A Short Peptide in Rice Glutelin Directs Trafficking of Protein into the Protein Storage Vacuoles of the Endosperm Cells

Several protein vacuolar sorting determinants (VSDs) have been identified in higher plants. Glutelin as a major storage protein in rice endosperm cells is transported to a protein storage vacuole (PSV). How glutelin sort to PSV and the mechanism of the intracellular trafficking has remained unknown. Here, a sequence-specific vacuolar sorting determinant (ssVSD) is identified by serial deletions of rice glutelin and its role in the protein-sorting process analyzed by transgenic approaches and transient assays. The ssVSD consists of six residues (QRLKHN) within the β-subunit of glutelin is sufficient to direct the glutelin to the protein body II in the rice endosperm cells. We found that protein-sorting via the ssVSD takes place by a ～680-kDa sorting complex containing the receptor Oryza sativa receptor-like membrane Ring-H2 3 (OsRMR3). Further study indicated that OsRMR3 and the ssVSD are essential for glutelin trafficking. Furthermore, site-directed mutagenesis showed that the leucine residues in the ssVSD are critical for protein sorting.     

The C-terminal region of α′ subunit of soybean β-conglycinin contains two types of vacuolar sorting determinants

In maturing seed cells, proteins that accumulate in the protein storage vacuoles (PSVs) are synthesized on the endoplasmic reticulum (ER) and transported by vesicles to the PSVs. Vacuolar sorting determinants (VSDs) which are usually amino acid sequences of short or moderate length direct the proteins to this pathway. VSDs identified so far are classified into two types: sequence specific VSDs (ssVSDs) and C-terminal VSDs (ctVSDs). We previously demonstrated that VSDs of α′ and β subunits of β-conglycinin, one of major storage proteins of soybean (Glycine max), reside in the C-terminal ten amino acids. Here we show that both types of VSDs coexist within this region of the α′ subunit. Although ctVSDs can function only at the very C-termini of proteins, the C-terminal ten amino acids of α′ subunit directed green fluorescent protein (GFP) to the PSVs even when they were placed at the N-terminus of GFP, indicating that an ssVSD resides in the sequence. By mutation analysis, it was found that the core sequence of the ssVSD is Ser-Ile-Leu (fifth to seventh residues counted from the C-terminus) which is conserved in the α and β subunits and some vicilin-like proteins. On the other hand, the sequence composed of the C-terminal three amino acids (AFY) directed GFP to the PSVs when it was placed at the C-terminus of GFP, though the function as a VSD was disrupted at the N-terminus of GFP, indicating that the AFY sequence is a ctVSD.     

Co-expression of soybean glycinins A1aB1b and A3B4 enhances their accumulation levels in transgenic rice seed

The soybean major storage protein glycinin is encoded by five genes, which are divided into two subfamilies. Expression of A3B4 glycinin in transgenic rice seed reached about 1.5% of total seed protein, even if expressed under the control of strong endosperm-specific promoters. In contrast, expression of A1aB1b glycinin reached about 4% of total seed protein. Co-expression of the two proteins doubled accumulation levels of both A1aB1b and A3B4 glycinins. This increase can be largely accounted for by their aggregation with rice glutelins, self-assembly and inter-glycinin interactions, resulting in the enrichment of globulin and glutelin fractions and a concomitant reduction of the prolamin fraction. Immunoelectron microscopy indicated that the synthesized A1aB1b glycinin was predominantly deposited in protein body-II (PB-II) storage vacuoles, whereas A3B4 glycinin is targeted to both PB-II and endoplasmic reticulum (ER)-derived protein body-I (PB-I) storage structures. Co-expression with A1aB1b facilitated targeting of A3B4 glycinin into PB-II by sequestration with A1aB1b, resulting in an increase in the accumulation of A3B4 glycinin.     

A C-terminal sequence of soybean beta-conglycinin alpha' subunit acts as a vacuolar sorting determinant in seed cells

In maturing seed cells, many newly synthesized proteins are transported to the protein storage vacuoles (PSVs) via vesicles unique to seed cells. Vacuolar sorting determinants (VSDs) in most of these proteins have been determined using leaf, root or suspension-cultured cells apart from seed cells. In this study, we examined the VSD of the alpha' subunit of beta-conglycinin (7S globulin), one of the major seed storage proteins of soybean, using Arabidopsis and soybean seeds. The wild-type alpha' was transported to the matrix of the PSVs in seed cells of transgenic Arabidopsis, and it formed crystalloid-like structures. Some of the wild-type alpha' was also transported to the translucent compartments (TLCs) in the PSV presumed to be the globoid compartments. However, a derivative lacking the C-terminal 10 amino acids was not transported to the PSV matrix, and was secreted out of the cells, although a portion was also transported to the TLCs. The C-terminal region of alpha' was sufficient to transport a green fluorescent protein (GFP) to the PSV matrix. These indicate that alpha' contains two VSDs: one is present in the C-terminal 10 amino acids and is for the PSV matrix; and the other is for the TLC (the globoid compartment). We further verified that the C-terminal 10 amino acids were sufficient to transport GFP to the PSV matrix in soybean seed cells by using a transient expression system.     

Purification and reconstitution of the proton-pumping ATPase of fungal and plant plasma membranes

The 11S storage protein (glycinin) of soybean [Glycine max (L.) Merr., cv. Raiden] was studied by polyacrylamide gel electrophoresis and amino acid sequence analysis. It contained the following subunits composed of acidic (A) and basic (B) polypeptides: A1aB2, A1bB1b, A2B1a, and A3B4. However, it lacked polypeptides A4, A5, and B3 which are present in many other cultivars. A new acidic polypeptide called A6 was present in a low amount and was characterized by amino acid sequence analysis. It was homologous to A4, although of a smaller apparent molecular weight. Since Raiden has an average protein content of about 40% and its glycinin fraction can be purified as a 350,000 D complex which is typical of other cultivars, the results imply polymorphism with respect to glycinin subunit composition. Because there is a wide variation in the methionine content of the various subunits, these findings suggest the possibility of genetically manipulating the nutritional quality of soybean seed protein by altering glycinin subunit composition.     

Improved bile acid-binding ability of soybean glycinin A1a polypeptide by the introduction of a bile acid-binding peptide (VAWWMY)

We have previously identified a potential bile acid-binding peptide sequence (VAWWMY) in acidic polypeptide A1a of the soybean glycinin A1aB1b subunit (Choi, S. K., et al., Biosci. Biotechnol. Biochem., 66, 2395-2401 (2002)). In this study, we introduced the nucleotide sequence encoding this peptide in the coding DNA which corresponds to amino acids between 251 and 256, and 282 and 287 into the A1a polypeptide by replacement to respectively give modified versions A1aM1 and A1aM2. A fluorescence analysis demonstrates that their bile acid-binding ability was improved compared to A1a. Moreover, modified proglycinin A1aB1b with the VAWWMY sequence at the same sites as those of A1aM1 and A1aM2 was judged to assume the correct conformation. These results suggest the possibility of developing transgenic crops to accumulate the modified glycinin.     

Soybean glycinin A1aB1b subunit has a molecular chaperone-like function to assist folding of the other subunit having low folding ability

Soybean (Glycine max L.) glycinin is composed of five subunits which are classified into two groups (group I: A1aB1b, A1bB2, and A2B1a; group II: A3B4 and A5A4B3). All the common soybean cultivars contain both group I and II subunits (Maruyama, N. et al., Phytochemistry, 64, 701-708 (2003)). The biosynthesis of group I starts earlier compared with that of the A3B4 subunit during seed development (Meinke, D.W. et al., Planta, 153, 130-139 (1981)). We have revealed that group I A1aB1b was mostly expressed as a soluble protein, but that A3B4 was expressed mainly as an insoluble protein in Escherichia coli under the same expression conditions; namely, A1aB1b had higher folding ability than A3B4. We therefore assumed that A1aB1b assists folding of group II subunits like a molecular chaperone does. In order to ascertain this, A1aB1b and A3B4 were co-expressed in E. coli. All of the expressed proteins of A3B4 were recovered in a soluble fraction. To confirm this result, we also co-expressed A1aB1b with modified A3B4 versions having extremely low folding ability. All expressed modified A3B4 versions were soluble. These results clearly suggest that A1aB1b has a molecular chaperone-like function in their folding.     

Improvement of nutritional value and functional properties of soybean glycinin by protein engineering

Glycinin is one of the predominant storage proteins of soybean. To improve its functional properties (heat-induced gelation and emulsification) and/or nutritional value, the A1aB1b proglycinin subunit was modified on the basis of genetically variable domains suggested from the comparison of amino acid sequences of glycinin-type globulins from various legumes and nonlegumes and the relationships between the structure and the functional properties of glycinin. Thus, nucleotide sequences corresponding to each of the variable domains were deleted from the cDNA encoding the A1aB1b proglycinin, and a synthetic DNA encoding four continuous methionines was inserted into the cDNA region corresponding to each of the variable domains. Expression plasmids carrying the modified cDNAs were constructed and expressed in Escherichia coli strain JM105. Some of the modified proteins were accumulated as soluble proteins in the cells at a high level and self-assembled. They exhibited functional properties superior to those of the native glycinin from soybean, which establishes the possibility of creating theoretically designed novel glycinins with high food qualities.     

Improved Bile Acid-binding Ability of Soybean Glycinin A1a Polypeptide by the Introduction of a Bile Acid-binding Peptide (VAWWMY)

We have previously identified a potential bile acid-binding peptide sequence (VAWWMY) in acidic polypeptide A1a of the soybean glycinin A1aB1b subunit (Choi, S. K., et al., Biosci. Biotechnol. Biochem., 66, 2395–2401 (2002)). In this study, we introduced the nucleotide sequence encoding this peptide in the coding DNA which corresponds to amino acids between 251 and 256, and 282 and 287 into the A1a polypeptide by replacement to respectively give modified versions A1aM1 and A1aM2. A fluorescence analysis demonstrates that their bile acid-binding ability was improved compared to A1a. Moreover, modified proglycinin A1aB1b with the VAWWMY sequence at the same sites as those of A1aM1 and A1aM2 was judged to assume the correct conformation. These results suggest the possibility of developing transgenic crops to accumulate the modified glycinin.     

Demonstration in yeast of the function of BP-80, a putative plant vacuolar sorting receptor

BP-80, later renamed VSR(PS-1), is a putative receptor involved in sorting proteins such as proaleurain to the lytic vacuole, with its N-terminal domain recognizing the vacuolar sorting determinant. Although all VSR(PS-1) characteristics and in vitro binding properties described so far favored its receptor function, this function remained to be demonstrated. Here, we used green fluorescent protein (GFP) as a reporter in a yeast mutant strain defective for its own vacuolar receptor, Vps10p. By expressing VSR(PS-1) together with GFP fused to the vacuolar sorting determinant of petunia proaleurain, we were able to efficiently redirect the reporter to the yeast vacuole. VSR(PS-1) is ineffective on GFP either alone or when fused with another type of plant vacuolar sorting determinant from a chitinase. The plant VSR(PS-1) therefore interacts specifically with the proaleurain vacuolar sorting determinant in vivo, and this interaction leads to the transport of the reporter protein through the yeast secretory pathway to the vacuole. This finding demonstrates VSR(PS-1) receptor function but also emphasizes the differences in the spectrum of ligands between Vps10p and its plant equivalent.     

Expression of soybean glycinin subunit precursor cDNAs in Escherichia coli

As the cDNAs encoding A1aB1b and A2B1a subunit precursors of the glycinin A2 subfamily contain a unique NcoI site sequence, (A)CCATGG, occurring at their translation initiation sites, plasmids were constructed to direct the synthesis of those precursor proteins by inserting NcoI/PstI fragments derived from those cDNA clones into the NcoI/PstI-pKK233-2 expression vector in Escherichia coli MV1190, respectively. The resultant plasmids directed the expression of 57-kDa protein components that have molecular masses in agreement with those of the in vitro translation products directed by glycinin A2 subfamily mRNAs, by the addition of isopropyl beta-D-thiogalactoside. These proteins, which comprised as much as 1% of the total bacterial protein, are immunoprecipitable with rabbit antibodies specific for glycinin subunits. This procedure makes glycinin subunits available as a model for studying structure-function relationships in seed proteins using site-directed mutagenesis. This is the first expression of glycinin-like storage protein in E. coli.     

Expression of the Stress-Related Genes for Glutathione S-Transferase and Ascorbate Peroxidase in the Most-Glycinin-Deficient Soybean Cultivar Tousan205 during Seed Maturation

Global analysis of gene expression profiles in most-glycinin-deficient cultivar Tousan205, was performed by DNA microarray analysis. It was confirmed that Tousan205 lacks mRNA expression of three glycinin subunit precursor genes, G1 (A1aB1x), G2 (A2B1a), and G5 (A3B4), and lacks G4 (A5A4B3) protein. Most glycinin subunits were deficient in mature seeds of Tousan205. We compared the gene expression of Tousan205 with those of parent cultivar, Tamahomare, which was used for crossbreeding of Tousan205. As a result, Tousan205 exhibited higher expression of some seed maturation proteins, and stress-related genes such as glutathione S-transferase and ascorbate peroxidase. This result indicates the possibility that the decrease of main storage protein, glycinin causes stress in soybean.     

Regeneration of a lytic central vacuole and of neutral peripheral vacuoles can be visualized by green fluorescent proteins targeted to either type of vacuoles

Protein trafficking to two different types of vacuoles was investigated in tobacco (Nicotiana tabacum cv SR1) mesophyll protoplasts using two different vacuolar green fluorescent proteins (GFPs). One GFP is targeted to a pH-neutral vacuole by the C-terminal vacuolar sorting determinant of tobacco chitinase A, whereas the other GFP is targeted to an acidic lytic vacuole by the N-terminal propeptide of barley aleurain, which contains a sequence-specific vacuolar sorting determinant. The trafficking and final accumulation in the central vacuole (CV) or in smaller peripheral vacuoles differed for the two reporter proteins, depending on the cell type. Within 2 d, evacuolated (mini-) protoplasts regenerate a large CV. Expression of the two vacuolar GFPs in miniprotoplasts indicated that the newly formed CV was a lytic vacuole, whereas neutral vacuoles always remained peripheral. Only later, once the regeneration of the CV was completed, the content of peripheral storage vacuoles could be seen to appear in the CV of a third of the cells, apparently by heterotypic fusion.     

A rice glutelin and a soybean glycinin have evolved from a common ancestral gene

A cDNA clone covering the entire coding region for a glutelin subunit precursor has been identified from a library of endosperm-developing rice cDNA clones using a mixed oligodeoxynucleotide probe, and then by immunoprecipitation of hybrid-selected translation product with an antiserum against the acidic polypeptides of the glutelin. Analysis of the cDNA insert revealed that rice glutelin is synthesized as precursor polypeptides which undergo post-translational processing to form the nonrandom polypeptide pairs, like glycinin precursors of soybean. By comparing the predicted protein sequence of this precursor from monocots with that of glycinin A1aB1b precursor from dicots, it was found that the overall 32% of the amino acid positions are identical in both proteins. Because regions which show identities are dispersed throughout both molecules, the similarity is not due to convergent evolution, but to divergence evolution from a common ancestral gene.     

Proteome rebalancing in soybean seeds can be exploited to enhance foreign protein accumulation

Seeds possess a high intrinsic capacity for protein production that makes them a desirable bioreactor platform for the manufacture of transgenic products. One strategy to enhance foreign protein production involves exchanging the capacity to produce intrinsic proteins for the capacity to produce a high level of foreign proteins. Suppression of the alpha/alpha' subunit of beta-conglycinin storage protein synthesis in soybean has been shown previously to result in an increase in the accumulation of the glycinin storage protein, some of which is sequestered as proglycinin into de novo endoplasmic reticulum (ER)-derived protein bodies. The exchange of glycinin for conglycinin is quantitative, with the remodelled soybeans possessing a normal protein content with an altered proteome. The green fluorescent protein (GFP)-kdel reporter was transferred in a construct using the glycinin promoter and terminator to mimic glycinin gene expression. When expressed in soybean seeds, GFP-kdel accreted to form ER-derived protein bodies. The introgression of GFP-kdel into the alpha/alpha' subunit of the beta-conglycinin suppression background resulted in a fourfold enhancement of GFP-kdel accumulation to > 7% (w/w) of the total protein in soybean seeds. The resulting seeds accumulated a single population of ER membrane-bound protein bodies that contained both GFP-kdel and glycinin. Thus, the collateral proteome rebalancing that occurs with the suppression of intrinsic proteins in soybean can be exploited to produce an enhanced level of foreign proteins.     

Signal sequence of preproglycinin affects production of the expressed protein in Escherichia coli

The effect of the signal peptide portion on the bacterial production of preproglycinin, a precursor of soybean storage protein, was examined. Nucleotide sequences corresponding to the signal peptide and the mature N-terminal region were deleted stepwise from the cDNA encoding the glycinin A1aB1b subunit precursor, and the deleted cDNAs were placed under the control of trc promoter in an expression vector pKK233-2. When the amounts of the protein products in Escherichia coli from each expression plasmid were determined, no accumulation of preproglycinin was observed from the plasmids with the full length or the five amino acids of the signal sequence. However, significant accumulation of the preproglycinin homologue proteins was noted from the plasmids retaining less than three amino acids of the signal sequence depending on the extent of deletion. N-terminal amino acid sequences of the products coincided with those predicted from the deleted cDNAs. The preproglycinin homologue proteins expressed from the mutant plasmids assembled into trimers of about 8S.     

Introduction of enterostatin (VPDPR) and a related sequence into soybean proglycinin A1aB1b subunit by site-directed mutagenesis

Enterostatin (VPDPR), having anoretic and hypocholesterolemic activities, and its homologue LPYPR, a hypocholesterolemic peptide found in the glycinin A5A4B3 subunit, were introduced into the corresponding site (TNGPQ) of the proglycinin A1aB1b subunit by site-directed mutagenesis. Modified proglycinins were expressed in E. coli and recovered from the insoluble fraction. VPDPR and LPYPR were released by the action of chymotrypsin and trypsin as expected. The overall yields of purified VPDPR and LPYPR were 40% and 62%, respectively.     

Vps10p transport from the trans-Golgi network to the endosome is mediated by clathrin-coated vesicles

A native immunoisolation procedure has been used to investigate the role of clathrin-coated vesicles (CCVs) in the transport of vacuolar proteins between the trans-Golgi network (TGN) and the prevacuolar/endosome compartments in the yeast Saccharomyces cerevisiae. We find that Apl2p, one large subunit of the adaptor protein-1 complex, and Vps10p, the carboxypeptidase Y vacuolar protein receptor, are associated with clathrin molecules. Vps10p packaging in CCVs is reduced in pep12 Delta and vps34 Delta, two mutants that block Vps10p transport from the TGN to the endosome. However, Vps10p sorting is independent of Apl2p. Interestingly, a Vps10C(t) Delta p mutant lacking its C-terminal cytoplasmic domain, the portion of the receptor responsible for carboxypeptidase Y sorting, is also coimmunoprecipitated with clathrin. Our results suggest that CCVs mediate Vps10p transport from the TGN to the endosome independent of direct interactions between Vps10p and clathrin coats. The Vps10p C-terminal domain appears to play a principal role in retrieval of Vps10p from the prevacuolar compartment rather than in sorting from the TGN.     

Integral Membrane Protein Sorting to Vacuoles in Plant Cells: Evidence for Two Pathways

Plant cells may contain two functionally distinct vacuolar compartments. Membranes of protein storage vacuoles (PSV) are marked by the presence of α-tonoplast intrinsic protein (TIP), whereas lytic vacuoles (LV) are marked by the presence of γ-TIP. Mechanisms for sorting integral membrane proteins to the different vacuoles have not been elucidated. Here we study a chimeric integral membrane reporter protein expressed in tobacco suspension culture protoplasts whose traffic was assessed biochemically by following acquisition of complex Asn-linked glycan modifications and proteolytic processing, and whose intracellular localization was determined with confocal immunofluorescence. We show that the transmembrane domain of the plant vacuolar sorting receptor BP-80 directs the reporter protein via the Golgi to the LV prevacuolar compartment, and attaching the cytoplasmic tail (CT) of γ-TIP did not alter this traffic. In contrast, the α-TIP CT prevented traffic of the reporter protein through the Golgi and caused it to be localized in organelles separate from ER and from Golgi and LV prevacuolar compartment markers. These organelles had a buoyant density consistent with vacuoles, and α-TIP protein colocalized in them with the α-TIP CT reporter protein when the two were expressed together in protoplasts. These results are consistent with two separate pathways to vacuoles for membrane proteins: a direct ER to PSV pathway, and a separate pathway via the Golgi to the LV.