Protein engineeringV. Zoete and O. Michielin
Development of in silico approaches for protein engineeringV. Zoete and O. Michielin
Computational molecular modelling methods provide an insight on the structural and energetic effect of mutations on protein-protein association and on protein structural stability. Free energy simulations are the most accurate methods for studying these effects. However, despite increased computational power, these methods are still quite time consuming and cannot be applied to study the role of a large number of residues. Among the simplified approaches that have been developed to address this issue, the molecular mechanics – Generalized Born surface area (MM-GBSA) approach is one of the most promising and widely used to estimate the importance of each residue on protein-protein association, opening the way to the rational design of engineered proteins. We have implemented this approach using the CHARMM package and the efficient GB-MV2 solvation model. Also, we have developed a new intuitive approach to decompose approximately the vibrational entropy term into atomic contributions, increasing significantly the correlation between the calculated and experimentally determined free energies (Zoete V and Michielin O, Proteins 2007, 67, 1026). New methodological developments are ongoing to increase the efficiency and speed of the method. We are actually applying this approach to design modified TCR (see further). Also, we continue the improvement of the relative approach that has been developed by V. Zoete and M. Meuwly to study the impact of mutations on protein structural stability and determine the most important residues for the protein fold. This approach has been applied successfully to the study of insulin (Zoete V and Meuwly M, J. Comput. Chem. 2006, 27, 1843), p53 (Yip YL, Zoete V, Scheib H and Michielin O, Hum. Mutat. 2006, 27, 926) and PPAR (Michalik L, Zoete V, Krey G, Grosdidier A, Gelman L, Chodanowski P, Feige JN, Desvergne B, Wahli W and Michielin O, J. Biol. Chem. 2007, 282, 9666) structural stability.
Rational optimization of TCRV. Zoete, M. Irving, M. Ferber, M. Andre, and O. Michielin, in collaboration with P. Romero, D. Speiser, D. Hacker, N. Rufer and V. Cerundolo
Recognition by the T-cell receptor (TCR) of immunogenic peptides (p) presented by class I major histocompatibility complexes (MHC) is the key event in the immune response against virus infected cells or tumor cells. The major determinant of T cell activation is the affinity of the TCR for the peptide-MHC complex, though kinetic parameters are also important. A study of the 2C TCR/SIYR/H-2Kb system using a binding free energy decomposition based on the MM-GBSA approach (see above) has been performed to assess the approach on this system. The results show that the TCR-p-MHC binding free energy decomposition including entropic terms provides a detailed and reliable description of the interactions between the molecules at an atomistic level (Zoete V and Michielin O, Proteins 2007, 67, 1026). We will now apply the method to study of the interactions between TCR and the histocompatibility leukocyte antigens (HLA)-A2 tumor epitope NY-ESO-1. NY-ESO-1 is a cancer testis antigen expressed not only in melanoma, but also on several other types of cancers. It has been observed at high frequencies in melanoma patients with unusually positive clinical outcome and, therefore, represents an interesting target for adoptive transfer with modified TCR. Sequence modifications of TCR potentially increasing the affinity for this epitope will be proposed based on theoretical approaches and tested experimentally, opening the way to adopted transfer-based immunotherapy.
Modelling of the TCR-peptide-MHC complex. Study of the structural properties of the TCR repertoire specific for selected tumor associated peptidesM. Ferber, A. Leimgruber, V. Zoete and O. Michielin, in collaboration with P. Romero, D. Speiser, N. Rufer and V. Cerundo
Following the determination in 1996 of the first crystal structure of a T cell receptor (TCR) bound to its ligand, an antigenic peptide (p) complexed with the major histocompatibility complex (MHC) of class I, our knowledge of the TCR recognition principles has been greatly enhanced. Due to the central role of the TCR-p-MHC interaction in immunology, and to the advent of techniques like p-MHC tetramers that allow the extraction from blood of all lymphocytes whose TCRs are specific for a given p-MHC, researchers have generated large numbers of TCR sequences that require structural information to be interpreted. Our general aim is to build an expert system to produce high quality models of TCR-p-MHC for any given experimental sequence, and to be able to study energetic and structural properties from these models. Our approach makes use of homology modelling for known TCR-p-MHC sequences. It is based on a continuously growing set of resolved crystal structures of TCR-p-MHC. This approach is complemented by ab initio predictions for the variable regions, like the CDR loops of the TCR that are known to make the key interactions with the peptide-MHC. These predictions are taking into account the existence of canonical structures for CDR loops, which allows one to improve the modelling process by adding conformational restraints specific to a given CDR loop. Comparison of the TCR-p-MHC models allows the detection of conserved structural motifs that are not detectable at the sequence level. These data are used to understand the rules governing the selection of the TCR immune repertoire.
Recently we were able to use a large number of TCR sequences that were provided through a close collaboration with the Ludwig Institute for Cancer Research and the Lausanne University Hospital (CHUV). These TCR are known to be specific for particular antigenic peptides (Melan-A and NY-ESO-1) relative to some melanoma cancers. Currently we are building models from theses sequences, in order to study the affinity of the specific TCR repertoires.