Abstract:
In this thesis several mass spectrometric methods were used within the scope of different projects for the identification of MHC class I ligands. One main objective of this dissertation was the development and validation of two differential approaches, acetylation and a combination of guanidinylation and nicotinylation, for a quantitative analysis of HLA ligands from two different sources by mass spectrometry. The application of the acetylation method to a tumor/normal tissue pair derived from a colon carcinoma patient led to the identification of two HLA ligands overpresented in tumor tissue and derived from the ribosomal protein L24 and the tumor-associated protein beta-catenin. By applying the guanidinylation/ nicotinylation method on an Awells/Keratin 18 transfectant a new HLA-A*0201 ligand derived from the tumor marker Keratin 18 was identified. Also, for the first time an MHC class I ligand, which was derived from preproinsulin in tranfected PRIESS cells was characterised. Additionally, the relative mRNA expression of a tumor/normal tissue pair of a renal cell carcinoma patient was compared to the relative HLA ligand presentation, which can be used in further studies for the evaluation of a correlation between mRNA expression and HLA ligand presentation.
Furthermore, numerous tumor ligands from different solid tumors, such as colon carcinoma, stomach carcinoma and glioblastoma were identified by mass spectrometric techniques. In the case of renal cell carcinoma a clinical study was initiated in which patient-specific vaccines based on HLA ligand identification and gene expression data were applied.
Systematic searches for disease-associated HLA ligands were performed using the predict, calibrate, detect method. These analyses led to the identification of an HLA ligand of the malaria-associated MSP-1 protein and for the first time to the characterisation of two hepatitis-C-virus epitopes on a molecular level.
The peptide motif of the woodchuck MHC molecule WLA.1-1 as a model system for hepatitis-B-virus infection was determined for T cell epitope studies. In addition, the peptide motifs of the HLA alleles A*2601, A*24, B*0702, and B*4402 were refined by newly identified natural HLA ligands, enabling an optimized T cell epitope prediction.