In vitro characterisation of ERAP1 mechanism and antigen processing

Abstract

The aminopeptidase ERAP1 is crucial for processing peptides for presentation by MHC class I, starting from sub-optimal N-terminally extended peptides. MHC I and ERAP1 both have genetic associations with autoimmune diseases and the reasons are poorly understood. Both proteins have peptide length and sequence preferences, and which protein ultimately determines the antigenic peptide repertoire is unknown. Presented here are studies through NMR and X-ray crystallography which investigate the importance of individual interactions between MHC I and the peptide, and how NMR can be used to probe the mechanism of ERAP1.

Crystal structures of four single chain constructs based on MHC I allele H-2Kb with A pocket mutations were solved. The mutations caused no difference to structure or peptide binding with respect to the unmutated form, and rearrangement of neighbouring side chains and water molecules in the A pocket were apparent in response to the removal of hydrogen bonding side chains. These structures enabled collaborators to perform assays and computational analysis which showed impaired binding of the peptide in the mutants permitted peptide trimming by ERAP1.

A peptide library was devised and characterised with proton chemical shifts assignments, and the peptides were expressed in E. coli with 13C/15N isotope labelling. HSQC spectra of an isotope-labelled peptide confirmed the proton assignments and offers potential for future experimentation with ERAP1 and MHC I.

From the peptide assignments, proton NMR reaction monitoring was used to comprehensively assay three ERAP1 variants to uncover differences in peptide specificity. This method offers scope to examine multiple steps in an aminopeptidase reaction which commonly used ERAP1 assays do not provide. A series of 1D proton NMR spectra obtained throughout the reaction were found to be an effective way of interrogating ERAP1 aminopeptidase activity. Wild type and 5SNP variant ERAP1 appeared to show different substrate length preferences. ERAP1-peptide binding was analysed with saturation transfer difference NMR and indicated the primary points of interaction were the peptide C-terminal and hydrophobic side chains. This matches findings on ERAP1 peptide preferences by other authors, but the interactions between peptide and ERAP1 have never been directly investigated before and have enabled tentative modelling of ERAP1-peptide binding.

More information

Identifiers

Catalogue record

Export record