Search

While the antigenic specificity and pathogenic relevance of immunologic reactivity to gluten in celiac disease have been extensively researched, the immune response to nongluten proteins of wheat has not been characterized. We aimed to investigate the level and molecular specificity of antibody response to wheat nongluten proteins in celiac disease. Serum samples from patients and controls were screened for IgG and IgA antibody reactivity to a nongluten protein extract from the wheat cultivar Triticum aestivum Butte 86. Antibodies were further analyzed for reactivity to specific nongluten proteins by two-dimensional gel electrophoresis and immunoblotting. Immunoreactive molecules were identified by tandem mass spectrometry. Compared with healthy controls, patients exhibited significantly higher levels of antibody reactivity to nongluten proteins. The main immunoreactive nongluten antibody target proteins were identified as serpins, purinins, α-amylase/protease inhibitors, globulins, and farinins. Assessment of reactivity toward purified recombinant proteins further confirmed the presence of antibody response to specific antigens. The results demonstrate that, in addition to the well-recognized immune reaction to gluten, celiac disease is associated with a robust humoral response directed at a specific subset of the nongluten proteins of wheat.

Background: Wheat glutenin polymers are made up of two main subunit types, the high- (HMW-GS) and low- (LMW-GS) molecular weight subunits. These latter are represented by heterogeneous proteins. The most common, based on the first amino acid of the mature sequence, are known as LMW-m and LMW-s types. The mature sequences differ as a consequence of three extra amino acids (MET-) at the N-terminus of LMW-m types. The nucleotide sequences of their encoding genes are, however, nearly identical, so that the relationship between gene and protein sequences is difficult to ascertain.It has been hypothesized that the presence of an asparagine residue in position 23 of the complete coding sequence for the LMW-s type might account for the observed three-residue shortened sequence, as a consequence of cleavage at the asparagine by an asparaginyl endopeptidase., Results: We performed site-directed mutagenesis of a LMW-s gene to replace asparagine at position 23 with threonine and thus convert it to a candidate LMW-m type gene. Similarly, a candidate LMW-m type gene was mutated at position 23 to replace threonine with asparagine. Next, we produced transgenic durum wheat (cultivar Svevo) lines by introducing the mutated versions of the LMW-m and LMW-s genes, along with the wild type counterpart of the LMW-m gene.Proteomic comparisons between the transgenic and null segregant plants enabled identification of transgenic proteins by mass spectrometry analyses and Edman N-terminal sequencing., Conclusions: Our results show that the formation of LMW-s type relies on the presence of an asparagine residue close to the N-terminus generated by signal peptide cleavage, and that LMW-GS can be quantitatively processed most likely by vacuolar asparaginyl endoproteases, suggesting that those accumulated in the vacuole are not sequestered into stable aggregates that would hinder the action of proteolytic enzymes. Rather, whatever is the mechanism of glutenin polymer transport to the vacuole, the proteins remain available for proteolytic processing, and can be converted to the mature form by the removal of a short N-terminal sequence.

Starch granule surface-associated proteins were separated by HPLC and identified by direct protein sequencing. Among the proteins identified was one that consisted of two polypeptide chains of 11 and 19 kDa linked by disulfide bonds. Sequencing of tryptic peptides from each of the polypeptides revealed similarities between some of the peptides and avenin-like b proteins encoded by partial cDNAs in NCBI. To identify a contiguous sequence that matched all of the peptides, contigs encoding three avenin-like b proteins were constructed from ESTs of the cultivar Butte 86. All peptide sequences were found in a protein encoded by one of these contigs that had not been identified previously. Protein and DNA sequences indicated that the two polypeptide chains were derived from a parent protein that had been cleaved at the C-terminal position of an asparagine residue. The name farinin is suggested for this protein and other avenin-like b proteins. Evolutionary relationships of the protein are discussed and a simple computer molecular model was constructed. On the basis of its sequence, the new protein was likely to be allergenic but unlikely to be active in celiac disease.

In response to the suggestion that an increase in the incidence of celiac disease might be attributable to an increase in the gluten content of wheat resulting from wheat breeding, a survey of data from the 20th and 21st centuries for the United States was carried out. The results do not support the likelihood that wheat breeding has increased the protein content (proportional to gluten content) of wheat in the United States. Possible roles for changes in the per capita consumption of wheat flour and the use of vital gluten as a food additive are discussed.

Wheat starch is used to make baked products for celiac patients in several European countries but is avoided in the United States because of uncertainty about the amounts of associated grain storage (gluten) proteins. People with celiac disease (CD) must avoid wheat, rye, and barley proteins and products that contain them. These proteins are capable of initiating damage to the absorptive lining of the small intestine in CD patients, apparently as a consequence of undesirable interactions with the innate and adaptive immune systems. In this study, starch surface-associated proteins were extracted from four commercial wheat starches, fractionated by high-performance liquid chromatography and gel electrophoresis, and identified by tandem mass spectrometry analysis. More than 150 proteins were identified, many of which (for example, histones, purothionins, and glutenins) had not been recognized previously as starch-associated. The commercial starches were analyzed by the R-5 enzyme-linked immunosorbent assay method to estimate the amount of harmful gluten protein present. One of these starches had a low gluten content of 7 ppm and actually fell within the range proposed as a new Codex Alimentarius Standard for naturally gluten-free foods (maximum 20 ppm). This low level of gluten indicates that the starch should be especially suitable for use by celiac patients, although wheat starches with levels up to 100 ppm are deemed safe in the proposed Codex standards.

The high-molecular-weight glutenin subunits (HMW-GS) of wheat gluten in their native form are incorporated into an intermolecularly disulfide-linked, polymeric system that gives rise to the elasticity of wheat flour doughs. These protein subunits range in molecular weight from about 70 K-90 K and are made up of small N-terminal and C-terminal domains and a large central domain that consists of repeating sequences rich in glutamine, proline, and glycine. The cysteines involved in forming intra- and intermolecular disulfide bonds are found in, or close to, the N- and C-terminal domains. A model has been proposed in which the repeating sequence domain of the HMW-GS forms a rod-like beta-spiral with length near 50 nm and diameter near 2 nm. We have sought to examine this model by using noncontact atomic force microscopy (NCAFM) to image a hybrid HMW-GS in which the N-terminal domain of subunit Dy10 has replaced the N-terminal domain of subunit Dx5. This hybrid subunit, coded by a transgene overexpressed in transgenic wheat, has the unusual characteristic of forming, in vivo, not only polymeric forms, but also a monomer in which a single disulfide bond links the C-terminal domain to the N-terminal domain, replacing the two intermolecular disulfide bonds normally formed by the corresponding cysteine side chains. No such monomeric subunits have been observed in normal wheat lines, only polymeric forms. NCAFM of the native, unreduced 93 K monomer showed fibrils of varying lengths but a length of about 110 nm was particularly noticeable whereas the reduced form showed rod-like structures with a length of about 300 nm or greater. The 110 nm fibrils may represent the length of the disulfide-linked monomer, in which case they would not be in accord with the beta-spiral model, but would favor a more extended conformation for the polypeptide chain, possibly polyproline II., (Copyright 2005 Wiley Periodicals, Inc.)