Functional Interaction of the Ankylosing Spondylitis–Associated Endoplasmic Reticulum Aminopeptidase 2 With the HLA–B27 Peptidome in Human Cells

• Published in: Arthritis & rheumatology (Hoboken, N.J.) N.J.), 2016-10, Vol.68 (10), p.2466-2475
• Main Authors: Martín‐Esteban, Adrian, Guasp, Pablo, Barnea, Eilon, Admon, Arie, López de Castro, José A.
• Format: Article
• Language: English
• Subjects: Aminopeptidases - genetics; Aminopeptidases - metabolism; Aminopeptidases - pharmacology; Blotting, Western; Cell Line; Endoplasmic reticulum; Genotype; HLA-B27 Antigen - metabolism; Humans; Ligands; Lymphocytes - metabolism; Minor Histocompatibility Antigens - genetics; Minor Histocompatibility Antigens - pharmacology; Peptides; Peptides - metabolism; Polymerase Chain Reaction; Recombinant Proteins; Spondylitis, Ankylosing - metabolism; Tandem Mass Spectrometry
• ISSN: 2326-5191; 2326-5205
• DOI: 10.1002/art.39734

Objective To determine the influence of endoplasmic reticulum aminopeptidase 2 (ERAP‐2) expression on the HLA–B*27 peptidome in live cells. Methods Using immunoaffinity chromatography and acid extraction, HLA–B*27:05–bound peptides were isolated from 2 ERAP‐2–negative lymphoblastoid cell lines and 1 ERAP‐2–positive lymphoblastoid cell line expressing functionally indistinguishable ERAP‐1 variants. More than 2,000–4,000 B*27:05 ligands were identified from each cell line, and their relative abundance was established by quantitative tandem mass spectrometry and MaxQuant‐based peptide analyses. Pairwise comparisons were used to determine the structural features of peptides whose relative abundance was dependent on the presence of ERAP‐2. Synthetic peptide digestions were performed with recombinant ERAP‐1 and ERAP‐2. Peptide affinity was estimated with standard algorithms. Results The B*27:05 peptidome from ERAP‐2–positive cells showed 3–4% fewer peptides with N‐terminal basic residues than did the peptidome from ERAP‐2–negative cells. Among the shared peptides, those most abundant in the presence of ERAP‐2 included more nonamers, fewer decamers, and fewer N‐terminal basic residues than the peptides predominant in ERAP‐2–negative cells. These ERAP‐2–dependent changes did not alter the global affinity of the B*27:05 peptidome. Conclusion ERAP‐2 significantly influences the B*27:05‐bound peptidome by destroying some ligands and decreasing the abundance of many more ligands with N‐terminal basic residues, while increasing the abundance of nonamers. The former effects are best explained by direct ERAP‐2 trimming. The effects on peptide length might be attributed to ERAP‐2–induced activation of ERAP‐1 trimming. These data support the notion of a peptide‐mediated mechanism as the basis for the association of ERAP‐2 with ankylosing spondylitis. Analogous effects on other major histocompatibility complex class I peptidomes might explain the involvement of ERAP‐2 in HLA–B27–negative spondyloarthritis.