosynthetic organ and the endosperm as a non-photosynthetic organ, because light and sugars such as sucrose might be important possible factors regulating gene expressions in the leaf and the endosperm, respectively. A set of enzymes is required for the synthesis of the tandem-cluster structure of amylopectin (an organized arrangement of amylopectin and amylose molecules in the starch granule), and for determining the shape and size of the starch granule, which is organ and species specific. The involvement of different sets of enzymes in starch biosynthesis in source and sink organs might result in the different molecular structures of amylopectin and starch granule morphology, in agreement with the well-known fact that starch structure differs between source and sink organs to meet the physiological significance of starch as temporarily stored assimilatory material in source organs and as reserve material in sink organs (Tomlinson et al., 1997).
The present data are basically consistent with the previous studies on the expression patterns of these genes in the seed and/or the leaf of rice plants (Mizuno et al., 2001; Hirose and Terao, 2004; Jiang et al., 2004; Dian et al., 2005; Suzuki et al., 2005).
Subunits of AGPase may form homotetrameric or heterotetrameric enzymes
Plant AGPase is a heterotetrameric enzyme composed of two large and two small subunits and its heteromeric nature has been extensively examined (Smith-White and Preiss, 1992; Greene and Hannah, 1998). AGPase catalyses the first committed reaction for starch biosynthesis by supplying SS with ADPglucose in plant tissues (Preiss and Sivak, 1996). There is compelling evidence indicating that, in some cereals such as maize, barley, rice, and wheat, cytosolic AGPase accounts for the major AGPase activity in the endosperm (Denyer et al., 1996; Thorbjørnsen et al., 1996a; Beckles et al., 2001; Hannah et al., 2001; Sikka et al., 2001; Johnson et al., 2003; Tetlow et al., 2003). It is assumed that in rice endosperm, OsAGPS2b and OsAGPL2 associate to form a tetrameric cytosolic AGPase, because both proteins are the only known cytosolic AGPase small and large subunits, and are apparently co-ordinately expressed during endosperm development (Fig. 2A), consistent with the fact that maize cytosolic AGPase is composed of Brittle2 protein (small subunit) and Shrunken2 protein (large subunit) (Greene and Hannah, 1998). On the other hand, the expression patterns of plastidic OsAGP genes in the seed suggest that OsAGPS1 associates with OsAGPL1 to constitute the plastidic AGPase that functions at the onset of seed developmental stage up to 5 DAF, beyond which OsAGPL3 possibly replaces OsAGPL1 in the endosperm (Fig. 2A). In leaves, only the OsAGPS2a and OsAGPL3 transcripts are abundant, the OsAGPS1 gene being slightly expressed, whereas the transcript levels of OsAGPL1, OsAGPL2, and OsAGPL4 are very low, and the OsAGPS2b transcripts are absent (Fig. 3A). These results suggest that in rice leaves, the major plastidial AGPase is a heterotetramer composed of OsAGPS2a and OsAGPL3, the less probable minor form being an AGPS1-AGPL3 heterotetramer, and it is likely that a cytosolic AGPase does not exist.
Reduced AGPS2b transcripts exert dramatic pleiotropic effects on the transcript levels of other AGPase subunit genes
The defective OsAGPS2 gene in the shrunken mutation causes the complete loss of the OsAGPS2b transcript (Fig. 4) and the OsAGPS2 protein in rice endosperm (data not shown), and a reduction of the total AGPase activity to about 20% of the wild type (Kawagoe et al., 2005). The shrunken mutation dramatically enhances the expression of other OsAGP genes in both seeds and leaves (Fig. 4). The expression of OsAGPL2, possibly the counterpart of OsAGPS2b for the cytosolic AGPase in the wild-type cultivar, is remarkably increased in the seed but not in the leaf, consistent with the observation by Giroux et al. (1994) in maize endosperm that the mutation of either the AGPS2 (Brittle2) or the AGPL2 (Shrunken2) gene causes a marked increase in the transcript level of the counterpart gene. One possible explanation might be that in the cytosol of the mutant endosperm, OsAGPL2 forms a homotetramer that may or may not be catalytically active. The shrunken mutation also elevates the transcript levels for OsAGPS1, OsAGPS2a, OsAGPL1, and OsAGPL3, suggesting that the mutation enhances the plastidial AGPase expression and hence activity, presumably to compensate for the absence of AGPS2b. This may be the reason why the residual activity of AGPase in the mutant is quite high (about 20% of the wild-type activity), and the mutant can store starch in endosperm to some extent (Yano et al., 1984; Kawagoe et al., 2005). It is also interesting to note that the expressions of OsAGPS1 and OsAGPL1 remarkably increase in the leaf of the mutant, although the physiological significance and mechanism of these elevated AGPase gene expressions to the leaf chloroplasts are unknown.
Conclusions
The present results collectively provide concrete evidence that the expressions of the numerous genes involved in starch metabolism are highly co-ordinated temporally and spatially in both source and sink tissues of rice plants, keeping distinct expression patterns depending on individual genes; and that the changes in the expressions of some genes such as OsAGPS2 due to the shrunken mutation can greatly affect the patterns of gene expression of other members of both its subclass (AGPS) and the other subclass (AGPL). Since every class of starch-synthesizing enzymes has multiple isoforms or subunits and all these enzymes are usually labile, it is impossible or very difficult to quantify separately all of these enzymatic activities or protein amounts in plant tissues, especially in the presence of interfering enzymes such as hydrolytic enzymes like amylases. Transcriptome analysis has an advantage of quantifying the changes in transcript levels of these genes at different developmental stages in wild-type species or their mutants and in cultivars with a different genetic background. Therefore, expression profiling of genes could lay a foundation for the identification of genes involved in the regulation of starch metabolism and provide valuable insights into the mechanism of metabolic regulation of starch biosynthesis under various physiological conditions. Nevertheless, it should be kept in mind that the transcript levels are not always related to their enzymatic activities or protein levels, considering post-transcriptional and post-translational controls.
The starch-biosynthesizing system in higher plants demands the concerted expression of many genes at any particular developmental stage of a particular tissue. In addition to maintaining the activity levels of individual enzymes, several combinations of protein–protein interactions such as those among the wheat proteins for BEI, BEIIb, and PHO (Tetlow et al., 2004) might play an essential role in the metabolic regulation of starch biosynthesis. Since it is difficult to measure separately the protein level or activity of each enzyme in the enzyme complex, transcriptome analysis would be a more convenient tool to reveal the regulatory mechanism for starch synthesis that determines when and where multiple genes are co-expressed and what proteins may interact.
Supplementary material
Supplementary figures are available online at www.jxb.oxfordjournals.org.
We thank Drs Naoko Fujita and Akiko Kubo for their comments and suggestions, and the staff of the CREST-Akita Plant Molecular Science Satellite Laboratory, Life Science Research Support Center in Akita Prefectural University for their help in DNA sequencing and analysis. This work is supported in part by Grant-in-Aid for Scientific Research (no. 15580016) from the Ministry of Education, Culture, Sports, Science and Technology of Japan to TH.
References
Ball SG, Morell MK.
Beckles DM, Smith AM, ap Rees T.
Bhave MR, Lawrence S, Barton C, Hannah LC.
Burton RA, Johnson PE, Beckles DM, Fincher GB, Jenner HL, Naldrett MJ, Denyer K.
Bustin SA.
Cao H, Imparl-Radosevich J, Guan H, Keeling PL, James MG, Myers AM.
Chang S, Puryear J, Cairney J.
Colleoni C, Dauvillée D, Mouille G, et al.
Critchley JH, Zeeman SC, Takaha T, Smith AM, Smith SM.
Delrue B, Fontaine T, Routier F, Decq A, Wieruszeski JM, Van Den Koornhuyse N, Mddelein M-L, Fournet B, Ball S.
Denyer K, Dunlap F, Thorbjørnsen T, Keeling P, Smith AM.
Dian W, Jiang H, Wu P.
Fujita N, Kubo A, Suh D-S, Wong K-S, Jane J-L, Ozawa K, Takaiwa F, Inaba Y, Nakamura Y.
Gachon C, Mingam A, Charrier B.
Gallant DJ, Bouchet B, Baldwin PM.
Gao M, Fisher DK, Kim K, Shannon JC, Guiltinan MJ.
Giroux MJ, Boyer C, Feix G, Hannah LC.
Greene TW, Hannah LC.
Hannah LC, Shaw JR, Giroux MJ, Reyss A, Prioul JL, Bae JM, Lee JY.
Hirose T, Terao T.
Hizukuri S.
Hussain H, Mant A, Seale R, et al.
James MG, Denyer K, Myers AM.
Jiang H, Dian W, Liu F, Wu P.
Johnson PE, Patron NJ, Bottrill AR, Dinges JR, Fahy BF, Parker ML, Waite DN, Denyer K.
Kawagoe Y, Kubo A, Satoh H, Takaiwa F, Nakamura Y.
Kubo A, Fujita N, Harada K, Matsuda T, Satoh H, Nakamura Y.
Kubo A, Rahman S, Utsumi Y, et al.
Lu Y, Sharkey TD.
Mizuno K, Kawasaki T, Shimada H, Satoh H, Kobayashi E, Okumura S, Arai Y, Baba T.
Mizuno K, Kobayashi E, Tachibana M, Kawasaki T, Fujimura T, Funane K, Kobayashi M, Baba T.
Morell MK, Blennow A, Kosar-Hashemi B, Samuel MS.
Mutisya J, Sathish P, Sun C, Andersson L, Ahlandsberg S, Baguma Y, Palmqvist S, Odhiambo B, Åman P, Jansson C.
Nakamura Y.
Nakamura Y.
Nakamura Y, Francisco Jr PB, Hosaka Y, Sato A, Sawada T, Kubo A, Fujita N.
Nakamura Y, Umemoto T, Ogata N, Kuboki Y, Yano M, Sasaki T.
Nakamura Y, Yuki K.
Nakamura Y, Yuki K, Park S, Ohya T.
Nishi A, Nakamura Y, Tanaka N, Satoh H.
Okagaki RJ.
Preiss J, Sivak M.
Satoh H, Nishi A, Fujita N, Kubo A, Nakamura Y, Kawasaki T, Okita TW.
Satoh H, Nishi A, Yamashita K, Takemoto Y, Tanaka Y, Hosaka Y, Sakurai A, Fujita N, Nakamura Y.
Shimomura S, Nagai M, Fukui T.
Sikka VK, Choi SB, Kavakli IH, Sakulsingharoj C, Gupta S, Ito H, Okita TW.
Smith SM, Fulton DC, Chia T, Thorneycroft D, Chapple A, Dunstan H, Hylton C, Zeeman SC, Smith AM.
Smith-White BJ, Preiss J.
Steup M.
Sun C, Sathish P, Ahlandsberg S, Jansson C.
Suzuki K, Hattori A, Tanaka S, Masumura T, Abe M, Kitamura S.
Tanaka N, Fujita N, Nishi A, Satoh H, Hosaka Y, Ugaki M, Kawasaki S, Nakamura Y.
Tetlow IJ, Davies EJ, Vardy KA, Bowsher CG, Burrell MM, Emes MJ.
Tetlow IJ, Wait R, Lu Z, Akkasaeng R, Bowsher CG, Esposito S, Kosar-Hashemi B, Morell MK, Emes MJ.
Thompson DB.
Thorbjørnsen T, Villard P, Denyer K, Olsen O, Smith AM.
Thorbjørnsen T, Villard P, Kleczkowski LA, Olsen OA.
Tomlinson K, Lloyd JR, Smith AM.
Umemoto T, Yano M, Shomura A, Satoh H, Nakamura Y.
Yamanouchi H, Nakamura Y.
Comments