Functional Properties of Wheat Gliadins. I. Effects on Mixing Characteristics and Bread Making Quality

The effects of addition of total gliadin and gliadin subfractions on the 2 g Mixograph parameters and loaf volume data of cv. Hereward base flour were studied. The addition of increasing levels of total gliadin and gliadin subgroups to cv. Hereward base flour decreased the overall dough strength, as evidenced by decreases in the values for mixing time (MT), mixing stability (MS) and work input (WI). The decreasing order of these parameters for different gliadins was: ω1- >, γ-, > β-, > α-gliadins. The mixing tolerance, as measured by resistance breakdown (RBD) and bandwidth breakdown (BWBD), decreased as a result of addition of different fractions. However, Peak dough resistance (PDR) values increased with addition of individual groups of gliadins and gluten to the base flour. A linear relationship was found between the PDR and loaf volume when individual groups of gliadin were added to the base flour. The ω-gliadins produced the least positive effects on PDR. Addition of total gliadin and its subgroups substantially improved loaf volumes of pan breads. The ω-gliadins again resulted in a smaller increase in loaf volume.     

Antigenicity of Wheat Prolamins: Detailed Epitope Analysis using a Panel of Monoclonal Antibodies

Putative continuous epitopes, recognised by five panels of monoclonal antibodies (MAb) with differing specificities for gliadins and glutenin subunits, were identified using overlapping nonapeptides. These peptides corresponded to the entire sequence of an α/β-gliadin, a γ-gliadin, an ω-prolamin (homologous to ω-gliadin), a low molecular weight glutenin subunit (L M_rGS) and several high molecular weight glutenin subunits (HM_r GS). Antibodies that bound to γ- or ω-gliadins, L M_rGS or HM_r GS bound to the peptides at similar concentrations used normally in direct ELISA, but little binding to the peptides was seen for several antibodies that bound specifically to small groups of α/β-gliadins. Epitopes for these antibodies in α/β-gliadin may be discontinuous (i.e. derived from amino acid residues that are brought together by folding of the polypeptide chain or by juxtaposition of two polypeptide chains), since binding of these antibodies to gliadins was greatly decreased following the reduction of intra-molecular disulphide bonds. While some regions in particular subunits were immunodominant, such as the cysteine–cysteine containing peptide found in the central domain of many prolamins, a diversity of reaction patterns was found. Cross-reaction of antibody with peptides from other prolamin families was often due to binding to a peptide having significant sequence homology, but in some cases no homology was obvious. Some major trends were as follows. Antibodies which bound to most or all H M_rGS recognised the central repeat region, while those that were selective for one or two subunits bound to epitopes in the unique N- and/or C-terminal domains. A high proportion of the epitopes recognised by MAb to α-, β-, ω-gliadins and L M_rGS contained cysteine; these MAb may be useful in detecting covalent binding sites within or between subunits. Although a number of MAb bound a wide range of gliadins and GS, several of these recognised single (and differing) epitopes in the target proteins. However, comparatively few MAb recognised epitopes from either the N- or C-terminal regions of the target proteins. Several explanations are possible; either these regions are buried in the immunogen and not accessible for antibody production or alternatively the repeat sequences are immunodominant.     

Effects of α-, β-, γ- and ω-Gliadins on the Dough Mixing Properties of Wheat Flour

Gliadins were extracted from wheat and individual groups (α-, β-, γ-, ω-1 and ω-2) purified. The effects of the individual groups of gliadin on the mixing properties of doughs from low and high protein flours were measured on a 2-g Mixograph and a prototype microextension tester. The addition of all groups of gliadin resulted in a decrease in dough strength. The relative weakening effects were ω-1>ω-2≈α-≈β->γ- in the Mixograph, and γ->α-≈β-≈ω-2≈ω-1 in the Extensograph.     

Effect of Barley and Barley Components on Rheological Properties of Wheat Dough

The effects of addition of whole barley and barley components (starch, β-glucans and arabinoxylans) on rheological properties of dough prepared from wheat flours with variable gluten quality (cv. Glenlea, extra-strong; cv. Katepwa, very strong; cv. AC Karma, strong; and cv. AC Reed, weak) were investigated in these studies using Mixograph and dynamic rheological measurements. Whole barley meal, starch and non-starch polysaccharides from hulless barley with variable starch characteristics (normal, high amylose, waxy, and zero amylose waxy) were tested. Upon addition of either β-glucans or arabinoxylans, significant increases in peak dough resistance, mixing stability, and work input were recorded in all flours. The addition of starch to various wheat flours reduced the strength of the respective flour-water doughs. The improvement of dough strength upon addition of waxy or zero amylose waxy barley meal was associated with the high content of total and soluble β-glucans present in barley samples. The addition of arabinoxylans or β-glucans increased the G′ of wheat doughs; arabinoxylans had a greater effect than β-glucans. Starch substantially decreased the elastic modulus of dough prepared from cv. Glenlea but waxy and high amylose starches increased the G′ of dough prepared from cv. AC Karma. A combination of the high amounts of non-starch polysaccharides and unusual starch characteristics in barley seems to balance the negative effects associated with gluten dilution brought about by addition of barley into wheat flour.     

Salt-Induced Disaggregation/Solubilization of Gliadin and Glutenin Proteins in Water

The effect of salt concentration used in preparing gluten, on the subsequent dissolution of gluten in water, was examined. Flour from a Canadian hard red spring wheat cultivar, Katepwa, was used to prepare glutens using three different solvents, i.e. distilled deionized water (DDW), 0·2% NaCl solution and 2% NaCl solution. The isolated wet glutens were extracted sequentially with DDW, providing four water soluble fractions and an insoluble residue. The amount of protein in each fraction was determined and respective compositions were assessed electrophoretically under reducing and non-reducing conditions. Surprisingly, DDW extracts of gluten prepared with 2% NaCl contained almost all the gliadins, except some ω-gliadin components, and most of the polymeric glutenin. For the gluten prepared with 0·2% NaCl, most of the gliadin, but only a small portion of glutenin, was extracted. For gluten prepared with DDW, only part of the gliadins and almost no glutenin was extractable with water. The DDW solubilities of gluten proteins prepared in DDW, 0·2% NaCl and 2% NaCl were 27, 52, and 85%, respectively, after four sequential extracts with DDW. The large increases in the solubility of gliadin and glutenin proteins in DDW when the gluten is prepared in salt solution (after removal of most of the salt) can be explained on the basis of a salt-induced conformational change of the proteins, which renders water a more effective solvent.     

Influence of sulphur fertilisation on quantities and proportions of gluten protein types in wheat flour

Although different supplies of sulphur (S) during wheat growth are known to influence the quantitative composition of gluten proteins in flour, an effect on the amount and on the proportions of single protein types has yet not been determined. Therefore, wholemeal flours of the spring wheat ‘Star’ grown on two different soils and at four different levels of S fertilisation (0, 40, 80, 160 mg S per container) were analysed in detail using an extraction/HPLC procedure. The results demonstrated that the amount of total gluten proteins as well as of the crude protein content of flour was little influenced, whereas amounts and proportions of single protein types were strongly affected by the different S fertilisation. The changes were clearly dependent on the Cys and Met content of each protein type. The amount of S-free ω-gliadins increased drastically, and that of S-poor high-molecular-weight (HMW) glutenin subunits increased moderately in the case of S deficiency. In contrast, the amounts of S-rich γ-gliadins and low-molecular-weight (LMW) glutenin subunits decreased significantly, whereas the amount of α-gliadins was reduced only slightly. S deficiency resulted in a remarkable shift of protein proportions. The gliadin/glutenin ratio increased distinctly; ω-gliadins became major components, and γ-gliadins minor components, whereas the ratio of HMW to LMW glutenin subunits was well-balanced.     

Effect of hydrostatic pressure and temperature on the chemical and functional properties of wheat gluten: Studies on gluten, gliadin and glutenin

The effect of hydrostatic pressure (0.1–800 MPa) in combination with various temperatures (30–80 °C) on the chemical and physical properties of wheat gluten, gliadin and glutenin was studied. Chemical changes of proteins were determined by extraction, reversed-phase high-performance liquid chromatography (HPLC), sodium dodecylsulphate (SDS) polyacrylamide gel electrophoresis (PAGE), circular dichroism (CD) spectroscopy, thiol measurement and studies on disulphide bonds. Rheological changes were measured by extension tests and dynamic stress rheometry. Treatment of gluten with low pressure (200 MPa) and temperature (30 °C) increased the proportion of the ethanol-soluble fraction (ESF) and decreased gluten strength. The enhancement of both pressure and temperature provoked a strong reduction of the ESF and the thiol content of gluten. Within gliadin types, cysteine containing α- and γ-gliadins, but not cysteine-free ω-gliadins were sensitive to pressure and were transferred to the ethanol-insoluble fraction. Disulphide peptides isolated from treated gluten confirmed that cleavage and rearrangement of disulphide bonds were involved in pressure-induced reactions. Increased pressure and temperature induced a significant strengthening of gluten, and under extreme conditions (e.g. 800 MPa, 60 °C), gluten cohesivity was lost. Isolated gliadin and glutenin reacted differently: solubility, HPLC and SDS-PAGE patterns of gliadin having a very low thiol content were not influenced by pressure and heat treatment; only conformational changes were detected by CD spectroscopy. In contrast, the properties of isolated glutenin having a relatively high thiol content were strongly affected by high pressure and temperature, similar to the effects on total gluten.     

The impact of heating and cooling on the physico-chemical properties of wheat gluten–water suspensions

The rapid visco analysis (RVA) system was used to measure rheological behaviour in 20% (w/v) gluten-in-water suspensions upon applying temperature profiles. The temperature profiles included a linear temperature increase, a holding step, a cooling step with a linear temperature decrease to 50 °C, and a final holding step at 50 °C. Temperature and duration of the holding phase both affected RVA viscosity and protein extractability. Size-exclusion and reversed-phase HPLC showed that increasing the temperature (up to 95 °C) mainly decreased glutenin extractability. Holding at 95 °C resulted in polymerisation of both gliadin and glutenin. Above 80 °C, the RVA viscosity steadily increased with longer holding times while the gliadin and glutenin extractabilities decreased. Their reduced extractability in 60% ethanol showed that γ-gliadins were more affected after heating than α-gliadins and ω-gliadins. Enrichment of wheat gluten in either gliadin or glutenin showed that both gliadin and glutenin are necessary for the initial viscosity in the RVA profile. The formation of polymers through disulphide bonding caused a viscosity rise in the RVA profile. The amounts of free sulphydryl groups markedly decreased between 70 and 80 °C and when holding the temperature at 95 °C.     

The Role of Gluten Proteins in the Baking of Arabic Bread

Fractionation and reconstitution/fortification techniques were utilised to study the role of gluten in Arabic bread. Glutens from two wheat cultivars of contrasting breadmaking quality were fractionated by dilute HCl into gliadin and glutenin. Gluten, gliadin and glutenin doughs from the good quality flour had higher G ′ and lower tan δ values than those from the poor quality flour at all the frequencies examined. Interchanging the gliadin and glutenin fractions between the reconstituted flours showed that the glutenin fraction is largely responsible for differences in the breadmaking performance. Fortification of an average quality flour with the gliadin and glutenin fractions from the poor and good quality flours, at the levels of 1% and 2% (protein to flour mass), induced marked differences in the mechanical properties of bread. The resilience of the loaves was not adversely affected by the addition of gliadins and increased, with a concomitant significant (p<0·05) improvement in quality, at the 2% level of fortification with gliadins from the good quality flour. Addition of glutenin resulted in loaves with leather-like properties that became particularly apparent at the higher level of fortification; the observed deterioration in quality paralleled the increase in the elastic character of the doughs. It is suggested that highly-elastic doughs are not compatible with the rapid expansion of gases at the high-temperature short-time conditions employed in the baking of Arabic bread and that there exists a threshold in dough elasticity beyond which a rapid decline in quality takes place.     

The dynamic rheological properties of glutens and gluten sub-fractions from wheats of good and poor bread making quality

The dynamic rheological properties of glutens and gluten fractions (gliadin and glutenin) of two U.K.-grown wheat cultivars, Hereward and Riband, having good and poor bread quality, respectively, were studied. Gluten and glutenin doughs from cv. Hereward had higher G' and lower tan δ values than those from cv. Riband at all frequencies studied. A more pronounced difference in G' and tan δ was observed between the glutenin doughs of the two wheats than between their respective gluten doughs. The rheological properties, i.e. G' and tan δ values, of gliadin doughs were similar for both wheats. Varying the gliadin/glutenin ratio by adding the isolated gliadin or glutenin sub-fractions to the parent glutens showed that the G' values decreased and the tan δ values increased as the gliadin/glutenin ratio was increased for both cultivars, indicating a considerable decrease in elasticity as the gliadin/glutenin ratio increased. The decrease in G' may be attributed to a plasticising effect of gliadin and ‘interference’ of gliadin with glutenin-glutenin interactions. The reduction in G' was much more pronounced when the gliadin/glutenin ratio was increased between 0.15 and 1.0 than between 1.0 and above. Gluten from cv. Hereward had higher G' and lower tan δ values than cv. Riband gluten at all gliadin/glutenin ratios, indicating that cv. Hereward gluten had greater elastic character than cv. Riband gluten. Although significant effects of other non-protein hydrocolloid components cannot be discounted, these observations are consistent with the view that the viscoelasticity of the glutenin sub-fraction of gluten and differences in the ratio of gliadin to glutenin are the main factors governing inter-cultivar differences in the viscoelasticity of wheat gluten.     

Comparative Studies of High MrSubunits of Rye and Wheat. II. Partial Amino Acid Sequences

A panel of anti-peptide antibodies specific for each of the different N-terminal sequence types of B- and C-low molecular mass glutenin subunits (L M_rGS) were utilised in immunoblotting studies to identify the chromosomal location of genes encoding different sequences and to characterise the allelic variation of the encoding loci. The MET-type sequences were predominantly found among the B- subunits, while the α- and γ- sequences predominated in the C- subunits. The quantitatively major SHIPGLERPS sequence was found in both the B- and C- mobility regions. Using either biotypes in the cultivar, Aroona or genetic lines containing double rye chromosome 1 substitutions and thus expressing only single LM_r GS alleles, the sequences were determined for most of the major polypeptides expressed by each LM_r GS allele. The L M_rGS from different genomes encoded different numbers of each sequence type. Furthermore, different polypeptides within a particular «block» of subunits encoded by a given allele often had differing N-terminal sequences. However, subunits of similar electrophoretic mobilities encoded by different alleles at each locus usually had identical N-terminal sequences, suggesting that they may instead differ in the number of repeats. In Chinese Spring, genes encoding the SHIPGLERPS and METSHIPGL sequence types were predominantly present on chromosomes 1B and 1D, while the related METSRVPGL sequence was only encoded on 1D. In contrast, the METSCIPGL, α- and γ-sequences were encoded on each of chromosomes 1A, 1B and 1D. Several different electrophoretic and immunoblotting approaches using null lines suggested that some of the α-type L M_rGS may also be encoded by group 6 chromosomes, particularly 6D. The anti- SHIPGLERPS antibody also recognised chromosome 1B encoded β-, γ- and ω-gliadins, while the anti-METSRVPGL antibody recognised 1D encoded α- and β-gliadins. The absence of sequences within the major gliadin families that are highly homologous to the latter two N-terminal L M_rGS sequences may suggest that some monomeric L M_rGS could exist within the electrophoretically-resolved gliadins. These antibodies will provide valuable reagents for the study of the roles of particular L M_rGS families in the structure and function of the glutenin macropolymer, the role of different LM_r GS types in determining the influence of allelic variation of L M_rGS composition on dough properties, and potentially in the development of diagnostics for these flour components.     

Effect of hydrostatic pressure and temperature on the chemical and functional properties of wheat gluten II. Studies on the influence of additives

Cysteine, N-ethylmaleinimide, radical scavengers, various salts or urea were added to wheat gluten. After treatment at increasing pressure (0.1–800 MPa) and temperature (30–80 °C) the resulting material was analysed by micro-extension tests and an extraction/HPLC method to measure protein solubility. Furthermore, cysteine was added to isolated gliadin and glutenin prior to high-pressure treatment and protein solubility was determined. The resistance to extension of gluten strongly increased and the solubility of gliadin in aqueous ethanol decreased with increasing pressure and temperature. As compared to experiments without additive the observed effects were much stronger. Isolated gliadin turned largely insoluble in aqueous ethanol when cysteine was added prior to high-pressure treatment. The S-rich α- and γ-gliadins were much more strongly affected than the S-poor ω-gliadins pointing to a disulphide related mechanism. Monomeric gliadin components were completely recovered after reduction of the aggregates with dithioerythritol. In contrast, samples without free thiol groups such as isolated gliadins or with SH groups, which had been blocked by N-ethylmaleinimide, were hardly affected by high-pressure treatment. The addition of radical scavengers to gluten showed no effect in comparison to the control experiment, indicating that a radical mechanism of the high-pressure effect can be excluded. The observed effects can be explained by thiol-/disulphide interchange reactions, which require the presence of free thiol groups in the sample. The addition of salts and urea showed that unfolding of the protein due to weakening of interprotein hydrogen bonds is strongest for ions with a high radius (e.g. thiocyanate). This leads to weakening of gluten at ambient pressure but it facilitates high pressure induced reactions, e.g. of disulphide bonds.     

Characterization of two 1D-encoded ω-gliadin subunits closely related to dough strength and pan bread-making quality in common wheat (Triticum aestivum L.)

Gliadin proteins of 113 common or bread wheat (Triticum aestivum L.) cultivars and advanced lines from China and other countries, were analyzed by high performance capillary electrophoresis (HPCE) and reversed-phase high performance liquid chromatography (RP-HPLC). A major protein peak migrating at 3 min by HPCE and eluting at about 20 min by RP-HPLC was identified in the ω-gliadin region. It was present in cultivars with good pan bread-making quality, whereas most cultivars with poor bread-making quality lacked this protein peak. Quality testing and statistical analysis showed that this ω-gliadin peak was significantly related to dough strength, loaf volume and loaf score. It was separated into two apparent protein components by one-dimensional SDS-PAGE and two-dimensional electrophoresis (2-DE). According to their relative mobilities on the gels, the proteins were designated ω-15 and ω-16, and their accurate molecular masses (42590.5 Da for ω-15 and 41684.1 Da for ω-16) were determined by MALDI-TOF-MS. The ω-15 and ω-16 gliadins possessed the N-terminal amino acid sequences of ARELNPSNKELQQQQ and KELQSPQQQF, and therefore they belonged to 1D-encoded ω-2 type and ω-1 type gliadins, respectively. Both gliadin subunits were always present together among the 86 cultivars analyzed, suggesting that they were encoded by two closely linked genes at Gli-D1 locus. The accumulative characteristics of gliadins during grain development indicated possible additive quantitative effects of ω-15+16 on dough strength. The ω-15 and ω-16 gliadins could be used as valuable genetic markers for wheat quality improvement.     

Silencing of γ-gliadins by RNA interference (RNAi) in bread wheat

RNA silencing is a sequence-specific RNA degradation system that is conserved in a wide range of organisms. The elucidation of the mechanism of RNA silencing has stimulated its use as a reverse genetics tool, because RNA silencing strongly down-regulates the expression of the target gene in a sequence-specific manner. The major protein fraction of wheat grain is gluten which is largely responsible for the functional properties of dough. Gliadins contribute mainly to the extensibility and viscosity of gluten and dough, with the polymeric glutenins being responsible for elasticity. The aim of this work was therefore to silence the expression of specific γ-gliadins by RNA interference, to demonstrate the feasibility of systematically silencing specific groups of gluten proteins. The sequence of a γ-gliadin gene was used to construct the pghp8.1 plasmid. The hpRNA silencing fragment was designed on the basis of 169 base pairs (bp) in sense and antisense orientation with the sequence of the Ubi1 intron as spacer region between the repeats. Two lines of bread wheat were transformed by particle bombardment. Gliadins were extracted from 30 mg of flour, separated by acid-PAGE and determined by matrix-assisted laser desorption/ionization time-of-flight mass spectrometry (MALDI-TOF MS). Seven transgenic lines were obtained and all of them showed reduced levels of γ-gliadins. All seven transgenic plants were fully fertile and their grain morphology and seed weight were comparable to the control lines. MALDI-TOF MS showed that six peaks, present in the untransformed line, were missing in transgenic lines of the BW208 genotype whereas three peaks were missing in the BW2003 genotypes. The proportion of γ-gliadins was reduced, by about 55–80% in the BW208 lines and by about 33–43% in the BW2003 lines. The ELISA assay based on the R5 antibody showed reductions in total gliadins (μg/mg flour) in three of the BW208 lines and in one BW2003 line, but an increase in one BW208 line (C613).     

Effects of α-, β-, γ- and ω-Gliadins on the Dough Mixing Properties of Wheat Flour

Gliadins were extracted from wheat and individual groups (α-, β-, γ-, ω-1 and ω-2) purified. The effects of the individual groups of gliadin on the mixing properties of doughs from low and high protein flours were measured on a 2-g Mixograph and a prototype microextension tester. The addition of all groups of gliadin resulted in a decrease in dough strength. The relative weakening effects were ω-1>ω-2≈α-≈β->γ- in the Mixograph, and γ->α-≈β-≈ω-2≈ω-1 in the Extensograph.     

Disulphide Bonds in Wheat Gluten Proteins

Disulphide bonds play a key role in determining the structure and properties of wheat gluten proteins. Comparison of the sequences of monomeric gliadins and polymeric glutenin subunits allows the identification of conserved and variant cysteine residues. Direct disulphide bond determination demonstrates that the conserved cysteine residues present in S-rich prolamins (α-type gliadins, γ-type gliadins and LMW subunits) form intra-chain disulphide bonds while additional cysteines residues present only in the LMW subunits form inter-chain bonds with cysteines in HMW subunits and other LMW subunits. Conserved and variant cysteine residues are also present in the HMW subunits but their patterns of disulphide bond formation are less well understood. Further information on the abilities of individual cysteine residues to form intra- and inter-chain disulphide bonds has also been obtained by heterologous expression of wild type and mutant proteins inE. coliand, in the case of the HMW subunits, by examination of the patterns of dimers recovered on partial reduction of glutenin or resulting from the expression of subunits in transgenic tobacco plants. Wheat gluten proteins are folded and assembled within the lumen of the endoplasmic reticulum of the developing endosperm cells, where disulphide bond formation and exchange may be catalysed by the enzyme protein disulphide isomerase. Similarly, disulphide bond reduction, for example to facilitate mobilisation during germination, may be catalysed by thioredoxinh. Understanding the mechanism and specificity of disulphide bond formation in gluten is crucial for the manipulation of its functional properties by genetic engineering or chemical modification.     

Heat and additive induced biochemical transitions in gluten from good and poor breadmaking quality wheats

Glutens from poor breadmaking quality wheat, cv. Riband, had a higher SDS extractability than glutens from good quality cv. Hereward. Heating of gluten, especially above 70 °C, caused a reduction in the amount of SDS-extractable gluten proteins. Treatment of gluten with redox additives (ascorbic acid, potassium bromate or glutathione) affected extractability, being highest for bromate treated glutens. The SH content of gluten was lower for poor breadmaking Riband and heating resulted in greater decrease in SH content of gluten from good breadmaking Hereward. Hereward gluten had a higher SS content than Riband. The alteration of SS content on heating was not significant and may indicate the heat-induced involvement of non-covalent interactions. SDS-PAGE revealed that oxidants, especially bromate, affect polypeptide composition leading to a more heat stable/tolerant protein structure.     

Rheological Behaviour of Wheat Glutens at Small and Large Deformations. Effect of Gluten Composition

Glutens derived from two wheat cultivars with a known difference in bread making quality, i.e. cv. Katepwa (good) and cv. Obelisk (poor), were fractionated into gliadin and glutenin. Cultivar Katepwa gluten contained more glutenin than cv. Obelisk gluten. Reconstituted glutens were prepared by mixing, in different ratios, gliadin and glutenin fractions that originated from one gluten type or from both glutens. The rheological properties of these mixtures, when hydrated, were studied at small deformations in shear and at large deformations in biaxial extension. The reconstitution of gluten in its original glutenin/gliadin ratio produced a composite that had a somewhat higher resistance to deformation and was more elastic than the unfractionated gluten. This was true for both gluten types. However, the difference between the rheological behaviour of both reconstituted gluten types was comparable with that found between the native glutens. From measurements with glutens reconstituted at various glutenin/gliadin ratios, it appeared that the main factor determining the rheological behaviour of hydrated gluten is the glutenin/gliadin ratio. By interchanging the gliadin and glutenin fractions of the two glutens, it was shown that the source from which the fractions originated, particularly that of the glutenin fraction, was also important.     

Protein bodies ontogeny and localization of prolamin components in the developing endosperm of wheat caryopses

During caryopsis development, prolamins are initially stored in individual protein bodies, then generate a protein matrix in the ripe caryopsis. The ontogeny of the protein bodies was analyzed by fluorescence and electron microscopy from 7 to 43 days after anthesis (dAA), a period of time from the cellularization of endosperm to its desiccation. A series of antibodies specific to each prolamin type (α/β-, γ-, ω-gliadins, low-molecular weight and high-molecular weight glutenin subunits) made it possible to localize and co-localize the different prolamins in organelles of endosperm cells at different developmental stages. Protein bodies containing prolamins were observed as early as 7 dAA. At the early developmental stages, protein bodies were spherical with diameters around 1–2 μm. Later, around 15 dAA, the PBs enlarged, and aggregation and/or coalescence were prominent at 21 dAA. From 33 dAA, individual PBs were no longer visible, but a protein matrix was confined in the space between starch granules. All prolamins were found in the same protein bodies, without any segregation according to their types. Immunochemical labelling of prolamins failed to reveal in TEM analyses any particular internal organization in protein bodies. Glutenin subunits and gliadins were observed in the Golgi apparatus at the early stages of endosperm development.