SiPep: a system for the prediction of tissue‐specific minor histocompatibility antigens

Abstract

Approximately 50 years ago it was found that inbred strains of mice were able to reject tumours and skin grafts from major histocompatibility complex (MHC) identical donors. They proposed that additional transplantation antigens must exist outside the MHC. These were described as minor histocompatibility antigens (mHAgs). Since then, related studies in humans have identified 16 human mHAgs. The aim of this work is to increase the number of known mHAgs by prediction of candidate minor histocompatibility loci by identifying coding single nucleotide polymorphisms (SNPs) where the amino acid variation lies within an MHC-binding peptide and alters the ability of that peptide to bind. We have developed an algorithm called SiPep which uses peptide sequences derived from the flanking regions of known non-synonymous SNPs, various MHC-binding and proteolytic cleavage evaluation methods and protein expression data to predict mHAgs. We have processed 45094 SNPs using the SiPep algorithm and have stored the results in a database called SNPBinder. The facilities to process submitted proteins through the SiPep algorithm as well as the SNPBinder database are available to the public. A set of peptides that are predicted as possible mHAgs by the SiPep algorithm have been tested using refolding assays and gel filtration and the results are presented in this paper. The SiPep tools and SNPBinder database are available free of charge via the internet. An HTML interface providing search facilities can be found at the following address: http://www.sipep.org/.

Minor histocompatibility antigens as determinants for graft‐versus‐host disease after allogeneic haematopoietic stem cell transplantation

“…Induction of remission by T lymphocytes specific for haematopoietic cell‐specific HA‐1 antigen (Marijt et al ., ) provides a promising example of the potential of minor H antigen mismatching. The minor H antigen loci identified to date have been estimated to represent only a minority of such antigens present in the human genome (Halling‐Brown et al ., ; Armistead et al ., ) as in principle any amino acid change can act as a minor H antigen and lead to alloresponse. The genes encoding for the known minor H antigens are located on several chromosomes and are inherited therefore independently of the HLA genes.…”

Minor histocompatibility antigens: past, present, and future

“…Alternatively, reverse immunology approaches may help in identifying clinically relevant minor H antigens. Datasets, software, and computational power seem to be sufficient to generate candidate minor H antigens . Although three autosomally encoded minor H antigens have been identified by the reverse immunology , the major bottleneck seems to be cellular confirmation of the predicted epitopes.…”

The immunoinformatics of cancer immunotherapy

“…mutations that disable activator proteins). The SiPep web service, which aims to predict mHags, used the dbMHC as well as data from the HapMap project to compute variant peptides (25). In this system, coding nonsynonymous mutations are considered.…”

“…Future applications, however, would do well to consider the other types of relevant mutations. Another mHag prediction system is SNEP (25), which also focuses on coding nonsynonymous SNPs. In this system, the CONFLICT and VARIANT annotations in SWISS‐PROT provide the polymorphism data.…”

“…In the context of a hematopoietic stem cell transplantation, such antigens can cause GvHD (57) as well as the therapeutic GvL effect (24). Finding target antigens that minimize the former and maximize the latter is the goal of immunoinformatic efforts, which combine antigen presentation with additional biological data including polymorphism and gene expression data (25,26). Polymorphism data are necessary to identify gene variants, which could result in the differential expression of peptides.…”