Synthetic amyloids as a new generic platform for PET imaging. From infectious disease to cancer imaging.

Experimental data demonstrated that amyloid-like aggregation of proteins is induced by short amyloidogenic Aggregation Prone Regions (APRs) within a protein sequence. These APRs are able to self-assemble into β-sheets, resulting in the typical cross-β structured backbone of amyloids. These amyloidogenic sequence segments normally consist of 5-15 amino acids and contain hydrophobic amino acids of low net charge. It was demonstrated that such APRs are sufficient to induce amyloid-like aggregation and are present in the vast majority of naturally occurring proteins, often buried inside the hydrophobic core of the folded protein. When APRs are solvent-exposed, self-assembly takes places via intermolecular association of APRs. The inclusion of non-homologous sequences into the densely in-register stacking of otherwise identical side chains is sterically impeded. Proteins will therefore usually aggregate predominantly with identical proteins because of this sequence specificity of amyloid aggregation. The β-sheet propensity of a certain peptide segment can be determined by APR prediction algorithms, such as the TANGO algorithm. Previously, it was established that APRs can be used to develop synthetic amyloidogenic peptides called Pept-insTM. Pept-in stands for "Peptide Interferor" and they have a maximum length of around 20 amino acids. Their sequences are based on APRs present in a target protein, which are identified via the TANGO algorithm. Instead of ligand-receptor interaction or epitope-antibody binding, Pept-ins interact with their target through their APR via highly specific β-sheet aggregation interactions. Pept-ins have previously been investigated for target-specific knockdown by inducing the aggregation of the target protein, suitable for agricultural and therapeutic applications in plants, prokaryotes and eukaryotes and for in vitro detection purposes. In this Doctoral Thesis, the potential of Pept-ins for in vivo diagnostic imaging was explored. More specifically, the use of Pept-ins radiolabelled with positron-emitting radionuclides for positron emission tomography (PET) imaging of cancer and infection was evaluated. Radionuclide imaging methods such as PET enable non-invasive imaging of biological processes and diseases, employing the high affinity and high selectivity interaction of radiolabelled compounds with their target in vivo. Radiolabelled vascin, [68Ga]Ga-NODAGA-PEG4-vascin, and radiolabelled P2, [68Ga]Ga-NODAGA-PEG2-P2, were evaluated as probes for PET imaging of cancer and infection, respectively, and showed good target-to-background PET image contrast. Vascin targets VEGFR2, a receptor important in tumour angiogenesis, and P2 induces a lethal aggregation cascade in E. coli. They were shown to accumulate in vivo at their target site, a phenomenon that is abolished when the APRs in the Pept-ins are disturbed by mutation of residues to the β-sheet structure breaking proline. These data suggest that the observed in vivo specificity depends on the Pept-in APR interaction with the target protein and underlines the importance of the specific design of Pept-ins. Furthermore, to exclude that the observed in vivo accumulation occurs through non-specific accumulation of self-aggregates via the enhanced permeability and retention (EPR) effect, cross-validation studies were performed. When radiolabelled P2 was injected into the mouse melanoma tumour model of vascin, no tumour accumulation was observed. Accordingly, when radiolabelled vascin was injected into the P2 mouse E. coli muscle infection model, no specific in vivo accumulation in infected tissue was detected. Pharmacokinetic studies proved fast clearance of radiolabelled vascin and P2 from healthy tissues. The promising in vivo specificity and pharmacokinetic properties of the radiolabelled Pept-ins observed in this work, in two different disease models, highlights the potential of a platform based on this technology for diagnostic applications. This work also validates the potential to exploit the specificity of amyloid-like self-assembly interactions for in vivo targeting purposes. The Pept-in sequence is based on that of the target protein, providing an easy rational design approach using the TANGO algorithm to identify APRs upfront. The omnipresence of target-specific APR segments in natural proteins adds a potential generic nature to a platform based on the Pept-in technology, similar to antibodies and derivatives. Pept-ins are produced by solid-phase chemistry technology, which is straightforward and fast, at a low production cost and high production throughput. Both vascin and P2 engage their target in the intracellular environment, profoundly extending the range of imaging disease markers beyond extracellular targets, as is the case for current radiotracers based on peptides, antibodies and antibody derivatives. PET imaging with radiolabelled VEGFR2-targeting vascin showed a high tumour-to-background contrast in the PET images, due to the remarkably increasing accumulation at the tumour site over the 3-hour scanning period. Subsequently, radiolabelled vascin could have potential as cancer PET imaging probe for clinical use. Radiolabelled P2 provided a clear distinction between bacterial infection and sterile inflammation, demonstrating the use of anti-bacterial Pept-ins as a new approach for molecular imaging of infectious disease. Considering the high and persisting accumulation in tumour tissue observed in the PET time-activity curves of vascin, Pept-ins labelled with α- or β-emitting radionuclides may have potential for radionuclide therapy.

Jaar van publicatie:2018

Toegankelijkheid:Open