Bean homologs of the mammalian glucose-regulated proteins: induction by tunicamycin and interaction with newly synthesized seed storage proteins in the endoplasmic reticulum

Treatment of developing bean cotyledons with the inhibitor of N-glycosylation tunicamycin enhanced the synthesis of at least two polypeptides with molecular mass 78 kDa and 97 kDa. Pulse-chase experiments and subcellular fractionation indicated that these are endoplasmic reticulum (ER) residents. The 78 kDa protein is a major component of the ER protein fraction and, by N-terminal sequencing, was identified as a bean homolog of the mammalian 78 kDa glucose-regulated protein (GRP78). This is a molecular chaperone that is probably involved in the folding and oligomerization of several animal and yeast proteins in the ER. When newly synthesized storage glycoproteins phaseolin, phytohemagglutinin or alpha-amylase inhibitor were immunoprecipitated from an ER preparation of tunicamycin-treated tissue, the GRP78 homolog was always co-precipitated. Bound GRP78 homolog could be released by ATP treatment. These results suggest that, at least when glycosylation is inhibited, this protein plays a role in the early stages of the synthesis of vacuolar storage proteins.     

In vivo and in vitro processing of seed reserve protein in the endoplasmic reticulum: evidence for two glycosylation steps

Cotyledons of the common bean (Phaseolus vulgaris L.) synthesize large amounts of the reserve protein phaseolin. The polypeptides are synthesized on membrane-bound polysomes, pass through the endoplasmic reticulum (ER) and accumulate in protein bodies. For a study of the biosynthesis and processing of phaseolin, developing cotyledons were labeled with radioactive amino acids, glucosamine and mannose, and isolated fractions (polysomal RNA, polysomes, and rough ER) were used for in vitro protein synthesis. Newly synthesized phaseolin present in the ER of developing cotyledons can be fractioned into four glycopolypeptides by SDS PAGE. In vitro synthesis with polysomal RNA results in the formation of two polypeptides by polysome run-off shows that glycosylation is a co-translational event. The two unglycosylated polypeptides formed by polysome run-off are slightly smaller than the two polypeptides formed by in vitro translation of isolated RNA, indicating that a signal peptide may be present on these polypeptides. Run-off synthesis with rough ER produces a pattern of four polypeptides similar to the one obtained by in vivo labeling. The two abundant glycopolypeptides formed by polysome run-off. This result indicates the existence of a second glycosylation event for the abundant polypeptides. Inhibition of glycosylation by Triton X-100 during chain-completion with rough ER was used to show that these two glycosylation steps normally occur sequentially. Both glycosylation steps are inhibited by tunicamycin. Analysis of carhohydrate to protein ratios of the different polypeptides and of trypsin digests of polypeptides labeled with [(3)H]glucosamine confirmed the conclusion that some glycosylated polypeptides contain two oligosaccharide chains, while others contain only one. An analysis of tryptic peptide maps shows that each of the unglycosylated polypeptides is the precursor for one glycosylated polypeptide with one oligosaccharide chain and one with two oligosaccharide chains.     

The Binding Protein Associates with Monomeric Phaseolin

The association of the binding protein (BiP) with newly synthesized proteins in the endoplasmic reticulum (ER) of developing bean (Phaseolus vulgaris) cotyledonary cells was investigated. ATP-sensitive association with many polypeptides was detected. The fraction of newly synthesized polypeptides associated with BiP varies among different proteins. The relationship between subunit assembly and binding to BiP was investigated in the case of the vacuolar trimeric glycoprotein phaseolin. In spite of the presence of a significant pool of phaseolin trimers in the ER, only monomeric phaseolin is found in association with BiP. On the whole, our results point to a general role of BiP in the synthesis of plant secretory proteins and indicate that, in the case of phaseolin, BiP binding sites are concealed during structural maturation in the ER, either before or upon formation of trimers. Our results also indicate that trimerization does not constitute a rate-limiting step in the transport of phaseolin to the protein storage vacuoles.     

In vivo endoproteolytically cleaved phaseolin is stable and accumulates in developing Phaseolus lunatus L. seeds

Phaseolin is the most abundant storage protein of bean seeds. To modify its amino-acidic composition by protein engineering, for the improvement of its nutritional value, regions which could be modified without detrimental effects on structural features of the protein must be identified. Data presented here, on the characterisation of the major storage protein of lima bean (Phaseolus lunatus L.) seeds, a phaseolin-like glycoprotein, provide good indications on one of such region. Phaseolus lunatus phaseolin consists of four major oligomers containing two subunit classes. Polypeptides of one class show a molecular mass ranging from 38.5 kDa to 32 kDa, while the molecular mass of polypeptides belonging to the other class ranges from 27 kDa to 21 kDa. The subunits originate from the cleavage of precursor forms, with molecular masses of 58 kDa and 54 kDa, which are still present — in residual amounts — in the mature protein. Comparison of their N-terminal sequences with those of the subunits demonstrate that cleavage occurs in a region of the molecule that instead remains uncleaved in phaseolins of the other species. Since this region can accommodate such a drastic modification, we suggest it could be a good candidate for in vitro manipulation.     

Mannose analog 1-deoxymannojirimycin inhibits the Golgi-mediated processing of bean storage glycoproteins

The asparagine-linked oligosaccharide chains of glycoproteins can be processed to form a wide variety of structures. The Golgi complex is the main compartment involved in this processing. In mammalian cells the first enzyme acting along the Golgi processing pathway is mannosidase I, whose action is a prerequisite for any further processing and which is inhibited by the mannose analog 1-deoxymannojirimycin (dMM). To have insights into the processing pathway in plant cells, we have studied the in vivo effect of dMM on the processing of the bean (Phaseolus vulgaris) storage proteins phaseolin and phytohemagglutinin, two well characterized plant glycoproteins. Cotyledons obtained from developing seeds were labeled with radioactive leucine, glucosamine, or fucose in the presence or absence of dMM. Treatment with dMM fully inhibited the acquisition of resistance to endo-β-N-acetylglucosaminidase H by phaseolin and phytohemagglutinin and the incorporation of fucose into protein. Furthermore, the apparent molecular weight of the polypeptides of phaseolin and phytohemagglutinin synthesized in dMM-treated cotyledons was consistent with the exclusive presence of oligommanose oligosaccharide chains which had not been processed in the Golgi complex. The inhibition of processing did not prevent exit from the Golgi complex, and most probably the storage proteins were correctly targeted to the protein bodies as indicated by the post-translational polypeptide cleavage of phaseolin. These results indicate that the action of a mannosidase is the first obligatory step of Golgi-mediated processing also in a plant cell and, together with data obtained in other laboratories on the in vitro specificity of glycosidases and glycosyltransferases present in the Golgi complex of plant cells, support the hypothesis that the key early reactions in Golgi-mediated processing are similar if not identical in plants and mammals.     

The Rate of Phaseolin Assembly Is Controlled by the Glucosylation State of Its N-Linked Oligosaccharide Chains

Many of the proteins that are translocated into the endoplasmic reticulum are glycosylated with the addition of a 14-saccharide core unit (Glc3Man9GlcNAc2) to specific asparagine residues of the nascent polypeptide. Glucose residues are then removed by endoplasmic reticulum-located glucosidases, with diglucosylated and monoglucosylated intermediates being formed. In this study, we used a cell-free system constituted of wheat germ extract and bean microsomes to examine the role of glucose trimming in the structural maturation of phaseolin, a trimeric glycoprotein that accumulates in the protein storage vacuoles of bean seeds. Removal of glucose residues from the N-linked chains of phaseolin was blocked by the glucosidase inhibitors castanospermine and N-methyldeoxynojirimycin. If glucose trimming was not allowed to occur, the assembly of phaseolin was accelerated. Conversely, polypeptides bearing partially trimmed glycans were unable to form trimers. The effect of castanospermine on the rate of assembly was much more pronounced for phaseolin polypeptides that have two glycans but was also evident when a single glycan chain was present, indicating that glycan clustering can modulate the effect of glucose trimming on the rate of trimer formation. Therefore, the position of glycan chains and their accessibility to the action of glucosidases can be fundamental elements in the control of the structural maturation of plant glycoproteins.     

Proteinase A-like enzyme from germinated kidney bean seeds. Its action on phaseolin and vicilin

A cysteine proteinase that possibly participates in the degradation of phaseolin, the main storage protein of kidney bean ( Phaseolus vulgaris L. cv. Moldavian) was isolated from germinating kidney bean seeds and partially characterized. According to its properties it may be classified as a member of a group of homologous cysteine proteinases A, also present in germinating seeds of a number of other plants. The proteinase of this group hydrolyze storage proteins to short peptides. Similarly, the kidney bean proteinase hydrolyzes vicilin, the reserve protein of vetch ( Vicia sativa ). However, its action on phaseolin is limited to the cleavage of subunits into two approximately equal parts and to the splitting off a small number of short peptides. An explanation of phaseolin resistance to the action of this proteinase is proposed on the basis of the differences of its structure from that of other homologous 7S proteins.     

Purification of a seed glycoprotein: N-terminal and deglycosylation analysis of phaseolin

Phaseolin, the major storage protein of the French bean Phaseolus vulgaris cv. Tendergreen, has been isolated and purified by either ion-exchange chromatography or reversed-phase HPLC. These purification procedures were used to fractionate the native protein aggregate into its characteristic subunit components. Amino-terminal sequence analysis was performed on the intact peptide subunits. Native phaseolin was chemically cleaved at a unique tryptophan residue which is proximal to the N-terminal region of the protein with BNPS-skatole and the resulting peptide fragments were isolated via reversed-phase HPLC. Chemical and enzymatic sequence results obtained from these peptide fragments are in full agreement with the results obtained for the full length peptide subunits. These N-terminal analyses show that the signal peptide cleavage process is somewhat random resulting in the phaseolin polypeptides having possibly three or four different N-termini. Native phaseolin and purified subunits were chemically deglycosylated with trifluoromethanesulphonic acid in the presence of an anisole scavenger. One-dimensional SDS-PAGE analysis of the deglycosylated products show that differential glycosylation is largely responsible for much of the observed molecular weight heterogeneity found among phaseolin polypeptides.     

A phaseolin domain involved directly in trimer assembly is a determinant for binding by the chaperone BiP

The binding protein (BiP; a member of the heat-shock 70 family) is a major chaperone of the endoplasmic reticulum (ER). Interactions with BiP are believed to inhibit unproductive aggregation of newly synthesized secretory proteins during folding and assembly. In vitro, BiP has a preference for peptide sequences enriched in hydrophobic amino acids, which are expected to be exposed only in folding and assembly intermediates or in defective proteins. However, direct information regarding sequences recognized in vivo by BiP on real proteins is very limited. We have shown previously that newly synthesized monomers of the homotrimeric storage protein phaseolin associate with BiP and that phaseolin trimerization in the ER abolishes such interactions. Using different phaseolin constructs and green fluorescent protein (GFP) fusion proteins, we show here that one of the two alpha-helical regions of polypeptide contact in phaseolin trimers (35 amino acids located close to the C terminus and containing three potential BiP binding sites) effectively promotes BiP association with phaseolin and with secretory GFP fusions expressed in transgenic tobacco or in transfected protoplasts. We also show that overexpressed BiP transiently sequesters phaseolin polypeptides. We conclude that one of the regions of monomer contact is a BiP binding determinant and suggest that during the synthesis of phaseolin, the association with BiP and trimer formation are competing events. Finally, we show that the other, internal region of contact between monomers is necessary for phaseolin assembly in vivo and contains one potential BiP binding site.     

Immunocytochemical localization of phaseolin and phytohemagglutinin in the endoplasmic reticulum and Golgi complex of developing bean cotyledons

Development of legume seeds is accompanied by the synthesis of storage proteins and lectins, and the deposition of these proteins in protein-storage vacuoles (protein bodies). We examined the subcellular distribution, in developing seeds of the common bean, Phaseolus vulgaris L., of the major storage protein (phaseolin) and the major lectin (phytohemagglutinin, PHA). The proteins were localized using an indirect immunocytochemical method in which ultrathin frozen sections were immunolabeled with rabbit antibodies specific for either PHA or phaseolin. Bound antibodies were then localized using goat-anti-rabbit immunoglobulin G adsorbed onto 4- to 5-nm colloidal gold particles. The sections were post-fixed with OsO₄, dehydrated, and embedded in plastic on the grids. Both PHA and phaseolin exhibited a similar distribution in the storage-parenchyma cells, being found primarily in the developing protein bodies. Endoplasmic reticulum and Golgi complexes (cisternal stacks and associated vesicles) also were specifically labeled for both proteins, whereas the cytosol and other organelles, such as mitochondria, were not. We interpret these observations as supporting the hypothesis that the transport of storage proteins and lectins from their site of synthesis, the rough endoplasmic reticulum, to their site of deposition, the protein bodies, is mediated by the Golgi complex.Abbreviations ER endoplasmic reticulum - IgG immunoglobulin G - PBS phosphate-buffered saline - PHA phytohemagglutinin     

2-DE-based proteomic analysis of common bean (Phaseolus vulgaris L.) seeds

Common bean (Phaseolus vulgaris L.) is the most important grain legume for direct human consumption. Proteomic studies in legumes have increased significantly in the last years but few studies have been performed to date in P. vulgaris. We report here a proteomic analysis of bean seeds by two-dimensional electrophoresis (2-DE). Three different protein extraction methods (TCA-acetone, phenol and the commercial clean-up kit) were used taking into account that the extractome can have a determinant impact on the level of quality of downstream protein separation and identification. To demonstrate the quality of the 2-DE analysis, a selection of 50 gel spots was used in protein identification by mass spectrometry (MALDI-TOF MS and MALDI-TOF/TOF). The results showed that a considerable proportion of spots (70%) were identified in spite of incomplete genome/protein databases for bean and other legume species. Most identified proteins corresponded to storage protein, carbohydrate metabolism, defense and stress response, including proteins highly abundant in the seed of P. vulgaris such as the phaseolin, the phytohemagglutinin and the lectin-related α-amylase inhibitor.     

Crosslinking of microsomal proteins identifies P-9000, a protein that is co-transported with phaseolin and phytohemagglutinin in bean cotyledons

Developing cotyledons of the common bean, Phaseolus vulgaris L., transport within their secretory system (endoplasmic reticulum and Golgi apparatus) the abundant vacuolar proteins, phaseolin and phytohemagglutinin. To identify proteins that may play a role in vacuolar targeting, we treated cotyledon microsomal fractions with a bifunctional crosslinking reagent, dithiobis(succinimidyl propionate), isolated protein complexes with antibodies to phaseolin and phytohemagglutinin, and analysed the polypeptides by sodium dodecylsulfate polyacrylamide gel electrophoresis. This allowed us to identify a protein of M_r=9000 (P-9000) that was crosslinked to both phaseolin and phytohemagglutinin. P-900 is abundantly present in the endoplasmic reticulum. The aminoterminus of P-9000 shows extensive sequence identity with the amino-terminus of PA1 (M_r=11 000), a cysteine-rich albumin whose processing products accumulate in the vacuoles of pea (Pisum sativum L.) cotyledons. Like PA1, P-9000 is synthesized as a pre-proprotein that is posttranslationally processed into smaller polypeptides. The possible functions of P-9000 are discussed.Abbreviations DSP dithiobis(succinimidyl propionate) - EDTA ethylenediaminetetraacetic acid - ER endoplasmic reticulum - kDa kilodalton - M_r relative molecular mass - PHA phytohemagglutinin - SDS sodium dodecylsulfate - PAGE polyacrylamide gel electrophoresis     

Glycosylation is not needed for the intracellular transport of phytohemagglutinin in developing Phaseolus vulgaris cotyledons and for the maintenance of its biological activities

Approximately 10% of the total protein contained in Phaseolus vulgaris L. cv. Greensleeves seeds is composed of the glycoprotein lectin, phytohemagglutinin. We have investigated whether the presence of N-linked oligosaccharide side chains is a prerequisite for the correct intracellular transport of this protein and whether unglycosylated phytohemagglutinin maintains its biological activities. Excised developing cotyledons were incubated in the presence of tunicamycin to prevent glycosylation "in vivo", and the fate of the unglycosylated protein synthesized in such cotyledons determined. It was found that unglycosylated phytohemagglutinin reaches its normal site of accumulation, the protein bodies, and maintains erythro-agglutinating and mitogenic activities.     

Analysis of N-linked glycosylation of hantaan virus glycoproteins and the role of oligosaccharide side chains in protein folding and intracellular trafficking

The membrane glycoproteins Gn and Gc of Hantaan virus (HTNV) (family Bunyaviridae) are modified by N-linked glycosylation. The glycoproteins contain six potential sites for the attachment of N-linked oligosaccharides, five sites on Gn and one on Gc. The properties of the N-linked oligosaccharide chains were analyzed by treatment with endoglycosidase H, peptide:N-glycosidase F, tunicamycin, and deoxynojirimycin and were confirmed to be completely of the high-mannose type. Ten glycoprotein gene mutants were constructed by site-directed mutagenesis, including six single N glycosylation site mutants and four double-site mutants. We determined that four sites (N134, -235, -347, and -399) on Gn and the only site (N928) on Gc in their ectodomains are utilized, whereas the fifth site on Gn (N609), which faces the cytoplasm, is not glycosylated. The importance of individual N-oligosaccharide chains varied with respect to folding and intracellular transport. The oligosaccharide chain on residue N134 was found to be crucial for protein folding, whereas single mutations at the other glycosylation sites were better tolerated. Mutation at glycosylation sites N235 and N399 together resulted in Gn misfolding. The endoplasmic reticulum chaperones calnexin and calreticulin were found to be involved in HTNV glycoprotein folding. Our data demonstrate that N-linked glycosylation of HTNV glycoproteins plays important and differential roles in protein folding and intracellular trafficking.     

Two Kinds of Protein Glycosylation in a Cell-Free Preparation from Developing Cotyledons of Phaseolus vulgaris

Membrane preparations from developing cotyledons of red kidney bean (Phaseolus vulgaris L.) transferred radioactive mannose from GDP-mannose (U-[¹⁴C]mannose) to endogenous acceptor proteins. The transfer was inhibited by the antibiotic tunicamycin, suggesting the involvement of lipidoligosaccharide intermediates typical of the pathway for glycosylation of asparagine residues. This was supported by the similarity of the linkage types of radioactive mannose in lipid-oligosaccharide and glycoprotein products; both contained labeled 2-linked mannose, 3,6-linked and terminal mannose typical of glycoprotein “core” oligosaccharides. As expected for “core” glycosylation, the transfer of labeled N-acetylglucosamine (GlcNAc) from UDP-GlcNAc (6-[³H]GLcNAc) to 4-linkage in endogenous glycoproteins could also be demonstrated. However, most of the radioactive GlcNAc was incorporated into terminal linkage, in a reaction insensitive to tunicamycin. The proteins receiving “core” oligosaccharide in vitro were heterogeneous in size, in contrast to those receiving most of the GlcNAc (which chiefly comprised the seed reserve-proteins phaseolin and phytohemagglutinin). It is suggested that following “core” glycosylation, single GlcNAc residues are attached terminally to the oligosaccharides of these seed proteins, without the involvement of lipid-linked intermediates. Phaseolin from mature seeds does not possess a significant amount of terminal GlcNAc and so it is possible that these residues are subsequently removed in a processing event.     

Expression,glycosylation and secretion of phaseolin in a baculovirus system

In this report, we describe the efficient expression and glycosylation, in insect cells, of -phaseolin polypeptides (M _r 45 and 48 kDa) from Phaseolus vulgaris, by means of a baculovirus expression vector. N-terminal sequence analysis demonstrated that the signal peptide was efficiently processed. Tunicamycin treatment suppressed both phaseolin bands seen in untreated or control cells, and resulted in a single species (M _r 43 kDa). We provide evidence that the observed size heterogeneity arises by asymmetric glycosylation of a single, high-molecular weight precursor. These results also indicate that differential glycosylation of phaseolin polypeptides can occur on the product of a single gene, and, in that sense, is not dependent on amino acid sequence variations. Phaseolin accumulates to a very high level (90 µg/10⁶ cells), 90% of it being secreted into the culture medium. Immuno-gold staining and electron microscopy demonstrated phaseolin polypeptides in electron-dense, membrane-bound vesicles seen at the periphery of the cytoplasm of infect cells and in cytoplasmic multivesicular bodies. The effect on protein accumulation of a single-basepair transversion (G»C) at position +6 is also described. This study constitutes, to our knowledge, one of the first instances of a plant protein being expressed in insect cells and suggests possible differences in the sorting mechanisms of glycoproteins from legume seeds and those from Spodoptera frugiperda cell line Sf9.     

Signal peptide specificity in posttranslational processing of the plant protein phaseolin in Saccharomyces cerevisiae.

We linked the cDNA coding region for the bean storage protein phaseolin to the promoter and regulatory region of the Saccharomyces cerevisiae repressible acid phosphatase gene (PHO5) in multicopy expression plasmids. Yeast transformants containing these plasmids expressed phaseolin at levels up to 3% of the total soluble cellular protein. Phaseolin polypeptides in S. cerevisiae were glycosylated, and their molecular weights suggested that the signal peptide had been processed. We also constructed a series of plasmids in which the phaseolin signal-peptide-coding region was either removed or replaced with increasing amounts of the amino-terminal coding region for acid phosphatase. Phaseolin polypeptides with no signal peptide were not posttranslationally modified in S. cerevisiae. Partial or complete substitution of the phaseolin signal peptide with that from acid phosphatase dramatically inhibited both signal peptide processing and glycosylation, suggesting that some specific feature of the phaseolin signal amino acid sequence was required for these modifications to occur. Larger hybrid proteins that included approximately one-half of the acid phosphatase sequence linked to the amino terminus of the mature phaseolin polypeptide did undergo proteolytic processing and glycosylation. However, these polypeptides were cleaved at several sites that are not normally used in the unaltered acid phosphatase protein.     

Structure, position, and biosynthesis of the high mannose and the complex oligosaccharide side chains of the bean storage protein phaseolin

Phaseolin, the major storage protein of the common bean (Phaseolus vulgaris), is a glycoprotein which is synthesized during seed development and accumulates in protein storage vacuoles or protein bodies. The protein has three different N-linked oligosaccharide side chains: Man9(GlcNAc)2, Man7(GlcNAc)2, and Xyl-Man3(GlcNAc)2 (where Xyl represents xylose). The structures of these glycans were determined by 1H NMR spectroscopy. The Man9(GlcNAc)2 glycan has the typical structure found in plant and animal glycoproteins. The structures of the two other glycans are shown below. (Formula; see text) Phaseolin was separated by electrophoresis on denaturing gels into four size classes of polypeptides. The two abundant ones have two oligosaccharides each, whereas the less abundant ones have only one oligosaccharide each. Polypeptides with two glycans have Man7(GlcNAc)2 attached to Asn252 and Man9(GlcNAc)2 attached to Asn341. Polypeptides with only one glycan have Xyl-Man3(GlcNAc)2 attached to Asn252. Both these asparagine residues are in canonical glycosylation sites; the numbering starts with the N-terminal methionine of the signal peptide of phaseolin. The presence of the Man7(GlcNAc)2 and of Xyl-Man3(GlcNAc)2 at the same asparagine residue (position 252) of different polypeptides seems to be controlled by the glycosylation status of Asn341. When Asp341 is unoccupied, the glycan at Asn252 is complex. When Asn341 is occupied, the glycan at Asn252 is only modified to the extent that 2 mannosyl residues are removed. The processing of the glycans, after the removal of the glucose residues, involves enzymes in the Golgi apparatus as well as in the protein bodies. Formation of the Xyl-Man3(GlcNAc)2 glycan is a multistep process that involves the Golgi apparatus-mediated removal of 6 mannose residues and the addition of 2 N-acetylglucosamine residues and 1 xylose. The terminal N-acetylglucosamine residues are later removed in the protein bodies. The conversion of Man9(GlcNAc)2 to Man7(GlcNAc)2 is a late processing event which occurs in the protein bodies. Experiments in which [3H]glucosamine-labeled phaseolin obtained from the endoplasmic reticulum (i.e. precursor phaseolin) is incubated with jack bean alpha-mannosidase show that the high mannose glycan on Asn252, but not the one on Asn341, is susceptible to enzyme degradation. Incubation of [3H] glucosamine-labeled phaseolin obtained from the Golgi apparatus with jack bean beta-N-acetylglucosaminidase results in the removal of the terminal N-acetylglucosamine residues from the complex chain.(ABSTRACT TRUNCATED AT 400 WORDS)     

Role of the endoplasmic reticulum in the synthesis of reserve proteins and the kinetics of their transport to protein bodies in developing pea cotyledons

Developing pea (Pisum sativum L.) cotyledons were labeled with radioactive amino acids, glucosamine, and mannose in pulse an pulse- chase experiments to study the synthesis, glycosylation, and transport of the reserve proteins vicilin and legumin to the protein bodies. Tissue extracts were fractionated on sucrose gradients to isolate either the endoplasmic reticulum (ER) or the protein bodies. Immunoaffinity gels were used to determine radioactivity in the reserve proteins (legumin and vicilin). After pulse-labeling for 45 min with amino acids, about half the total incorporated radioactivity coincided closely with the position of the ER marker enzyme NADH-cytochrome c reductase at a density of 1.13 g . cm-3 on the sucrose gradient. Both radioactivity and enzyme activity shifted to a density of 1.18 g . cm-3 in the presence of 3 mM MgCl2 indicating that the radioactive proteins were associated with the rough ER. Approximately half of the incorporated radioactivity associated with the rough ER was in newly synthesized reserve protein and this accounted for 80% of the reserve protein synthesized in 45 min. Trypsin digestion experiments indicated that these proteins were sequestered within the ER. In pulse-chase experiments, the reserve proteins in the ER became radioactive without appreciable lag and radioactivity chased out of the ER with a half-life of 90 min. Radioactive reserve proteins became associated with a protein body-rich fraction 20-30 min after their synthesis and sequestration by the ER. Pulse-chase experiments with radioactive glucosamine and mannose in the presence and absence of tunicamycin indicated that glycosylation of vicilin occurs in the ER. However, glycosylation is not a prerequisite for transport of vicilin from ER to protein bodies. Examination of the reserve protein polypeptides by SDS PAGE followed by fluorography showed that isolated ER contained legumin precursors (Mr 60,000-65,000) but not the polypeptides present in mature legumin (Mr 40,000 and 19,000) as well as the higher molecular weight polypeptides of vicilin (Mr 75,000, 70,000, 50,000, and 49,000). The smaller polypeptides of vicilin present in vicilin extracted from protein bodies (Mr 12,000-34,000) were absent from the ER. The results show that newly synthesized reserve proteins are preferentially and transiently sequestered within the ER before they move to the protein bodies, and that the ER is the site of storage protein glycosylation.