27 26 D I A M O N D L I G H T S O U R C E A N N U A L R E V I E W 2 0 2 1 / 2 2 D I A M O N D L I G H T S O U R C E A N N U A L R E V I E W 2 0 2 1 / 2 2 Macromolecular Crystallography Group Beamline I04-1 XChemfragment screening finds newbinding sites on vital tubulin protein Related publication: Mühlethaler, T., Gioia, D., Prota, A. E., Sharpe, M. E., Cavalli, A., & Steinmetz, M. O. Comprehensive analysis of binding sites in tubulin. Angewandte Chemie International Edition 60 , 13331–13342 (2021). DOI: 10.1002/anie.202100273 Publication keywords: Crystallographic fragment screening; Molecular dynamics simulations; Tubulin; Microtubules; Protein-ligand interactions T he tubulin protein plays an essential role in vital cell functions, including cell division. Tubulin molecules form tube-like structures, called microtubule filaments, that give cells their shape and help transport proteins and other cellular components. Tubulin can bind to many proteins and small molecules, but the total number of binding sites it has was unknown. Researchers at the Paul Scherrer Institute and the Italien Institute of Technology used a unique combination of computer simulations and crystallographic fragment screening performed at the XChem facility and Macromolecular Crystallography (MX) beamline (I04-1) to investigate that fundamental question. For the fragment screening, the teamexposed hundreds of tubulin crystals to solutions containing fragments of molecules. Then, they used beamline I04-1 to create X-ray diffraction patterns for each soaked crystal, showing which molecule fragments have bound to the tubulin, and where. Their results uncovered 11 previously unknown binding sites on the protein and identified 56 fragments that bind to tubulin and could be used in future drug development. The team’s approach can also be used to investigate other proteins and could help to discover new binding sites in other pharmaceutically important molecules. Microtubules are dynamic cytoskeletal filaments, which are assembled from and disassembled into their αβ-tubulin heterodimeric building blocks (referred to as tubulin from here onwards). A fundamental property of microtubules and tubulin is their ability to bind a plethora of regulators. The main activity of these regulators is to modulate microtubule dynamics and organisation, and consequently microtubule function. In cells, an array of proteins bind microtubules to control fundamental microtubule cytoskeleton- based physiological processes in all eukaryotes ranging from cell division, cell motility, cell polarity to intracellular trafficking. In addition, a large number of chemically diverse, small molecule ligands bind to seven so far identified, distinct binding sites in tubulin. Notably, compounds that interfere with microtubule function have been very successfully used to treat human pathologies including gout and cancer, however, they are also generally employed in basic research studies aimed at understanding microtubule cytoskeleton-based cellular processes (reviewed in 1 ). The observation that tubulin can interact with a plethora of regulators raises the intriguing question of how many different binding sites do actually exist in the tubulin dimer. Here, the researchers addressed this question using a combined computational and crystallographic fragment screening approach. They initially performed a molecular dynamics simulation in explicit solvent with a high-resolution X-ray crystal structure of the tubulin dimer 2 . They then computationally identified pockets in tubulin, analysed their relative dynamics and persistency, and assessed their communication networks by tracking the exchange of atoms between adjacent pockets. With the two-fold objective to (i) validate experimentally their computational predictions, and (ii) to identify potential ligands able to bind into novel tubulin pockets, they next conducted an X-ray crystallography-based fragment screen using the XChem facility and beamline I04-1 at Diamond Light Source 3 . A fragment is a small, ~200 Da chemical entity that in combination with a crystal structure of the fragment complexed to its target has been recognised as a powerful tool for structure- based drug design 4 . They soaked individual tubulin crystals with 708 different fragments, collected 672 X-ray diffraction data sets, and solved 503 structures with a resolution of better than 4.0 Å. The data obtained from their combined computational and experimental approach revealed a total of 27 distinct binding sites in tubulin. Remarkably, all major, previously characterised tubulin-drug binding sites (reviewed in 5 ) were readily detected. Furthermore, several key contact points between tubulin dimers in microtubules as well as between tubulin and secondary structural elements of regulatory protein partners were revealed. Importantly, they found 18 sites that are not targeted by any of the antitubulin drugs that have been structurally characterised to date. 11 out of those (seven in b -tubulin and four in a -tubulin) represent completely new and thus undescribed binding sites in the tubulin dimer. They further found an intricate, dynamic communication network between different pockets located also remote from each other in both the a - and b -tubulin monomers (Fig. 1). Finally, they identified 56 chemically diverse fragments that bound to a total of 10 different tubulin pockets (Fig. 2). Their results have important implications. For example, it is well known that the vast majority of structurally characterised tubulin-binding ligand and protein partners target b -tubulin. An open question in the antitubulin drug development field has thus been whether a -tubulin can at all be considered as a target for the development of small molecule modulators of microtubule dynamics. Their analysis now presents several sites in and fragments able to bind to a -tubulin, which can be exploited in future antitubulin ligand development campaigns. Their results further disclose several fragment binding sites in the tubulin dimer whose residue composition notably differ amongst human tubulin isotypes. This observation offers a unique basis for the design of isotype-selective antitubulin ligands. This is of particular interest in the context of chemotherapy since the upregulation of specific tubulin isotypes by cancer cells is a widely recognised resistancemechanismagainst antitubulin drugs. Finally, the majority of structurally characterised ligands and proteins that target tubulin typically do not share binding sites, which is rather surprising. Their experimental data now reveal four sites that are targeted by both fragments and secondary structural elements of major physiological microtubule regulators. In conclusion, their data and analyses provide a comprehensive description of the shape, chemical property, and dynamics of small molecule-binding sites in tubulin. Until now, drug discovery efforts were directed towards interfering with microtubule dynamics. Our results not only offer a platform for the innovative design of more selective antitubulin ligands with novel mechanisms of action with respect to microtubule dynamics modulation, they also provide a structural basis for the rational design of inhibitors of tubulin- protein interactions. In more general terms, our combined computational and crystallographic fragment screening approach used in this study offers a protocol that may help identifying new ligand-binding sites in any other pharmaceutically relevant target and analyse them in terms of chemical tractability and allosteric modulation. References: 1. Dumontet, C. et al. Microtubule-binding agents: a dynamic field of cancer therapeutics. Nature Reviews Drug Discovery 9 , 790–803 (2010). DOI: 10.1038/nrd3253 2. Prota, A. E. et al. Molecular mechanism of action of microtubule- stabilizing anticancer agents. Science 339 , 587–590 (2013). DOI: 10.1126/science.1230582 3. Douangamath, A. et al . Achieving efficient fragment screening at XChem facility at Diamond Light Source. Journal of Visualised Experiments 171 (2021). DOI: 10.3791/62414 4. O’Reilly, M. et al. Crystallographic screening using ultra-low-molecular- weight ligands to guide drug design. Drug Discovery Today 24 , 1081–1086 (2019). DOI: 10.1016/j.drudis.2019.03.009 5. Steinmetz, M. O. et al. Microtubule-targeting agents: strategies to hijack the cytoskeleton. Trends in Cell Biology 28 , 776–792 (2018). DOI: 10.1016/j.tcb.2018.05.001 Funding acknowledgement: iNEXT (PID2692), funded by the Horizon 2020 program of the European Union, the Regione Lombardia (ID 239047 NEON), and the Swiss National Science Foundation (31003A_166608 and 310030_192566). Corresponding authors: Prof. Michel O. Steinmetz, Paul Scherrer Institut,
[email protected] Prof. Andrea Cavalli, Italian Institute of Technology,
[email protected] Figure 2: Fragment-binding sites in tubulin determined by X-ray crystallography. Shown are the two a b -tubulin heterodimers as observed in their previously developed crystal system 2 . The α- and β-tubulin monomers are shown in dark and light grey ribbon representation, respectively. Fragment binding sites are highlighted in different colours. Figure 1: Tubulin pockets and their communication networks predicted by MD simulation. Predicted pockets and their communication networks in β-tubulin (top panel) and α-tubulin (bottom panel). Marine blue lines depict connected network nodes; their widths are displayed proportional to the respective communication frequency between two nodes. Spheres represent centre of masses of the pockets (corresponding to network nodes) and are shown in different colours. Identical colours indicate pockets that are often merged during the simulation. Spheres coated with yellow rings highlight novel sites identified in this study.