Skip to main content
Premium Trial:

Request an Annual Quote

Oslo Team Uses Mass Spec to Characterize Gluten Peptide Binding in HLAs Linked to Celiac Disease


NEW YORK(GenomeWeb) –University of Oslo researchers have used mass spectrometry to characterize the gluten peptide binding motifs of human leukocyte antigen (HLA) types associated with celiac disease.

Detailed in a paper published last month in Immunogenetics, the study provides basic scientific insights into processes underlying celiac disease and could also prove useful in developing therapies for the disease, Ludvig Sollid, a University of Oslo researcher and senior author on the paper, told GenomeWeb.

Celiac disease is an autoimmune disorder in which the body generates an immune response to gluten proteins found in grains including wheat, rye, and barley. In celiac patients, HLA molecules – which are responsible for presenting antigens to T cells – bind and present gluten peptides as foreign antigens, triggering the inappropriate immune response characteristic of the disease.

Nearly all celiac sufferers possess one of four HLA types: DQ2.5, DQ2.2, DQ8, or DQ7.5. Each of these variants present gluten peptides to T cells, but each bind and present different repertoires of gluten peptides, and each are associated with different severities of celiac disease.

Study of these peptide repertoires has been significantly enabled in recent years by advances in mass spec technology, Sollid noted.

"What has been a game changer in this area has been the sensitivity of the mass spec instruments, and also that you can do relative quantitation of peptides to look at abundance as well," he said.

One issue facing the Oslo team was the fact that, while most mass spec-based proteomics tools are built with searching tryptic peptides in mind, the endogenous gluten peptides they targeted in their work were not tryptic.

"It was a challenge," Sollid said. "But with the MS/MS spectra you can get decent sequencing information, and coupling that with a good database it was possible to assign many of the spectra we got."

At the beginning of the project the researchers limited their search space to secreted proteins in order to reduce the computation demands of the process, he said, adding that "during the course of the study we got access to better computing capacity, and then we redid the analysis [to search] the whole human proteome."

Using a Thermo Fisher Q Exactive, the researchers looked at peptides eluted from DQ2.5, DQ2.2, and DQ7.5 molecules affinity purified from a panel of nine cell lines. They identified 4,267 peptides in the eluates of the DQ2.5 molecules, 7,395 peptides in the eluates from the DQ2.2 molecules, and 7,380 in the eluates from the DQ7.5 molecules.

Looking at the variations in the peptides bound to each HLA type, Sollid and his colleagues found that the identity and quantity of peptides varied significantly between types. This, he said, was expected in the case of DQ7.5, which researchers had previously determined to have different binding motifs than the DQ2.2 and DQ2.5 molecules.

Researchers had generally thought that these two DQ2 molecules, on the other hand, had similar binding motifs given their high level of homology. The Oslo team's analysis, however, demonstrated that they, in fact bound rather different peptide repertoires, and, by using the neural network tool NNAlign, they were able to determine that the molecules had distinct peptide binding motifs.

"The differences in the [DQ2] proteins are relatively small, and for that reason you might think the [peptide] binding might be much the same," Sollid said. "But the analysis indicated that they are quite different, and looking at the sequences of the peptide repertoires using neural analysis, we concluded that [the DQ2 proteins] have distinct binding motifs." 

The DQ proteins' different binding motifs and, consequently, different peptide repertoires are associated with different levels of risk of developing celiac, Sollid noted. DQ2.5, for instance, is strongly associated with the disease, while DQ2.2 has only a weak association.

Knowledge of these binding motifs will help researchers identify other gluten peptides that play a role in driving to immune response in celiac disease, Sollid said. "We know what [structural characteristics] they should carry to bind to these HLA molecules and be presented to T-cells.

While this is interesting from a basic research perspective, it could also be valuable for development of therapies for the disease, he noted.

One of the main therapeutic approaches currently being explored in celiac disease by companies such as Cambridge, Mass-based immunotherapy firm ImmusanT is peptide vaccination, wherein the goal is to introduce gluten peptides to patients so as to reeducate their immune systems not to generate a response to these antigens.

The findings in the Immunogenetics paper, suggest the importance of selecting the right peptides for a patient's specific DQ type, Sollid said. "It suggests you have to tailor-make the therapy based on which HLA molecule the patient is carrying."

Looking forward, he said he and his colleagues were interested in using mass spec to identify additional gluten peptides involved in celiac.

"There are still gluten T-cell epitopes we don't know the identity of," Sollid said, adding that they could use HLA molecules as an affinity matrix for capturing peptides that they would then identify via mass spec.

"That is something we will perhaps pursue to get a more complete list of which parts of gluten are harmful to people with celiac disease," he said.