Karine Murat
Conformational transitions have found broad interest due to their impact on protein misfolding and self-assembly as key events leading to the development of neurodegenerative diseases. However, investigation of these dynamic events has been limited so far mainly due to the intrinsic tendency of the involved polypeptides for self-association and aggregation. Consequently, the elucidation of β-sheet and fibril-forming processes related to degenerative diseases remains a difficult task, often giving contradictory results. In order to overcome these difficulties and to investigate early steps of misfolding and subsequent fibril and plaque formation, a new concept termed "Switch-peptides" is presented in this thesis, opening new perspectives for the rational design of drugs preventing Alzheimer's disease. Figure. Elaboration of the "Switch-peptides" concept via S to N acyl migration. As shown in the Figure, the switch-peptides represent stable, self-contained folding precursors, in which folding and self-assembly is blocked by the presence of the Ser-, Thr-, Cys- derived switch elements S dissecting the regular peptide backbone by an (thio)ester and a flexible C-C bond (Soff). Removal of the protecting group Y by enzymatic cleavage or by change of pH, triggers X,N acyl migration restoring the native peptide backbone in situ, and setting off the folding process in the nascent state ("in statu nascendi", ISN) of the parent molecule. These resulting conformational changes i.e. from a random coil to a folded state, can be used to study the onset of biological activity (A) onset (B) and disruption (C) of secondary and tertiary structures. The work described in the present thesis elaborates the structural and chemical foundations of the concept, notably in the induction of structural and conformational properties of peptides. Special attention is focused on the S to N acyl migration in elaborating an in situ ligation methodology. For potential applications in vitro and in vivo, we explore a new thioester solubilizing group to realize efficient native chemical ligations at physiological conditions (Figure). Besides the synthetic access of switch-peptides thioesters via solid-phase techniques, the impact of internal (steric hindrance) and external pH factors on the kinetics of the intramolecular acyl migration is evaluated by analytical HPLC. The major applications of the switch-peptides concept are explored via three main axis : In order to settle the foundations of strategy B, our study was based on the possibility of inducing in situ the formation of secondary and tertiary structures. For this purpose, an amphiphilic model peptide is designed to mimic conformational transitions relevant in degenerative disease, i.e. transitions of type random coil to β sheet. The role of the switch element S in the disruption of secondary structures due to the conformational decoupling of σ on P (Soff), is demonstrated by CD.The induced conformational transition to β sheet conformations is accompanied by a significant decrease in peptide solubility. Moreover, the controlled release of nucleophile within the S element can modulate the rate of structure transition over a broad time scale, as demonstrated by kinetic studies. The use of orthogonal protecting groups Y allows the sequential and individual triggering of multiple elements S, localized in strategic positions in more complex polypeptides. With this objective in mind, the concept is expanded to the induction of ββα supersecondary structures such as the Zinc finger domain (Zif). Actually, the simultaneous introduction of a chemically (pH) and enzymatically (DPPIV) cleavable group at the N- and C-terminus in a sequence derived of Zif1 268 permited the consecutive onset of secondary and tertiary structure by a hierarchical triggering of acyl migrations. The second part of the thesis is devoted to the elaboration of a new generation of β breaking switch-peptides (Figure, strategy C). For the design of such inhibitors, the concept of "separating" recognition (σ inactivated) and functional (σ activated) states appears essential. To construct molecules able to switch between recognition and functional state, we assemble different pseudoproline (ψ Pro)-containing building blocks as induction units σ for β-sheet breaking and a target peptide P. The impact of the ψ Pro-moiety is decoupled from the peptide chain via a flexible thioester bond (Soff) and set off after acyl migration (Son). As a consequence, the insertion and association into existing β-sheet templates should be promoted. Preliminary results by EM studies indicate that a controlled enzyme-induced acyl migration results in a significant improvement of the β-sheet breaking potential. For example, one of the designed switch-peptides slowed down the fibrillogenesis of an amyloid model, representing a significant step in the developement of therapeutically relevant compounds in Alzheimer's disease. The third part focuses on the in situ creation of structure and function of potentially bioactive peptide ligands (Figure, strategy A) via a novel ligation method ("Switch-ligation"). For the example of an angiotensin II-derived analogue, we explore the potential of chemoselective ligations for the onset of biological function under physiologically relevant conditions. Receptor (AT1) binding studies of [Cys5]Ang II in collaboration with PD. Dr. E. Grouzmann, CHUV reveal a fast and efficient ligation without the need of additives (IC50= 25 nM). As an alternative example, the covalent self-assembling of two precursors results in the in situ creation of an efficient peptide deformylase inhibitor. The perspective to extend these methodologies for in vivo applications has been tentatively explored as a new concept in prodrug design. In conclusion, the results of this thesis contribute substantially to the proof of concept of switch-peptides, opening new perspectives in drug design as well as in the study of peptide and protein misfolding relevant in degenerative diseases.
EPFL2006
