JXB Advance Access originally published online on April 4, 2006
Journal of Experimental Botany 2006 57(7):1591-1602; doi:10.1093/jxb/erj156
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
RESEARCH PAPER |
Proteome of amyloplasts isolated from developing wheat endosperm presents evidence of broad metabolic capability*
1Department of Plant and Microbial Biology, University of California, 111 Koshland Hall, Berkeley, CA 94720, USA
2US Department of Agriculture, Agricultural Research Service, Western Regional Research Center, Albany, CA 94710, USA
To whom correspondence should be addressed. E-mail: view{at}nature.berkeley.edu
Received 10 April 2005; Accepted 7 February 2006
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and methodsResults and discussionConcluding remarksSupplementary dataNote added in proofReferences |
|---|
By contrast to chloroplasts, our knowledge of amyloplasts—organelles that synthesize and store starch in heterotrophic plant tissues—is in a formative stage. While our understanding of what is considered their primary function, i.e. the biosynthesis and degradation of starch, has increased dramatically in recent years, relatively little is known about other biochemical processes taking place in these organelles. To help fill this gap, a proteomic analysis of amyloplasts isolated from the starchy endosperm of wheat seeds (10 d post-anthesis) has been conducted. The study has led to the identification of 289 proteins that function in a range of processes, including carbohydrate metabolism, cytoskeleton/plastid division, energetics, nitrogen and sulphur metabolism, nucleic acid-related reactions, synthesis of various building blocks, protein-related reactions, transport, signalling, stress, and a variety of other activities grouped under ‘miscellaneous’. The function of 12% of the proteins was unknown. The results highlight the role of the amyloplast as a starch-storing organelle that fulfills a spectrum of biosynthetic needs of the parent tissue. When compared with a recent proteomic analysis of whole endosperm, the current study demonstrates the advantage of using isolated organelles in proteomic studies.
Key words: Amyloplast proteins, amyloplast proteome, dithiothreitol, endosperm, Global Proteome Machine, GPM, isolated amyloplasts, membranes, wheat
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and methodsResults and discussionConcluding remarksSupplementary dataNote added in proofReferences |
|---|
Although known for many years, our understanding of amyloplasts, plant organelles functional in the synthesis and storage of starch in heterotrophic plant tissues, remains in its infancy. Aside from pathways leading to the synthesis and breakdown of starch, relatively little is known about the biochemistry of this organelle (Neuhaus and Emes, 2000). In their recent proteomic identification of 171 proteins in amyloplasts isolated from wheat starchy endosperm, Andon et al. (2002) provided a foundation for understanding processes taking place in the plastid. According to their work, however, the biochemical activities of amyloplasts seem relatively restricted; 85% of the proteins fell under the protein destination/storage, energy metabolism, and unknown categories. Thinking that, by analogy with chloroplasts (Kleffmann et al., 2004; van Wijk, 2004) and their etioplast relatives (von Zychlinski et al., 2005), amyloplasts should have broad metabolic capability, the question of the nature of their resident proteins was reopened. Using amyloplasts, also isolated from developing wheat endosperm, 289 proteins were identified. Of these, less than one-third fall in the protein destination/storage, energy metabolism, and unknown categories and more than half function in metabolism and response to stress. Particularly prominent are enzymes of amino acid, nucleic acid, and sulphur metabolism. In demonstrating the versatility of amyloplasts, the results add to our understanding of plastid function, and, by building on findings obtained with whole endosperm (Vensel et al., 2005), show the advantage of using isolated organelles for proteomic analysis.
|
Materials and methods |
|---|
|
TopAbstractIntroductionMaterials and methodsResults and discussionConcluding remarksSupplementary dataNote added in proofReferences |
|---|
Materials
Wheat (Triticum aestivum L. cv. Butte) was grown in a climate-controlled greenhouse under a 16 h light (supplemented with 100 W sodium lamps)/8 h dark, day/night regimen that had a maximum daytime temperature of 24 °C and a minimum night-time temperature of 17 °C (Altenbach et al., 2003). Water and fertilizer (Plantex 20-20-20, 500 ml of 0.6 g l–1 pot–1 d–1) were applied by drip irrigation. Heads were harvested 10 d post-anthesis (dpa) and used immediately for amyloplast preparation.
Amyloplast preparation
Intact amyloplasts were isolated using the procedure of Tetlow et al. (1993) as summarized in Fig. 1. Twenty heads were harvested and used within a 2 h period for each preparation. Embryos were cut from the grain, and endosperm was squeezed through the opening created and collected in ice-cold buffer (0.5 M sorbitol, 50 mM HEPES pH 7.5). The endosperm was transferred to plasmolysis buffer (0.8 M sorbitol, 50 mM HEPES pH 7.5, 1 mM EDTA, 1 mM KCl, 2 mM MgCl2) and incubated for 1 h on ice. Plasmolysed endosperm was chopped twice for 30 s with an electric knife, the blades of which were replaced with holders fitted with single-edge razor blades. The resulting homogenate was filtered through two layers of Miracloth and gently pipetted onto a 4 ml cushion of 2% Nycodenz (Nycomed, Oslo, Norway) in plasmolysis buffer in a 15 ml conical tube containing a 2 ml 1% agar pad at the bottom. Following centrifugation (30 g, 10 min, 4 °C) (Centrifuge 5810 R, Eppendorf, Westbury, NY, USA), the supernatant fraction was removed by aspiration and discarded. The pellet containing the amyloplasts was gently suspended in plasmolysis buffer and the Nycodenz procedure repeated once more. The purity of the amyloplast preparation was assessed previously by the group who devised the protocol (Tetlow et al., 1998). They found minimal contamination of amyloplast preparations by other cell components. Based on an assay of marker enzymes, mitochondria (<0.2%) were the major source of contamination. This conclusion was confirmed in the present study as small amounts of cytochrome oxidase (Table 1), one of the four classical mitochondrial marker proteins (Quail, 1979) were found, but the other three (fumarase, succinate, and cytochrome c reductase) were not detected.
|
|
Isolation of amyloplast proteins
The pellet containing intact amyloplasts was suspended in a small volume of plasmolysis buffer without sorbitol and supplemented with a protease inhibitor cocktail (Complete Mini; Roche, Basel, Switzerland). The suspension was frozen in liquid nitrogen and thawed three times to break the intact organelles. The lysate was centrifuged (10 000 g, 20 min, 4 °C) to collect the soluble proteins (supernatant fraction). The pellet, containing starch granules and other insoluble materials, was extracted with the same buffer plus 2% Triton X-100 and centrifuged to collect the membrane proteins (supernatant fraction). The supernatant solution of this second extraction yielded a fraction enriched in membrane proteins. Four volumes of cold acetone were added to each fraction, and following incubation overnight (–20 °C), precipitated proteins were recovered by centrifugation (Wong et al., 2004).
Two-dimensional electrophoresis (2-DE)
Isoelectric focusing and SDS/PAGE were performed using the systems of Invitrogen Corp. (Carlsbad, CA, USA) and Bio-Rad Laboratories (Hercules, CA, USA), according to the manufacturers' instructions. Proteins were solubilized in a solution containing 7 M urea, 2 M thiourea, 0.5% ampholytes, 2% ß-dodecyl maltoside, and 10 mM dithiothreitol (Luche et al., 2003). Isoelectric focusing was carried out using an IPG strip with a non-linear pH range of 3–10 (Invitrogen Corp. or Bio-Rad Laboratories, depending on the 2-D system used). The second dimension was developed with a NuPage 4–12% BIS-TRIS Zoom gel (Invitrogen Corp.) or a Criterion Precast gel (Bio-Rad Laboratories). Gels were stained with Coomassie brilliant blue G-250 (Kasarda et al., 1998).
Protein spot excision and digestion
Gels were scanned (Powerlook III, Umax) and the spots detected with the Progenesis software package (Nonlinear Dynamics Ltd, Newcastle upon Tyne, UK). Gels were transferred to a ProPic gel spot picker (Genomic Solutions, Ann Arbor, MI, USA) that excised and placed spots into 96-well reaction plates for subsequent in-gel tryptic digestion with an automated protein digester (DigestPro, Intavis, Langenfeld, Germany). The DigestPro was programmed to destain the gel piece and carry out reduction with dithiothreitol, alkylation with iodoacetamide, enzymatic digestion with trypsin, and elution of the generated peptides into a 96 well plate that was subsequently inserted into the autosampler of the mass spectrometer.
LC/MS/MS of tryptic peptides of proteins
A QSTAR PULSAR i quadrupole time-of-flight (TOF) mass spectrometer (Applied Biosystems/MDS Sciex, Toronto, Canada) equipped with a Proxeon Biosystems (Odense, Denmark) nano-electrospray source was used to perform ESI-MS of the tryptic peptides. HPLC peptide separation was carried out using a nano-flow liquid chromatograph (LC Packings/Dionex, Sunnyvale, CA, USA). The 96 well plate from the DigestPro was covered with a self-sealing silicone compression mat (catalogue no. CM-96-EXP; Axygen Scientific, Union City, CA, USA) and cooled to 10 °C to prevent sample evaporation. Aliquots (15 µl) of sample from the DigestPro 96 well receiving tray were loaded into a loop using an autosampler (FAMOSTM, LC Packings/Dionex) and pumped at a flow rate of 20 µl min–1 onto a C18, 5 µm, 300 A, Nano-PrecolumnTM (300 µm i.d. x 1 mm, P/N 160459, LC Packings/Dionex) with a loading pump (SWITCHOSTM, LC Packings/Dionex). The loading pump eluent was 0.5% acetic acid, 0.02% heptafluorobutyric acid in HPLC-grade water. After 2 min, the loading valve was switched to place the trap in-line with the LC pump (UltimateTM, LC Packings/DIONEX). Samples were eluted from the trap onto the C-18 monomeric, VydacTM EVERESTTM column (catalogue number 238EV5.07515; W.R. Grace & Co.) at a rate of 200–250 nl min–1. The spray tip was an 8 µm i.d. PicoTipTM emitter (catalogue no. FS360-75-8-CE-20, New Objective, Woburn, MA, USA). Spray voltage was maintained at 1800 V. A co-axial counter-current flow of ultra-high purity nitrogen (curtain gas) was used to shield the orifice of the mass spectrometer and contributed to charged droplet desolvation. Mobile phase A was 0.5% glacial acetic acid (Fisher Scientific, reagent grade) diluted in HPLC-grade water (Burdick and Jackson, Muskegon, MI, USA). Analytical column elution solvents were labelled A (0.5% acetic acid) and B (80% acetonitrile, 0.5% acetic acid). Samples were eluted with the following gradient profile: 8% B at 0 min to 80% B by 12 min through 13 min to 8% B by 14 min continuing at 8% B to 28 min.
The TOF mass analyser of the instrument was calibrated using the 187.0713 and 1245.5444 m/z fragment ions from the CID of the +2 charge state (m/z 785.8) of glufibrogen. Data were acquired using the IDA acquisition mode of Analyst QS software; from an initial survey scan of mass range m/z 400–1500, the most abundant double or triple charged ion above a threshold of 20 counts was selected for fragmentation. The quadrupole mass filter (Q1) was adjusted so that only ions of the selected m/z entered the collision cell of Q2. CID of the mass-selected ion in the collision cell of Q2 was carried out using ultra-high purity nitrogen as the collision gas. Analysis of the fragment ions by the TOF mass analyser was set to range of 70–2000 m/z. The precursor ion was precluded from further MS/MS experiments. An IDA script was used to determine the optimum collision energy for each precursor ion. Following the 3 s MS/MS fragmentation period, the MS survey scan was repeated until another MS/MS period was triggered.
Database analysis and identification of proteins
The WIFF acquisition files created by Analyst QS were converted to DTA files using a WIFF-to-DTA converter (Genomic Solutions). The software used to analyse the DTA files was obtained from the Global Proteome Machine (GPM) organization (http://www.thegpm.org/). A local installation of the GPM open-source software was used for visualization and analysis of the data. The spectrum modeler X! TANDEM (Fenyo and Beavis, 2003; Craig and Beavis, 2004) that is a part of the GPM software was used to match MS/MS fragmentation data to peptide sequences. The DTA files from each fraction were joined together into a single file and submitted to the locally installed copy of X! Tandem (version 2005.06.01.2
[EC]
) using the JAVA program XtandemAppend (authored by Jayson Falkner and posted to the LiveCD Project at http://www.thegpm.org/). The scripts were modified to process DTA files and recompiled on Windows XP. The resulting single DTA file for each fraction was searched against a flat file containing amino acid sequences of all plant proteins in the HarvEST:Wheat version 1.04 (http://harvest.ucr.edu/), NCBI non-redundant green plant database, NCBI Triticum aestivum: UniGene Build No. 37, and wEST Database (http://wheat.pw.usda.gov/wEST) (Lazo et al., 2004). The translation of the nucleotide sequences into amino acid sequences in all six reading frames was carried out using the Knexus suite of software (Genomics Solutions).
|
Results and discussion |
|---|
|
TopAbstractIntroductionMaterials and methodsResults and discussionConcluding remarksSupplementary dataNote added in proofReferences |
|---|
Identification of proteins
Amyloplasts were resolved into soluble and membrane fractions, proteins separated by 2-DE, and gel spots analysed by LC/MS/MS. Three gels, including one overloaded to visualize proteins of low abundance, were developed for the soluble and insoluble protein fractions. In this way, >600 protein spots were analysed and 85 218 usable spectra obtained. The mass spectra fragment ion data were searched against two sets of databases—NCBI non-redundant and a wheat EST database combining HarvEST, Unigene, and wEST. The search against the NCBI database assigned 7037 spectra yielding 1001 identifications with an expectation value of <10–3; the search against the EST database assigned 9162 spectra for 1053 accepted identifications. These 2054 identifications were reduced to 289 unique proteins after exclusion of redundant and closely related homologues. The peptide sequences identified for each of the 289 proteins listed in Table 1 are included in the supplementary Table 1 (see www.jxb.oxfordjournals.org). Surprisingly, out of the 171 proteins identified in amyloplasts by Andon et al. (2002), only 31 were identified in the present study. The limited overlap between these two similar preparations could stem from several factors. First, the endosperm for the preparations in the present study was collected at 10 dpa and that for Andon et al. (2002) at 12 dpa. As the endosperm is a rapidly developing tissue, collection time, as well as growth regimen, could account for part of the difference observed. Secondly, the isolation procedures were slightly different. In the present study, the procedure followed was that described for wheat amyloplasts (Tetlow et al., 1993), whereas Andon et al. (2002) used a protocol developed for amyloplasts from maize (Neuhaus et al., 1993). Finally, since 2002, the wheat sequence database has been greatly increased, therefore enhancing the efficiency of protein identification by MS [see http://harvest.ucr.edu/ and Lazo et al. (2004) for more information].
Purity of preparation
Identified proteins were determined to be of plastid origin on the basis either of their designation in the SwissProt database or as a result of a search on PubMed. On this basis, about 10% of the proteins discussed below appear not to be of plastid origin—the most obvious being the storage proteins (e.g. gliadins and alpha-amylase/trypsin inhibitors) housed in protein bodies that may co-purify with amyloplasts. In this connection, a subsequent extraction of the insoluble fraction of the amyloplast preparation with the detergent Triton X-100 yielded a membrane-enriched fraction (M fraction in Table 1) which contained a high level of non-plastidial proteins such as storage proteins. Further, certain proteins in the preparations in the present study, such as those associated with ribosomes, possibly adhered to starch granules released from broken amyloplasts that co-purify with intact amyloplasts. Indeed, there is evidence that starch granules bind endosperm proteins in a non-specific manner when the tissue is homogenized (F DuPont, W Hurkman, D Kasarda, unpublished observations). The presence of members of the citric acid cycle raises the question of mitochondrial contamination. While possible for certain entries, one of these, fumarate hydratase 2, is annotated as plastid targeted in the SwissProt database. Several proteins functional in the synthesis and transformation of ATP, for example, ATP synthase subunits, and electron transport were also identified. Whereas certain of these could be mitochondrial contaminants, most are typical of chloroplasts. Accordingly, low levels of these proteins may be present in amyloplasts owing to a lack of strict control over their expression such as seems to occur for a number of photosynthetic proteins (e.g. 33 kDa oxygen-evolving enhancer protein, chlorophyll a/b-binding protein, and Rubisco large subunit) (Table 1). In short, while certain of the proteins in question are likely to be contaminants, the relevance of others to amyloplasts remains unclear.
Soluble versus membrane-enriched fractions
The isolated amyloplasts were subfractionated by centrifugation into soluble and insoluble fractions. The insoluble fraction was extracted with buffer containing the non-ionic detergent, Triton X-100, yielding a fraction enriched in membrane proteins. Analysis of the data showed that, of the 289 proteins identified, 87 were present in both fractions, 119 were unique to the soluble fraction, and 83 to the membrane-enriched fraction (Table 1). Fractionation of the amyloplast proteins into soluble and membrane-enriched fractions allowed the identification of low-abundance proteins that were not identified previously in the endosperm. The majority of these were soluble proteins that function in carbohydrate metabolism, nitrogen/sulphur metabolism, nucleotide biosynthesis, and protein assembly/folding and turnover. There were also many integral or membrane-bound proteins unique to the membrane-enriched fraction, including proteins that function in energetics (ATPase subunits), electron transport (oxygen-evolving enhancer), transport (porins), and signalling (GTP-binding proteins). However, the membrane-enriched fraction also contained a number of proteins that are not known to be associated with membranes. As would be expected, granule-bound starch synthase was recovered exclusively in the membrane-enriched fraction because of the insolubility of the starch granules. Storage proteins were also found in this fraction owing to limited solubility in the extraction buffer. Other non-membrane proteins may be present in this fraction as they bind non-specifically to the starch granules during amyloplast breakage (see above).
Origin of the amyloplast proteins
To estimate the relative proportion of proteins of plastid and nuclear origin, the 289 proteins identified in the present study were compared with the list of known plastid-encoded proteins in plants published on the HAMAP proteome website (http://www.expasy.org/sprot/hamap/plastid.html). Seventy-six genes are described in the wheat plastid genome, whereas rice plastids encode 94 polypeptides. Most of these proteins are participants in the photosynthetic electron chain and the biogenesis/assembly of its components (40 in wheat) or in plastid protein translation (22 in wheat). As amyloplasts are organelles without photosynthetic activity, it is not surprising that few plastid-encoded proteins were identified. Indeed, only three such proteins were detected in the amyloplast preparation (ATP synthase alpha and beta chains and Rubisco large subunit). It is of interest that no subunits of the prokaryotic-type ribosome were identified, suggesting that in organello protein synthesis is limited in amyloplasts. This observation agrees with early evidence showing that only a small number of plastid-encoded genes are required in non-green plastids (Harris et al., 1994).
Function of proteins
Table 1 lists the 289 unique proteins categorized according to function, SwissProt accession number, and log of the expectation value. The proteins are grouped in 12 major categories: carbohydrate metabolism, cytoskeleton/division, energetics, nitrogen/sulphur metabolism, nucleic acid-related, building blocks (other than specified), protein-related, transport, signalling, stress-related, miscellaneous, and unknown. In certain cases, subcategories were devised for clarity: carbohydrate metabolism (starch, glycolysis, pentose phosphate pathway, citric acid cycle, malate valve), energetics (ATP/PP synthesis/transformation, electron transport), nitrogen/sulphur metabolism (amino acid, sulphate assimilation), nucleic acid-related (DNA/RNA, nucleotide biosynthesis), building blocks (isoprenoid, tetrapyrrole, vitamin/cofactor, lipid metabolism), protein-related (assembly/folding, turnover, storage), transport (membrane, ion), signalling (hormone, phosphorylation, GTP-linked, other), stress-related (thiol-linked, ascorbate-linked, other).
A graphical view of the functional distribution of the amyloplast proteins from 10 dpa wheat endosperm highlights four biochemical processes—carbohydrate and nitrogen/sulphur metabolism, nucleic acid- and protein-related reactions—that together comprised 
50% of the identifications (Fig. 2). Twelve per cent of the proteins have no known function, and, although another 4% have defined activity, they do not fit into one of the present categories and are thus placed in the miscellaneous group. In the following section, certain processes are highlighted, especially with respect to relevance to developing cereal endosperm.
|
Carbohydrate metabolism
Among all the biochemical reactions of the amyloplast, the biosynthesis of starch is generally considered the major activity, since the newly formed starch granule will ultimately occupy the bulk of the organelle. Accordingly (Table 1), most of the enzymes of starch biosynthesis (Smith, 2001; James et al., 2003) were identified in the preparation described here. However, somewhat surprisingly, counterparts catalysing the degradation or rearrangement of starch, such as
-1,4-glucan phosphorylase (Smith et al., 2005) were also found along with those associated with glycolysis, the oxidative pentose phosphate pathway, and malate valve. The presence of these components suggests that a significant part of the carbon imported by amyloplasts is diverted to the production of reducing power and ATP (Neuhaus and Emes, 2000) to support myriad biosynthetic reactions taking place within the organelle (more below). The overall distribution of enzymes of carbohydrate metabolism is in accord with the conclusion that, at this early stage of development (10 dpa), the plastid actively catalyses processes in addition to starch synthesis (Vensel et al., 2005).
Cytoskeleton/division
Five of the proteins identified were assigned a role in determining plastid ultrastructure and division. Of these, two (actin 42 and tubulin) are problematic and, while assumed to be of plastid origin, could originate from the cytosol (for a discussion of this point, see Kleffmann et al., 2004). On the other hand, two of plastid origin, Arc6 and FtsZ, provide evidence that endosperm amyloplasts are actively dividing at 10 dpa (Marrison et al., 1999; Vitha et al., 2001).
Energetics
A number of proteins found were involved in electron transport and in the synthesis and transformation of ATP. The function of many of these is unclear, i.e. components associated with photosynthetic electron transport of chloroplasts such as the 33 kDa oxygen-evolving enhancer and chlorophyll a/b-binding proteins. By contrast, the role of others is obvious, for example, the ferredoxin and ferredoxin/NADP reductase isoforms specific to non-photosynthetic plastids catalyse the reverse transfer of reducing equivalents from NADPH to ferredoxin in order to support processes that in chloroplasts are driven by light (Onda et al., 2000). The finding of ferredoxin-thioredoxin reductase suggests the presence of the ferredoxin/thioredoxin system described for chloroplasts. However, the inability to detect thioredoxin in wheat amyloplast preparations in the present study shows the need for further work.
Nitrogen/sulphur metabolism
The plastid is the site of nitrogen and sulphur assimilation as well as most of the carbon skeletons of the amino acids (Neuhaus and Emes, 2000; Hofgen et al., 2001). The current finding of a large number of proteins (39) functional in the synthesis of amino acids is consistent with these activities and with the large quantity of these building blocks required for the synthesis of enzymes and storage proteins. At 10 dpa the endosperm is entering an extended phase of starch accumulation and storage protein biosynthesis (Altenbach et al., 2003). It appears that the plastids play an essential role in both processes.
Nucleic acid-related proteins
Two different groups of proteins linked to nucleic acids were identified in amyloplast extracts. Most are involved in the synthesis of DNA and RNA and in transcription and translation. Surprisingly, the eight ribosome subunits identified are of eukaryotic origin (Table 1). While the source of contamination is unclear, it is noted that Andon et al. (2002) obtained similar results. The inability in the present study to detect subunits of the 70S ribosome indicates that most proteins needed by amyloplasts are encoded in the nucleus (Leister, 2003; Jarvis and Robinson, 2004). The second subcategory related to nucleic acids comprises enzymes of nucleotide—purine and pyrimidine—biosynthesis. While the de novo synthesis of nucleotides is still poorly characterized in plants, the evidence suggests that most of the reactions are localized to plastids (Boldt and Zrenner, 2003). In the present study, the presence of nine enzymes of nucleic acid biosynthesis was confirmed: six members of the purine and three of the pyrimidine biosynthesis pathways.
Other building blocks
The metabolic versatility of amyloplasts is reflected in the large number of enzymes active in the biosynthesis of building blocks not discussed above: isoprenoids (four members), tetrapyrroles (seven), vitamins and cofactors (seven), and lipids (eight). Plants have two pathways for the synthesis of isoprenoids—one cytosolic and the other plastidic (Lichtenthaler, 1999). As expected, each of the four enzymes listed in Table 1 is a member of the plastid pathway (also called the glyceraldehyde 3-phosphate/pyruvate pathway). The synthesis of tetrapyrroles via the C5 pathway is also a process that takes place entirely in plastids. Six members of this pathway were identified: glutamate semialdehyde aminomutase, delta-aminolevulinic acid dehydratase, porphobilinogen deaminase, uroporphyrinogen decarboxylase, coproporphyrinogen oxidase, and ferrochelatase. One enzyme of haem degradation, haemoxygenase, was also found. Enzymes linked to the synthesis of chlorophyll were not detected, in agreement with the fact that, unlike chloroplasts, amyloplasts do not require photosynthetic pigments. Further, by contrast to the chloroplast electron carriers discussed above, the formation of enzymes active in chlorophyll biosynthesis appear to be under strict control by light.
In another category, enzymes needed for the biosynthesis of several vitamins and cofactors were found in the preparations, confirming the need of plastids for their biosynthesis. Finally, eight members functional in the de novo synthesis of fatty acids—an energy demanding pathway—were also identified (Rawsthorne, 2002). This finding is consistent with the need of plastids to satisfy the fatty acid requirement for the cell. The identification of members of these different biosynthetic pathways is in keeping with the idea that, at this early stage of development, the amyloplast functions as a factory that supplies essential metabolites and building blocks to the parent tissue.
Protein-related
This category, which includes components active in the folding, assembly, turnover, and storage of proteins is one of the major groups found in amyloplasts. Interestingly, the number of proteases (13) is equivalent to the number of chaperones (14), thus suggesting a rapid turnover of the plastid protein components—a feature in agreement with the dynamics of grain development (Vensel et al., 2005). As discussed above, the presence of several storage proteins residing in protein bodies (e.g. gliadins, amylase-trypsin/chymotrypsin inhibitors) is probably due to trace contaminants in the amyloplast preparations. Similar likely contaminants were reported by Andon et al. (2002).
Transport
Membrane transporters are critical not only for the import of metabolites, but also for the export of amyloplast products (Fischer and Weber, 2002). In wheat endosperm, one of the major amyloplast transporters is the ADP-glucose translocator that imports ADP-glucose formed in the cytosol (Shannon et al., 1998). In addition, 18 proteins essential for various pores, channels, and translocators were detected, reflecting the complexity of the transport machinery of the plastid envelope. Several proteins active in the sequestration of ions were also identified, in support of a role for amyloplasts in ion storage (e.g. ferritin and calreticulin).
Signalling
Proteins linked to hormones, phosphorylation, GTP, and other signalling pathways are well represented in 10 dpa amyloplasts. These proteins (16 in all) are likely to function in the regulation of plastid development, as well as in the partitioning of resources such as carbon between different pathways. The amyloplast is obviously connected to a number of complex signalling networks in cereals as a result of its central role in endosperm development.
Stress-related
Relatively few stress-related proteins were detected in the preparations, probably due to the early stage of development (10 dpa) that precedes the onset of grain maturation and drying. Those proteins identified were linked either to thiols (peroxiredoxins and enzymes of glutathione biosynthesis) or ascobate (enzymes detoxifying reactive oxygen species). Also catalase and superoxide dismutase—enzymes functional in reactive oxygen species removal—were identified. Of the stress-related proteins identified, only one, sti1, is not directly redox linked owing to its role as a component of a chaperone multicomplex (Hernandez Torres et al., 1995; Wegele et al., 2003).
Miscellaneous and unknown
Eleven proteins that had either limited functional description or did not fit into the classification categories are included under ‘Miscellaneous’. The characterization of these proteins, together with those in the ‘Unknown’ category, awaits further study.
Comparison with whole endosperm proteome
In a recent study, Vensel et al. (2005) identified 250 proteins in the salt-soluble (metabolic protein) fraction of whole wheat endosperm by using a proteomic approach similar to that of the current study. That fraction would contain the soluble proteins present in amyloplasts. A comparison of those results with the ones currently described can, therefore, be used to illustrate the effect of fractionation. As seen in Table 1, 46 proteins distributed throughout the various functional categories are common to the two studies (these are identified with an asterisk in Table 1). Among these, 40 reside in amyloplasts (six appear to be extraplastidic). Based on this comparison, the use of isolated amyloplasts enhanced the identification of organellar proteins by 7-fold, thus illustrating the advantage of organelle isolation in proteomic analysis. In this case, the isolation step eliminated highly abundant extraplastidic proteins, thereby allowing the detection of organellar polypeptides not detected within the parent tissue. A case in point relates to enzymes of methionine biosynthesis. The current study led to the identification of a plastidic enzyme (cystathionine ß-lyase), while the earlier total endosperm work revealed a member of the pathway that could be either cytosolic or plastidic (Ravanel et al., 2004). Analysis of the isolated organelle also gives information on functional processes that is not obvious when examining the parent tissue (Fig. 2).
The endosperm and amyloplast studies also differ in the nature of the proteins identified (Fig. 2). Thus, with whole endosperm, a greater percentage of proteins was linked to quantitatively prominent processes such as carbohydrate metabolism, protein-related functions (including storage proteins), and stress response. By contrast, with amyloplasts, more proteins were affiliated with nitrogen/sulphur metabolism, nucleic acids, transport, and signalling. Also there was a much greater percentage of proteins of unknown function with amyloplasts. The observed distributions emphasize the importance of amyloplasts in fulfilling biosynthetic functions essential for the parent cell.
|
Concluding remarks |
|---|
|
TopAbstractIntroductionMaterials and methodsResults and discussionConcluding remarksSupplementary dataNote added in proofReferences |
|---|
The present findings show that amyloplasts, starch-storing organelles of heterotrophic plant tissues, have broad biosynthetic capability that is presumably required not only for their own growth and development, but also for that of parent cells. The results indicate, for example, that amyloplasts are endowed with enzymes catalysing the synthesis of amino acids, isoprenoids, fatty acids, and tetrapyrroles via pathways that, in leaves, are known to take place in chloroplasts. However, despite this general similarity, there appear to be quantitative differences in the distribution of enzymes in these organelles (compare A and B in Fig. 3). Notably, chloroplasts have, relative to amyloplasts (i) a larger number of unknown (33% versus 12%), and nucleic acid-related proteins (17% versus 11%), and (ii) a lower number of proteins related to other processes, for example, carbohydrate metabolism (2% versus 12%), energetics (2% versus 7%), transport (4% versus 8%), and nitrogen/sulphur metabolism (2% versus 15%). It remains to be seen whether these percentages are linked to species or developmental stage, or whether they reflect fundamental differences in the organelles. Nonetheless, the results suggest that overall plastids display unity of biosynthetic function in supplying essential building blocks, whether the parent cell is photosynthetic or heterotrophic. The question that emerges is how the biosynthetic processes are regulated in amyloplasts. There is extensive knowledge of the mechanisms operative in chloroplasts where light interfaces with several elements to modulate biochemical processes (Buchanan and Balmer, 2005). The corresponding knowledge for amyloplasts is, however, fragmentary. In filling this heterotrophic gap, future efforts should focus on amyloplast processes whose chloroplast counterparts are regulated by light.
|
|
Supplementary data |
|---|
|
TopAbstractIntroductionMaterials and methodsResults and discussionConcluding remarksSupplementary dataNote added in proofReferences |
|---|
Supplementary data can be found at JXB online.
|
Note added in proof |
|---|
|
TopAbstractIntroductionMaterials and methodsResults and discussionConcluding remarksSupplementary dataNote added in proofReferences |
|---|
While this manuscript was in process, our work demonstrating the presence of a complete ferredoxin/thioredoxin system in isolated amyloplasts was published together with evidence on the identification of thioredoxin-lined proteins present in the organelle (Y. Balmer et al., 2006, A complete ferredoxin-thioredoxin system regulates fundamental processes in amyloplasts. Proc. Natl. Acad. Sci. USA 103, 2988–2993).
|
Footnotes |
|---|
* Disclaimer: The mention of a trademark or proprietary product does not constitute a guarantee or warranty of the product by the United States Department of Agriculture and does not imply its approval to the exclusion of other products that may be suitable.
|
References |
|---|
|
TopAbstractIntroductionMaterials and methodsResults and discussionConcluding remarksSupplementary dataNote added in proofReferences |
|---|
, , , . . . , –.Temperature, water and fertilizer influence the timing of key events during grain development in a US spring wheat. Journal of Cereal Science (2003) 37:9–20.
, , , , , . . . , –.Proteomic characterization of wheat amyloplasts using identification of proteins by tandem mass spectrometry. Proteomics (2002) 2:1156–1168.[CrossRef][Web of Science][Medline]
, . . . , –.Purine and pyrimidine biosynthesis in higher plants. Physiologia Plantarum (2003) 117:297–304.[CrossRef][Medline]
, . . . , –.Redox regulation: a broadening horizon. Annual Review of Plant Biology (2005) 56:187–220.[CrossRef][Medline]
, . . . , –.TANDEM: matching proteins with tandem mass spectra. Bioinformatics (2004) 20:1466–1467.
, . . . , –.A method for assessing the statistical significance of mass spectrometry-based protein identifications using general scoring schemes. Analytical Chemistry (2003) 75:768–774.[Medline]
, . . . , –.Transport of carbon in non-green plastids. Trends in Plant Science (2002) 7:345–351.[CrossRef][Web of Science][Medline]
, , , , , , . . . , –.In-depth analysis of the thylakoid membrane proteome of Arabidopsis thaliana chloroplasts; new proteins, functions and a plastid proteome database. The Plant Cell (2004) 16:478–499.
, , . . . , –.Chloroplast ribosomes and protein synthesis. Microbiology Reviews (1994) 58:700–754.
, , . . . , –.Isolation and characterization of gmsti, a stress-inducible gene from soybean (Glycine max) coding for a protein belonging to the TPR (tetratricopeptide repeats) family. Plant Molecular Biology (1995) 27:1221–1226.[CrossRef][Web of Science][Medline]
, , , . . . , –.Manipulation of thiol contents in plants. Amino Acids (2001) 20:291–299.[CrossRef][Web of Science][Medline]
, , . . . , –.Starch synthesis in the cereal endosperm. Current Opinion in Plant Biology (2003) 6:215–222.[CrossRef][Web of Science][Medline]
, . . . , –.Mechanisms of protein import and routing in chloroplasts. Current Biology (2004) 14:R1064–1077.[CrossRef][Web of Science][Medline]
, , . . . , –.Resolution of high molecular weight glutenin subunits by a new SDS-PAGE system incorporating a neutral buffer. Cereal Chemistry (1998) 75:70–71.
, , , , , , . . . , –.The Arabidopsis thaliana chloroplast proteome reveals pathway abundance and novel protein functions. Current Biology (2004) 14:354–362.[CrossRef][Web of Science][Medline]
, , , et al. . . , –.Development of an expressed sequence tag (EST) resource for wheat (Triticum aestivum L.): EST generation, unigene analysis, probe selection and bioinformatics for a 16 000-locus bin-delineated map. Genetics (2004) 168:585–593.
. . . , –.Chloroplast research in the genomic age. Trends in Genetics (2003) 19:47–56.[CrossRef][Web of Science][Medline]
. . . , –.The 1-deoxy-D-xylulose-5-phosphate pathway of isoprenoid biosynthesis in plants. Annual Review of Plant Physiology and Plant Molecular Biology (1999) 50:47–65.[CrossRef][Web of Science]
, , . . . , –.Evaluation of nonionic and zwitterionic detergents as membrane protein solubilizers in two-dimensional electrophoresis. Proteomics (2003) 3:249–253.[CrossRef][Web of Science][Medline]
, , , , , . . . , –.The distinctive roles of five different ARC genes in the chloroplast division process in Arabidopsis. The Plant Journal (1999) 18:651–662.[CrossRef][Web of Science][Medline]
, , , . . . , –.Purification of highly intact plastids from various heterotrophic plant tissues: analysis of enzymic equipment and precursor dependency for starch biosynthesis. Biochemical Journal (1993) 296:395–401.
, . . . , –.Nonphotosynthetic metabolism in plastids. Annual Review of Plant Physiology and Plant Molecular Biology (2000) 51:111–140.[CrossRef][Web of Science]
, , , , , . . . , –.Differential interaction of maize root ferredoxin:NADP(+) oxidoreductase with photosynthetic and non-photosynthetic ferredoxin isoproteins. Plant Physiology (2000) 123:1037–1045.
. . . , –.Plant subcellular fractions. Annual Review of Plant Physiology (1979) 30:424–484.
. . . , –.Carbon flux and fatty acid synthesis in plants. Progress in Lipid Research (2002) 41:182–196.[CrossRef][Web of Science][Medline]
, , , , , , . . . , –.Methionine metabolism in plants: chloroplasts are autonomous for de novo methionine synthesis and can import S-adenosylmethionine from the cytosol. J Biol Chem (2004) 279:22548–57.
, , , . . . , –.Brittle-1, an adenylate translocator, facilitates transfer of extraplastidial synthesized ADP-glucose into amyloplasts of maize endosperms. Plant Physiology (1998) 117:1235–1252.
. . . , –.The biosynthesis of starch granules. Biomacromolecules (2001) 2:335–341.[CrossRef][Web of Science][Medline]
, , . . . , –.Starch degradation. Annual Review of Plant Biology (2005) 56:73–98.[CrossRef][Medline]
, , . . . , –.A rapid method for the isolation of purified amyloplasts from wheat endosperm. Planta (1993) 189:597–600.[Web of Science]
, , . . . , –.Metabolite pools during starch synthesis and carbohydrate oxidation in amyloplasts isolated from wheat endosperm. Planta (1998) 204:100–108.[CrossRef][Web of Science]
. . . , –.Plastid proteomics. Plant Physiology and Biochemistry (2004) 42:963–977.[CrossRef][Web of Science][Medline]
, , , , , . . . , –.Developmental changes in the metabolic protein profiles of wheat endosperm. Proteomics (2005) 5:1594–1611.[CrossRef][Web of Science][Medline]
, , . . . , –.FtsZ ring formation at the chloroplast division site in plants. The Journal of Cell Biology (2001) 153:111–120.
, , , , , . . . , –.Proteome analysis of the rice etioplast: metabolic and regulatory networks and novel protein functions. Molecular and Cellular Proteomics (2005) 4:1072–1084.
, , , . . . , –.Sti1 is a novel activator of the Ssa proteins. Journal of Biological Chemistry (2003) 278:25970–25976.
, , , , , . . . , –.Thioredoxin reduction alters the solubility of proteins of wheat starchy endosperm: an early event in cereal germination. Plant and Cell Physiology (2004) 45:407–415.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  J. de Dios Barajas-Lopez, A. J. Serrato, A. Olmedilla, A. Chueca, and M. Sahrawy Localization in Roots and Flowers of Pea Chloroplastic Thioredoxin f and Thioredoxin m Proteins Reveals New Roles in Nonphotosynthetic Organs Plant Physiology, November 1, 2007; 145(3): 946 - 960. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. Lopez-Juez Plastid biogenesis, between light and shadows J. Exp. Bot., January 1, 2007; 58(1): 11 - 26. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. A. Siddique, J. Grossmann, W. Gruissem, and S. Baginsky Proteome Analysis of Bell Pepper (Capsicum annuum L.) Chromoplasts Plant Cell Physiol., December 1, 2006; 47(12): 1663 - 1673. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||