The Quantitative Imaging and Nanobiophysics Group
Our laboratory was established in 2013 to undertake research combining cell biology, optical physics and biochemistry. The group focuses on biological problems that cannot be addressed with current imaging technology, and thus aims to develop analytical, optical and biochemical approaches to address these questions. In cell biology we aim to understand how viruses enter cells, probing and remodelling membranes, and what are the structural changes viruses undergo during cell-entry, uncoating and morphogenesis. To do so, we are developing new classes of fluorescent probes, high-speed cell friendly Super-Resolution (SR) methods and computational modelling approaches that, although designed to answer questions of interest in the lab, will have broad applications in cell biology research.
The overarching goal of our lab is to develop technology broadly applicable to research in cell biology. However, these developments are generally primed by the absence of technology to address flagship biological questions, directly studied by our research group. We particularly focus on studying cell signalling and host-pathogen interactions. In particular, we use Super-Resolution (SR) to study the dynamic host-membrane organization of viral receptors, receptor mediate cell entry of HIV-1 and other viruses (collaboration with Mark Marsh). In tandem, we use SR and modelling approaches to study Vaccinia viral structure and host-factors involved in key viral replication checkpoints, such as entry, uncoating, replication, and assembly (collaboration with Jason Mercer).
To attain <30nm super-resolution fluorescence imaging on the dynamic nanoarchitecture of a biological system it will be necessary to develop small probes with high-specificity, minimal disruption of the target molecule behaviour and photophysics compatible with SR microscopy. The resolution attainable with SR imaging is influenced by a number of factors including the size of the labelling moieties used to target fluorescent dye molecules to the protein of interest: The smaller the labelling structure used, the closer the fluorophore’s location correlates to that of the target molecule, and the greater the precision that can be attached to its localisation. In addition, small labelling molecules such as single-domain antibodies (nanobodies) can more easily access target domains within densely packed environments with minimal stearic hindrance when compared to large antibody complexes. Our FlAsH-PALM approach (Lelek et al. 2012) has demonstrated that small probes are critical to explore the compact structure of viruses without hindering their structure and activity. We have discovered and are developing a novel type of modular SR probe of simple synthesis, allowing us to easily ‘plug-in’ most available synthetic fluorescent dyes into a linker scaffold that induces photoswitching. In collaboration with Mark Marsh, and Darren Tomlinson (U. Leeds) and Avacta Life Sciences, we are designing small binders (single-domain antibodies and Adhirons) to target both components of HIV-1 and viral target host receptors, with the aim of super-resolving the process of HIV entry into cells.
Optical Physics and Applied Mathematics
Over the last years we have developed optical and mathematical approaches to advance the SR field. One such example is the QuickPALM project, the first open-source real-time analysis algorithm for 3D PALM/STORM SR (Henriques et al. 2010).
Nevertheless, despite significant progress in the field, live-cell SR studies remain hampered by the requirement for intense illumination that is usually phototoxic. We are currently developing a new approach to tackle this problem - SR Radial Fluctuations (SRRF) - enabling SR in modern microscopes (TIRF, widefield and confocal) using conventional fluorophores such as GFP. SRRF is capable of live-cell imaging over timescales ranging from minutes to hours, using sample illumination orders of magnitude lower than methods such as PALM, STORM or STED.
Concurrently, our group is prototyping a novel SR apparatus. This instrument is the basis for the development of several new SR approaches featuring characteristics beyond the state-of-the-art and not available commercially. The system, LMCB-Alpha, is an experimental instrument optimized for high-speed non-phototoxic live-cell (Almada et al. 2015) and high-content (Pereira et al. 2015) SR microscopy. By featuring a modular design, the system allows easy and cost-effective upgrades. It has been engineered to seamlessly integrate the control of microfluidics, allowing for ‘lab-on-a-chip’ SR experiments. This feature enables approaches such as live-cell SR imaging followed by rapid fixation, allowing the same sample to be carried over for correlative SR and electron microscopy. Overall, this system has an acquisition 1000-fold faster and requires 100-fold less illumination to achieve data of comparable resolution to the one acquired in a commercial equivalent (e.g. Zeiss Elyra PS-1).
Image Data-Mining and Computational Modelling
Although the field of SR microscopy is starting to mature, there is still a lack of analytical tools capable of extracting statistically robust quantitative information from the generated data. We have developed high-content approaches to analyse large populations of cells through SR microscopy (Pereira et al. 2015), and have used this approach to map the topology of T-cell signalling complexes at the surface of the plasma membrane (Soares et at. 2013). We have a long-standing interest in studying dynamic changes in viral structure during the infectious process, a challenge well suited for SR microscopy as conventional light microscopy cannot accurately resolve the subviral architecture of mammalian viruses and EM is incapable of live-cell microscopy. While at the group of Christophe Zimmer (Inst. Pasteur, Paris), prior to the formation of the current LMCB research group, we were the first to combine the information of hundreds of super-resolved HIV-1 viruses in live cells to structurally map capsid uncoating (Lelek et al. 2012). Recently, in collaboration with Jason Mercer, we have started the VirusMapper project, a computational framework capable of mining thousands of viral images both from SR and EM, classifying and generating structural models of the different structural states and components of viruses.