Towards understanding the glycan code: next generation structural glycobiology for accurate description of protein-glycan complexes

The extraordinary properties of living systems emerge as a result of dynamic interactions between their biomolecular components (proteins, nucleic acid, carbohydrates or glycans, ...). Despite the complexity of a system of millions of molecular interactions, the whole network of interactions is highly regulated for healthy organisms, and alterations in the regulation processes underlie all diseases. In this network, different molecules interact with different strengths (affinities), where strong interacting partners produce stable biomolecular complexes and weak binding molecules produce transient assemblies. A well "orchestrated" system of protein-ligand interactions of very different affinities is indeed responsible for most of the regulation processes in living organisms. Strong complexes are formed when sustained biological signals are needed, whereas weak interactions are recalled when quick cellular responses are required after temporary stimuli (signal transduction, reversible cell-cell contacts, transient interactions in host/pathogen recognition, etc.). For a complete understanding of life processes, it is necessary to investigate both strong and weak protein-ligand interactions, which encourages the development of novel approaches to characterize the 3D molecular structures of weak protein complexes.

Strong protein-ligand interactions have been extensively investigated and many biologically relevant 3D complexes have been determined, as their intrinsic stability makes them amenable to a number of analytical techniques. However, for weak protein interactions many conventional approaches fail or become unreliable. For example, X-ray crystallography, a very powerful structural technique, show limitations for weak interactions as: (i) obtaining crystals of the complexes including the ligand is difficult, and (ii) the typically poorly defined electronic density that describe the ligand in the binding pocket. NMR spectroscopy, one of the most powerful techniques to study intermolecular interactions, has demonstrated its extraordinary capability for the detection of weak protein-ligand interactions in solution, through the use of ligand-based experiments, like Saturation Transfer Difference (STD) NMR spectroscopy. However, the translation from these experiments to 3D structures is currently not straightforward.

We have published some improvements in the set up of STD NMR experiments to determine protein-ligand affinities and study multiple binding modes of ligands in a protein binding pocket. The present proposal stems from recent results in our research group that allow us to propose that STD NMR spectroscopy can provide more structural information for weak interactions than the map of ligand contacts, or group epitope mapping (GEM). Here, we propose to obtain the orientation of the ligand in the binding pocket, a much needed piece of information. We plan to generate new experimental restraints to drive the structural calculations of the complexes, to get accurate 3D structures. Until now, these experimental restraints have remained unexplored.

Among the most biologically relevant weak interactions, those of proteins with glycans are essential steps in many cell-cell communication processes, and key in the infectivity of microbial pathogens. In particular, influenza virus exploits this for initial cell recognition, attachment, and release of new virions. In a different strategy, HIV covers its surface with host glycans to evade the immunological response. Interestingly, new broad neutralizing antibodies (bNAb) are being discovered which are able to stop infection and are elicited against those carbohydrates in that glycan shield. In the context of our collaborations with Prof. Rob Field (Norwich) and Dr. Katie Doores (London), we will apply the novel STD NMR approaches to investigate the molecular recognition processes in human and avian influenza virus, and in immunologically active bNAbs against HIV.

Weak specific protein-glycan interactions are biologically relevant as they constitute the translational elements of the so-called "glycan code" of high importance in the context of glycomics. However, their low affinity poses significant difficulties for the determination of 3D structures: X-ray crystallography can produce incorrect glycan structures, and NMR suffers from a lack of enough experimental restraints.

Many technical issues make difficult the structural calculations of 3D models of weak protein-ligand complexes:
(i) Intermolecular NOEs are extremely difficult to observe (the large excess of ligand "dilutes" the intensities of these NOEs among the signals of the large fraction of free ligand). Thus, the orientation of the ligand is uncertain, even if the structure of the protein (x-ray or NMR) and the ligand (trNOESY) are known.
(ii) Even if very favourable conditions allow determination of a few intermolecular NOEs, full NOE calculation of the structure would involve the full resonance assignment of the protein(iii) For very large receptor size, extremely fast transverse relaxation makes NMR detection not possible, precluding any complete NOE study

We propose to develop a novel "pharmacophoric STD NMR" approach for 3D structure calculation of weak protein-ligand complexes. We will use gradual modification of saturation power, frequencies, saturation times, and solvent composition, to get "functional" protein-ligand information to restrict conformational search of the ligand in the binding pocket. Finally, we propose to develop a protocol to obtain long sought protein-ligand distances restraints from STD NMR, based on a novel protocol for distance calibration. Calculations of 3D structures using the intermolecular semiambiguous restraints will follow using the very recently published "NMR2 molecular replacement" methodology.

Finally, the novel NMR approach will be used to study glycans interactions in flu and HIV infection.

Pathways to Impact

Overall impacts and beneficiaries:
Developments in the ability to determine accurate 3D molecular structures of weak protein-ligand complexes impact on all current research endeavours focused on deepening our understanding of transient or low affinity biomolecular interactions. Specifically, biologists interested in protein-glycan and protein-metabolite interactions will be significantly benefited from the outcomes from this project. The technical developments will be sought closely by the NMR spectroscopic community, including main companies manufacturing equipment, and the findings in the virus studies will impact pharma companies involved in vaccine design, as well as companies developing selective sensors for flu virus detection. Finally, the PDRA will gain an attractive skill set for future employment in academy or industry.

Impacting a BBSRC strategic priority area:
The development of novel NMR approaches to determine accurate 3D structures of weak biomolecular complexes constitutes a solution to a specific technological gap identified in world-class bioscience. The proposed project is focused on underpinning the "bioscience for health" research priority area. Along the course of the project, the PI will highlight the importance of specific low affinity protein-ligand interactions in key regulatory processes. After successful outcomes from the 3 objectives, the PI will communicate the findings with maximum dissemination to the public with the help of the UEA media relations office.

Novel scientific findings:
A critical deliverable of the project focus on transferring the new knowledge to scientists. This will be carried out through the usual channels of conference presentations and publications in high impact scientific journals in both chemistry and biology. The PI and PDRA will engage with UEA media relations office to announce the work to media through the corresponding press release, and on websites and social media like UEA Facebook page, or Twitter. In 2017 the PI will establish a dedicated webpage to document our studies on weak protein-ligand complexes for the general public. This will help to enhance public knowledge of the biological significance of carbohydrates as regulators of cell-cell and cell-pathogen interactions, and its link to health and disease.
The PI and PDRA will be collaborating with national and international experts and the development and strengthening of such collaboration will increase the global impact of our findings.


Impact on society:
There is a continuous demand from society to the research community for investigating alternative ways to eradicate infectious agents, or to neutralise their pathological effects on humans. Our findings will inform the design of novel compounds for improved molecular sensors for selective detection of human or avian influenza virus, and the design of potential novel immunogenic molecular structures that could lead to the development of broadly acting HIV vaccines.

Impact on industry:
The goal of the project is to provide novel NMR spectroscopic approaches for structure characterization of weak complexes, as well as to unveil structural details of the molecular recognition of glycans in very relevant infection processes, both goals of interest for companies. The PI has identified some of the relevant stakeholders that will be impacted by the project, and they have expressed their interest.

Training and development:
This project will employ a PDRA and contribute to their professional training and development in a variety of technical, academic, and engagement skills. The hired PDRA will be fully engaged in the impact agenda, and will spend a short time in an industrial setting which will provide the PDRA with a very attractive and unique skill set for future employment in academy and industry.

Full details are provided in the "Pathways to Impact" section of the full proposal.