A relaxed specificity in interchain disulfide bond formation characterizes the assembly of a low-molecular-weight glutenin subunit in the endoplasmic reticulum

Abstract

Wheat (Triticum spp.) grains contain large protein polymers constituted by two main classes of polypeptides: the high-molecular-weight glutenin subunits and the low-molecular-weight glutenin subunits (LMW-GS). These polymers are among the largest protein molecules known in nature and are the main determinants of the superior technological properties of wheat flours. However, little is known about the mechanisms controlling the assembly of the different subunits and the way they are arranged in the final polymer. Here, we have addressed these issues by analyzing the formation of interchain disulfide bonds between identical and different LMW-GS and by studying the assembly of mutants lacking individual intrachain disulfides. Our results indicate that individual cysteine residues that remain available for disulfide bond formation in the folded monomer can form interchain disulfide bonds with a variety of different cysteine residues present in a companion subunit. These results imply that the coordinated expression of many different LMW-GS in wheat endosperm cells can potentially lead to the formation of a large set of distinct polymeric structures, in which subunits can be arranged in different configurations. In addition, we show that not all intrachain disulfide bonds are necessary for the generation of an assembly-competent structure and that the retention of a LMW-GS in the early secretory pathway is not dependent on polymer formation.

Figure 1.

B11-33 and 1B LMW-GS structure. Schematic diagram showing the proposed disulphide structure of the B11-33 (Okita et al., 1985) and 1B (D'Ovidio et al., 1997) LMW-GS. Designation of Cys residues is according to Müller et al. (1998). Numbers indicate their position in the primary structure, starting from the first amino acid in the signal peptide. Nonrepetitive regions are shaded. I, II, and III indicate the three regions that constitute the nonrepetitive C-terminal domain. SP, Signal peptide.

Figure 2.

LMW-GS polymer formation in tobacco protoplasts. Tobacco protoplasts were transfected with a plasmid encoding the B11-33 LMW-GS or with vector as control. Protoplasts were labeled with [³⁵S]Cys and [³⁵S]Met for 1.5 h and then homogenized. Proteins immunoselected with an anti-glutenin antiserum were analyzed by nonreducing (OX, lanes 1 and 2) or reducing (RED, lanes 3 and 4) SDS-PAGE and fluorography. A_ox, Oxidized monomer; A_red, reduced monomer. B₁, B₂, B₃, C, D, and E indicate the positions of B11-33 containing complexes. The positions of molecular mass markers (kD) are shown on the right.

Figure 3.

Analysis of B11-33 Cys mutants. A, Tobacco protoplasts were transfected with vector DNA or plasmids encoding the wild-type B11-33 protein (WT) or the C25S, C230S, C(25, 230)S mutants, alone or in combination, as indicated. Protoplasts were labeled with [³⁵S]Cys and [³⁵S]Met for 1.5 h and then homogenized. Proteins immunoselected with an anti-glutenin antiserum were analyzed by SDS-PAGE under nonreducing conditions and fluorography. A_ox, Oxidized monomer. B₁, B₂, B₃, C, D, and E indicate the positions of the different B11-33-containing complexes. B, Tobacco protoplasts were transfected with vector DNA or plasmids encoding the C25S, C25S-HA, C230S, C230S-HA mutants, alone or in combination as indicated. Protoplast labeling, immunoselection, and analysis were performed as indicated in A. A_ox, Oxidized untagged monomer; B₁, B₃, untagged dimers; A_ox*, oxidized HA-tagged monomer; B₁*, B₃*, HA-tagged dimers. In both segments, the positions of molecular mass markers (kD) are shown on the right.

Figure 4.

Heterodimer formation in tobacco protoplasts. Tobacco protoplasts were transfected with plasmids encoding the 1B-HA protein or different mutants in which individual Cys residues were mutated to Ser. When indicated, the 1B-HA protein or its Cys mutants (C65S-HA, C278S-HA) were coexpressed with Cys mutants of the untagged B11-33 protein (C25S, C230S, C(25, 230)S). Protoplasts were labeled with [³⁵S]Cys and [³⁵S]Met for 1.5 h. Proteins were immunoselected from protoplast homogenates using an anti-HA monoclonal antibody and analyzed by non-reducing SDS-PAGE. The positions of molecular mass markers (kD) are shown on the left.

Figure 5.

Sedimentation behavior of monomeric and oligomeric forms of the B11-33 protein. Protoplasts were transfected with vector DNA or with plasmids encoding either the wild-type B11-33 protein or the C(25, 230)S mutant. Protoplasts were labeled with [³⁵S]Cys and [³⁵S]Met for 1.5 h. Protoplast homogenates were fractionated by sedimentation velocity centrifugation on Suc gradients. Fractions were collected from the top and proteins were immunoselected with an anti-glutenin antiserum from an aliquot of the total homogenate (T), from the material recovered at the bottom of the tube (P), and from each gradient fraction (1–14). Immunoselected proteins were analyzed by nonreducing SDS-PAGE and fluorography. The approximate position of sedimentation markers is shown at the top. cyt, Cytochrome C; ov, ovalbumin; BSA, bovine serum albumin; ald, aldolase; cat, catalase; fe, ferritin. The positions of molecular mass markers (kD) are shown on the left.

Figure 6.

Role of intrachain disulphide bonds in the assembly of B11-33 LMW-GS. Protoplasts were transfected with vector DNA or with plasmid encoding either the B11-33-HA protein or mutants of this protein in which specific pairs of Cys residues were mutated to Ser (C(134, 169)S-HA; C(142, 162)S-HA; C(170, 280)S-HA). Protoplasts were labeled with [³⁵S]Cys and [³⁵S]Met for 1.5 h. A, Protoplasts were homogenized under denaturing conditions in the absence (−) or in the presence (+) of DTT, as indicated, and proteins were immunoselected from protoplast homogenates using an anti-HA monoclonal antibody. Immunoselected proteins were analyzed by reducing SDS-PAGE. B, Protoplasts were homogenized under denaturing conditions in the absence of DTT, and proteins were immunoselected from protoplast homogenates using an anti-HA monoclonal antibody. Immunoselected proteins were analyzed by nonreducing SDS-PAGE. In both segments, the positions of molecular mass markers (kD) are shown on the left.

Figure 7.

Glutenin polymers are associated with the ER. Protoplasts expressing the B11-33-HA protein were labeled with [³⁵S]Cys and [³⁵S]Met for 1 h, and total homogenates were fractionated on a 16% to 55% Suc gradient. Fractions were collected from the top and proteins were immunoselected with an anti-HA monoclonal antibody from total homogenate (T), from each gradient fraction (1–19), and from the material recovered at the bottom of the gradient (P). The supernatant from the first immunoprecipitation was then subjected to a second round of immunoselection using an anti-BiP antiserum. Immunoselected proteins were analyzed by nonreducing (anti-HA) or reducing (anti-BiP) SDS-PAGE. The positions of molecular mass markers (kD) are shown on the right.

Figure 8.

Stability and intracellular transport of LMW-GS in tobacco protoplasts. A, Protoplasts were transfected with vector alone (vector) or constructs encoding either the B11-33-HA or the C(25, 230)S-HA protein, labeled for 1 h with [³⁵S]Cys and [³⁵S]Met and then chased for the times indicated. Proteins were immunoselected from cell homogenates with anti-HA antibodies and analyzed by reducing SDS-PAGE and fluorography. B, Protoplasts were transfected and pulse-labeled as in A and then chased for the times indicated. After fractionation into a membrane and a soluble fraction as in “Materials and Methods,” proteins were immunoselected with anti-HA or anti-BiP antibodies and analyzed by reducing SDS-PAGE and fluorography. The arrowhead indicates the position of the BiP polypeptide.

Figure 9.

Subcellular localization of glutenin subunits. Tobacco leaf protoplasts were transiently transformed with a plasmid encoding either the B11-33-HA (A–D) or the C(25, 230)S-HA (E–H) proteins and fixed with 4% paraformaldehyde 48 h after trasfection. The subcellular distribution of the two proteins and of the ER marker BiP was examined by confocal laser scanning microscopy using a combination of an anti-HA monoclonal antibody and a rabbit anti-BiP antiserum. A and E, Distribution pattern of the heterologous proteins revealed by the anti-HA and Alexa Fluor 488 goat anti-mouse antibodies. B and F, Reticular fluorescence of endogenous BiP visualized by the anti-BiP and Alexa Fluor 568 goat anti-rabbit secondary antibody. C and G, Chlorophyll autofluorescence. D, Overlaid images: A (green) + B (magenta). H, Overlaid images: E (green) + F (magenta). n, Nucleus; v, central vacuole. Scale bar = 5 μm.