MHC Peptidomics: HLA-B*27 vs Mamu-B*08


Indian rhesus macaques are arguably the most reliable animal models in AIDS research. In this species the MHC class I allele Mamu-B*08, among others, is associated with elite control of SIV replication. A similar scenario is observed in humans where the expression of HLA-B*27 or HLA-B*57 has been linked to slow or no progression to AIDS after HIV infection. Despite having large differences in their primary structure, it has been reported that HLA-B*27 and Mamu-B*08 display peptides with sequence similarity. To fine-map the Mamu-B*08 binding motif and assess its similarities with that of HLA-B*27 we affinity purified the peptidomes bound to these MHC class I molecules and analyzed them by LC-MS/MS identifying several thousands of endogenous ligands. Sequence analysis of both sets of peptides revealed a degree of similarity in their binding motifs, especially at peptide position 2 (P2) where arginine was present in the vast majority of ligands of both allotypes. In addition, several differences emerged from this analysis: (i) ligands displayed by Mamu-B*08 tended to be shorter and to have lower molecular weight, (ii) Mamu-B*08 showed a higher preference for glutamine at P2 as a suboptimal binding motif and (iii) the second major anchor position, found at P-omega, was much more restrictive in Mamu-B*08. In this regard, HLA-B*27 bound efficiently peptides with aliphatic, aromatic (including tyrosine) and basic C-terminal residues while Mamu-B*08 preferred peptides with leucine and phenylalanine in this position. These results deepen our understanding of the molecular basis of the presentation of peptides by Mamu-B*08 and can contribute to the detection of novel SIV epitopes restricted by this allotype.

Sample Processing Protocol

Lysates of stable transfectants of HLA-B*27 and mamu-B*08 in the HMy2-C1R (C1R) cell line were subjected to immunoprecipitation with the mAb W6/32 and acid extraction of the bound peptide pool before LC-MS/MS analysis. As a negative control, the same protocol was applied to untransfected C1R cells.

Data Processing Protocol

MS/MS data were converted to mgf files with the PeakView software (ABSciex, version 1.1) and searched against a concatenated target-decoy database containing the 88,293 Uniprot entries of the Homo sapiens complete proteome set (downloaded on October 21, 2013) and their corresponding reversed sequences. Four independent search engines were used: MASCOT (Matrix Science, version 2.5), OMSSA (National Center for Biotechnology Information, version 2.1.9), X!Tandem2 (The Global Proteome Machine Organization, version win-13-02-01-1), and X!Tandem2 with k-score plugin (LabKey Software, version 2.3-7806). Search parameters were set as follows: no enzyme, MS tolerance: 0.01 Da, MS/MS tolerance: 0.02 Da and oxidation of methionine, protein N-terminal acetylation and pyroglutamic acid formation from N-terminal glutamine as variable modifications. The individual outputs of the search engines were combined by converting each engine-specific scoring scheme to a common probability-based scale as previously described (Ramos-Fernández et al. Mol Cell Proteomics. 2008.7(9):1748-54). A subsequent metascoring step was applied to increase the rate of peptide identifications and on-the-fly stress tests were employed to guarantee unbiased, highly accurately FDR estimates


Miguel Marcilla, Proteomics Unit - Dpt. of Macromolecular Structures - Spanish National Biotechnolgy Centre (CNB-CSIC)
Alberto Paradela, Proteomics Unit. Spanish National Biotechnology Centre (CNB-CSC). Madrid. Spain. ( lab head )

Submission Date


Publication Date



Not available


TripleTOF 5600


Not available


Not available

Experiment Type

Shotgun proteomics


    Marcilla M, Alvarez I, Ramos-Fernández A, Lombardía M, Paradela A, Albar JP; Comparative Analysis of the Endogenous Peptidomes Displayed by HLA-B*27 and Mamu-B*08: Two MHC Class I Alleles Associated with Elite Control of HIV/SIV Infection., J Proteome Res, 2016 Mar 4, 15, 3, 1059-69, PubMed: 26811146