Developing ex vivo structural biology using natural abundance NMR: the role of conformational dynamics in regulating protein metastability

The efficient folding of proteins into their correct three-dimensional structures is essential for cellular function. In most cases this corresponds to the energetically most favourable state, but a number of metastable proteins fold instead to high energy conformations, which are primed to undergo large scale structural transformations when later required according to the particular function of the protein. Serpins are one such class of metastable proteins, of which the plasma glycoprotein alpha1-antitrypsin (AAT) is the prototypical example. Serpins comprise the most abundant family of protease inhibitors, and possess a molecular structure that is inherently dynamic: in the process of inhibiting their substrate protease, they undergo a dramatic change in shape from their initial metastable conformation. Clearly then, metastability is central to serpin function and conformational changes must be able to be triggered efficiently when required, and yet it is also their Achilles heel: spontaneous transitions can lead to misfolding or formation of polymeric aggregates, a process that is often associated with disease. Of the 35 serpin genes found in humans, nearly a third have a known involvement in hereditary disease, and five are known to form protein aggregates called polymers. However, despite many years of research the molecular mechanisms by which these transformations can be regulated remain poorly understood. It is our hypothesis that small-scale fluctuations ('dynamics') in the structure of the metastable native state may hold the key to this puzzle, and so this project is designed first to characterise the solution-state structure and dynamics of AAT molecules, and then to correlate these observations with the measured rates of conformational change.

Nuclear magnetic resonance (NMR) spectroscopy is an exceptionally powerful experimental technique for studying the structure and dynamics of proteins. However, NMR traditionally requires proteins to be expressed recombinantly within bacterial cells using specialised isotopic labelling techniques, and for a number of interesting molecules, including several variants of AAT, this is not currently possible. Instead, our preliminary data overturn this paradigm by showing we can measure high quality NMR spectra using AAT purified directly from human donors - including patients with rare, disease-associated mutations - without the need for isotopic labelling. Thus, for the first time we can study the solution-state structure and dynamics of ex vivo, natively glycosylated AAT molecules, and this has revealed widespread changes in the conformation of a disease-associated variant that were not observed using crystallographic approaches that confine molecules into a rigid lattice structure.

We propose to pursue these observations further, developing a new toolkit of NMR experiments to characterise structure and dynamics in these unlabelled ex vivo protein samples. We will investigate in detail the impact that mutations - associated with disease, or artificially designed - can have upon the structure and dynamics of the metastable serpin fold, and compare this with the effect the mutations have on both inhibitory activity and the misfolding and polymerisation processes.

In correlating the solution structure and dynamics of AAT variants with serpin function and dysfunction, our research will address the longstanding problem of how structural changes within metastable proteins can be regulated, and this may ultimately lead to a new mechanistic basis for the design of inhibitors of serpin misfolding. More broadly, the new NMR methodologies that we will develop in this project will provide a platform that can be readily extended to ex vivo structural biology of other previously inaccessible protein systems.

The conformational dynamics of metastable proteins (proteins that are kinetically but not thermodynamically stable) pose a fascinating problem of regulation, for while efficient transitions to the thermodynamic ground state are intrinsic to their function, spontaneous transitions must be effectively suppressed. Serpins are one such class, of which the plasma glycoprotein alpha1-antitrypsin (AAT) is the prototypical example. Mutations in AAT, identified from their association with pathology and the formation of inactive monomers or polymers, provide a valuable basis for understanding the interplay between structure, dynamics and function in this protein family.

NMR is a natural tool for studying protein dynamics, but many AAT variants cannot be expressed recombinantly with isotopic labelling. However, we show here that high quality spectra can be acquired using NMR measurements of ex vivo, patient-derived and natively glycosylated AAT at natural isotopic abundance. This reveals widespread changes in the solution-state conformation of a disease-associated variant not observed using crystallographic approaches.

We propose to pursue these observations further, to study the impact of mutations on the conformation and dynamics of AAT to regulate its function and dysfunction. We will define microscopic mechanisms of AAT conformational change, and test our hypothesis that access to different conformations is directly correlated with intermediate state populations determined through dynamical studies.

In correlating the solution structure and dynamics of AAT variants with serpin function and dysfunction, we will address the longstanding problem of how energy landscapes of metastable proteins can be sculpted and regulated, which may lead to new approaches for inhibiting serpin misfolding. More broadly, the NMR approaches we develop will provide a platform that can be readily extended to ex vivo structural biology of other previously inaccessible protein systems.

Our proposed research fits closely with the goal of the BBSRC to support fundamental discoveries in basic bioscience, and in particularly structural biology and technology development, but in addition to a wide range of direct academic beneficiaries our proposal has a number of potential economic and societal impacts:

1. Improving health and well-being.
The misfolding and polymerisation of the Z variant of AAT leads to its depletion from the blood and accumulation of polymers in the liver, resulting in liver cirrhosis and COPD respectively, and currently there are no specific therapies to block protein misfolding and polymerisation. The insights into AAT conformational dynamics and function that we will obtain may lead to the development of new generations of allosteric modulations and inhibitors of serpin function or misfolding, and the Lomas group is well positioned to support this development through an MRC programme grant and close links with industry and spin-out companies.
In addition, the new NMR methodologies we develop will be applicable to a range of other interesting systems, and may set the stage for future drug discovery strategies based upon ex vivo NMR spectroscopy. As a new field within NMR the potential for impact at this early stage is particularly high and to this end we will endeavour to disseminate and publicise our results and methods as widely as possible.

2. Enhancing research capacity, knowledge and skills within industry and the public sector.
In addition to potential applications towards drug discovery, the high-resolution natural abundance NMR methods we develop in this project may be of utility in areas of biotechnology, and particularly, to the characterisation and quality control of antibodies and biosimilars. We will attend industry conferences to disseminate our results, and identify and engage with relevant companies and organisations to whom such strategies will appeal (e.g. the National Institute of Biological Standards and Control, the National Institute of Standards and Technology (USA), and Prof Paul Dalby (UCL) and UCB, with whom interactions have been established via a previous BBSRC-BRIC grant).

3. Commercialisation and exploitation of scientific knowledge.
DAL and JI are involved in a spin-out company to pursue some promising ligand inhibitors of AAT polymerisation. The current proposal will facilitate a significant role for ex vivo NMR of AAT in such pursuits. Potential wealth creation to the UK could arise through successful commercialisation and the opportunity exists to take a similar approach to other diseases that result from protein misfolding and aggregation. If successful approaches could be adopted for related serpinopathies, particularly the dementia FENIB that results from polymerisation of mutants of neuroserpin.

4. Providing a scientifically well-trained professional workforce.
UCL has a very strong emphasis on research-based teaching - lectures and workshops on the results of the proposed study will potentially enthuse the researchers of the future. Additionally, through the exploration of challenging molecular systems that necessitate cutting edge NMR development at its core, this proposal addresses a shortage of highly skilled NMR spectroscopists in the workforce, scientists who can, for example leverage the recent major RCUK investment (>£20M) in high field NMR.

5. Increasing public engagement with research.
Public interest in science is inherently culturally enriching, and we will participate in public UCL lunchtime seminars and school outreach events (e.g., Sutton Residential Summer School). We will also engage directly with patients at the London AAT Deficiency Service and with the UK charity Alpha-1 Awareness (with which links are already closely established within the DAL group and, through previous fundraising activities, the researcher co-I Dr Chris Waudby) to raise awareness of our research and discoveries to this key stakeholder group.