My research interests are in using information about protein structures to design small molecule compounds that change the way proteins work. I divide my time between academic research within YSBL at York and applied research at the pharmaceutical company, Vernalis.
At Vernalis, we use the methods of structure-based drug discovery to discover new compounds which can be taken forward in clinical trials to treat various diseases and conditions, including cancer, inflammation, infection and neurodegeneration (see Vernalis Research Website for details).
At York we develop and apply these methods for both chemical biology and industrial biotechnology.
Most of the research in my York group is in fragment-based discovery – that is finding small chemical molecules that bind to a protein. These chemical fragments can give us some ideas about mechanism and function and also provide ideas about how to improve the compounds to change the way the protein works.
There are currently three principle areas of research:
Funding for these projects comes from grants from the BBSRC, industrial consortium and an EU ITN (fragnet) with pump priming and infrastructure support from institutional awards from EPSRC Wellcome Trust (C2D2).
In addition to these funded projects, there are continuing background interests in the analysis of how small ligands bind to proteins, and the development of molecular modelling methods such as the program rDock
A recent discovery is of compounds that actually increase the activity of an enzyme. We have found and optimised small molecules that increase the catalytic activity of a glycoside hydrolase (pictured) and have developed methods to permanently attach the small molecule to the protein.
This approach may be a way to improve the efficiency of action of enzymes which perform industrial processes and we are currently working on a number of different enzymes to see if we can affect their activity in a similar way.
We are collaborating with Prof Peter McGlynn in exploring new ways to interfere with how bacteria replicate their DNA. The McGlynn group can reconstitute this molecular machine, called the DNA replisome, in the test tube. We have identified small molecules (fragments) which disrupt this molecular machine and are currently exploring which proteins or sets of proteins in the replisome these fragments are interfering with. This could identify potential strategies for killing bacteria.
Many of the small molecules used in medicinal chemistry are flat rings – and there is general interest in making more three dimensional molecules. However, these have traditionally been difficult to make, requiring chemistry that can control stereochemistry. We are helping Prof Peter O’Brien to design and assess compounds made using some of his chemistry, in a consortium with various industrial partners. Details of the synthetic projects can be seen on his website.
Key recent publications
- Increase of enzyme activity through specific covalent modification with fragments, Darby et al (2017) Chem Sci DOI: 10.1039/C7SC01966A
- Erlanson, D., Fesik, S., Hubbard, R.E., Jahnke, W., Jothi, H. (2016) “Twenty years on: the impact of fragments on drug discovery”
Nature Reviews Drug Discovery, 15, 605-619, doi:10.1038/nrd.2016.109
- Discovery of Selective Small-Molecule Activators of a Bacterial Glycoside Hydrolase
Darby et al. (2014) Angew Chemie Int Ed, 53, 13419-13423
- rDock: A fast, versatile and open source program for docking ligands to proteins and nucleic acids
Ruiz-Carmona et al. (2014) PLoS Comp Biol 10 Article Number: e1003571
- Hsp90 inhibitors and drugs from fragment and virtual screening
Roughley et al. (2012) Top Current Chem, 317, 61-82
- Experiences in fragment-based lead discovery
Murray, J.M. and Hubbard, R.E. (2011) Methods in Enzymology, 493, 509-532
- Design of a fragment library that maximally represents available chemical space
Schulz et al. (2011) JCAMD, 25, 611-620
Apart from brief sabbaticals at Harvard, Rod Hubbard’s academic career has been at the University of York. During the 1980s he was a pioneer in the development of molecular graphics and modeling systems for studying protein structure (HYDRA and QUANTA), which introduced methods that are still in use today. In the 1990s, he helped to build (and directed) the Structural Biology Laboratory at York as a major centre, with over 80 scientists studying the structure and function of proteins. For the past fifteen years, his personal research interests have focused on understanding the relationship between structure, mechanism and function in various protein systems (including proteases, nuclear receptors and kinases) and experimental and theoretical studies of protein-ligand interactions. Since 2001, he has spent some of his time at the company Vernalis, where he helped establish and apply structure-based drug discovery methods. He is a consultant to a number of pharmaceutical and technology companies, sits on a number of Research Council committees and boards and is chair of various external advisory groups for large scale academic projects.
Over the past thirty years, the York group has determined the structures of a large number of proteins. These structures have increased our understanding of the mechanism of action and biological function of the proteins underpinning many areas of biological science. These insights have then been translated into new ideas for disease treatment (drug design) and for use in various industrial processes. Examples of important therapeutic projects include early protein engineering on modified insulins and humanised antibodies as well as important drug targets such as kinases, proteases and the estrogen receptor (pictured).
Examples in industrial biotechnology include the first structures of various enzymes such as lipases, cellulases and amylases, used worldwide in washing powders, paper treatment and recently being explored in biofuel production.
As well as exploring these different systems, the York group has a long track record of developing the methods which are used to carry out these studies. In the Hubbard group, this has led to the development of a range of structure-based methods that identify and help design new compounds that can bind to the proteins and change the way the proteins behave. The methods used include protein crystallography, high field NMR spectroscopy, surface plasmon resonance (SPR), fragment-based screening, computational docking, virtual screening and molecular modelling and design. The development and application of these methods is the major focus of our current research.
It is extremely difficult to design or discover a small molecule compound that binds strongly and selectively to a particular protein target. Just small changes in a compound can have big effects on binding making it very unlikely you can find effective molecules by random synthesis. Fragment-based methods work by identifying very small fragments that bind weakly to different parts of an active site. Determination of the crystal structures of these fragments bound to the protein can then be used to design composite molecules that will bind strongly. We have established an 1100 member fragment library and use NMR and other biophysical methods to identify which fragments bind. The crystal structures of the fragments bound to the protein (as in the figure) are then used to discover or design larger compounds which bind to the protein with greater affinity.
In virtual screening, relatively simple computational chemistry calculations are used to assess each compound from a large database of accessible molecules for its ability to bind to the active site. This rather crude screening identifies a subset of the compounds for further analysis or filtering. York is a partner in the development of the rDock program. Projects in the laboratory have included investigating and optimising new features in the software to deal with solvent presence and position, investigation of new features to dock against a family of protein structures.
Schematic showing the virtual screening process. Molecular docking calculations select which compounds from a large library can fit into the binding site of the protein to produce an initial list of virtual hits. Detailed molecular calculations are then used to decide which are the best hit compounds which are then selected for assay.