NMR and EPR methods for localizing peptide ligands
Lead Chief Investigator: Gottfried Otting, The Australian National University
Collaborating Chief Investigator: Richard Payne, The University of Sydney
Summary
Single-crystal X-ray structures present the gold-standard of structure determination of protein-ligand complexes but critically depend on the availability of single crystals and their capacity to diffract in the X-ray beam. In this project, new NMR and EPR methods are sought to identify the binding site of a high-affinity peptide ligand on its target protein.
Relevance to the Centre
This project aligns with the Centre’s aim to decode the function of peptides.
Report:
Decoding the 3D structure of proteins and protein-ligand complexes allows researchers to construct a visual map of the protein binding site. The 3D map informs chemists working on the development of novel inhibitors on a molecular scale. For example, knowing that the protein binding site has a star-shaped “lock” would influence the chemists to design star-shaped “keys” instead of circular or triangular-shaped molecules. The current gold standard for determining 3D structure relies heavily on single-crystal X-ray structures, which depend on the availability of single crystals and their ability to diffract X-rays. Unfortunately, in most cases X-ray crystallography is not an option. Therefore, a cross-nodal collaboration led by CI Gottfried Otting was established to produce target proteins incorporating molecular lanthanide tags. These modified proteins can be analyzed using newly developed magnetic resonance spectroscopy methodologies to identify the binding site of high-affinity peptide ligands on their target protein making it possible to decode the 3D structure of the target protein.
To unveil specific protein-protein and protein-ligand interactions, the team sought to develop a molecular tag that could be incorporated into the target protein during bacterial expression-based biosynthesis by genomic encoding, i.e., specifying the exact site of incorporation by the DNA sequence. To do this, CI Gottfried Otting and AI Thomas Huber at the Australian National University identified new aminoacyl-tRNA synthetase mutants that can incorporate unnatural amino acids with a chemically reactive handle during protein biosynthesis using bacterial expression systems.
Molecular tags including a lanthanide metal were chemically synthesized by research fellows Dr Leila Hill and AI Stephen Butler, led by CI Kate Joliffe at The University of Sydney. The magnetic field generated by the lanthanide is sensed by the protein surface nearby in nuclear magnetic resonance (NMR) experiments. The concept is akin to an aviation guiding beacon (lanthanide tag) on a tall building (protein) to make it more visible to pilots (biochemists) at night.
To accurately detect the positions of individual atoms relative to the lanthanide tags, tailored NMR methods were developed by CI Gottfried Otting and research fellow Dr Henry Orton. Unexpectedly, the theoretical and experimental data obtained revealed that, unlike a radar system that only measures distances, the directionality of magnetic fields greatly enhances the information content in the measurements and, hence, the accuracy with which the locations of individual atoms can be determined from measurements conducted with multiple tags.
Additionally, the team explored the possibility of assessing subtle changes in protein structure relative to the environment by measuring solvent accessibility of individual atoms by detecting the NMR impact of solutions made up with a paramagnetic lanthanide compound. This is important to biochemists because a protein is a dynamic three-dimensional unit that can adopt more than one form to carry out its biological function. To minimize electrostatic interactions between lanthanide compound and protein, an uncharged gadolinium complex was synthesized for measuring solvent-induced paramagnetic relaxation enhancements (sPRE) by NMR spectroscopy. Such measurements assess how exposed different parts of a protein are to their solvent environment.
A complementary technique employed electron paramagnetic resonance (EPR) spectroscopy in collaboration with AI Professor Nicholas Cox at the Australian National University, who set a world record in the range and accuracy of the distances that can be measured between a gadolinium label and fluorinated amino acids site-specifically installed in various proteins. As many FDA-approved drug molecules contain fluorine atoms, the proof-of-principle established will be of great value in detecting the binding mode of drugs on protein targets by EPR spectroscopy without relying on risky and time-consuming protein crystallography.
In a practical application of the protein tagging methods described above, the binding of site-specific peptides to the main protease of SARS-CoV-2 was explored in collaboration with CI Richard Payne at The University of Sydney.
During the pandemic, the SARS-CoV-2 main protease was discovered as an important target for drug development. This protease is made of two identical units fused together (homodimer), and its large molecular size makes it difficult for analysis using traditional NMR spectroscopy. To address this issue, a fluorine tag was installed into the SARS-CoV-2 main protease, enabling researchers to use NMR spectroscopy to confirm the binding site of a peptide inhibitor synthesized by CI Payne, which informs future drug development activities. In addition, NMR spectroscopy revealed that the site-specific binding of the peptide inhibitor induced a structural response throughout the protease, indicating an unusual degree of flexibility.
The key takeaway result of this project is the realization that the placement of multiple lanthanide tags at a single site of a protein delivers 3D structural information on proteins and protein-ligand complexes with great accuracy from straightforward NMR spectroscopic experiments. This finding greatly simplifies the use of paramagnetic tags for 3D structural determination by NMR spectroscopy.
Additionally, Professor Nick Cox successfully established a proof-of-principle for accurately measuring the distance between fluorine atoms and gadolinium ions using EPR experiments for distance ranges that are very difficult to access by any other technique.
To increase the robustness of these techniques, AI Stephen Butler at Loughborough University is now working on developing alternative lanthanide tags that are expected to be more synthetically accessible. CI Gottfried Otting and AI Thomas Huber also aims to create new unnatural amino acids bearing nitroxide radicals that will be compatible with aminoacyl-tRNA synthetases to incorporate during bacterial expression-based biosynthesis of peptides.
This project has shown that EPR spectroscopy can provide accurate distance measurements in proteins that are not suitable for established structural biology techniques such as X-ray crystallography, NMR spectroscopy or cryo-electron microscopy. This makes it a unique tool to prove (or disprove) models computed by AlphaFold and related software. EPR spectroscopy is also not limited by molecular weight, making it uniquely suitable for assessing structural models of protein-protein complexes, which are at the heart of biological signaling and disease.
As a spin-off from this project, CIPPS research fellow Dr Elwy Abdelkader also designed the viral protease expression vector (human rhinovirus 14 3C protease) used to process the SARS-CoV-2 main protease. This vector has been distributed to over 300 laboratories worldwide since its inclusion in the Addgene (USA) database.
CIPPS contributors: Professor Gottfried Otting (ANU), Professor Kate Joliffe (USyd), Professor Richard Payne (USyd), Professor Colin Jackson (ANU), Professor Nicholas Cox (ANU), Professor Thomas Huber (ANU), Dr Stephen Butler (Loughborough), Dr Leila Hill (USyd), Dr Elwy Abdelkader (ANU), Dr Henry Orton (ANU).