Imaging cellular structure and function beyond the diffraction limit
Funded by the MRC Next Generation Optical Microscopy Initiative (MR/K015834/1)
Professor Matthias Merkenschlager (PI), Professor Amanda Fisher, Dr Simona Parrinello, Professor David Carling, Dr Vincenzo De Paola, Dr Enrique Martinez-Perez - MRC Clinical Science Centre
Professor Paul French, Dr Christopher Dunsby, Professor Mark Neil, Professor Roy Taylor - Photonics Group, Physics Department
Professor Tony Magee - Facility for Light Microscopy (FILM), NHLI
Super-resolved microscopy (SRM) represents a revolution in optical imaging with profound implications for the life sciences. It allows researchers to visualise biological structure and function on almost the molecular scale in live cells and organisms. Typically, biological processes are visualised by attaching fluorescent molecules to proteins of interest forming images from the light emitted by these fluorescent "labels". The optical resolution, which defines the smallest structures that can be visualised, has long been regarded as fundamentally limited to >200 nm by the diffraction of light waves.
Recent breakthroughs in fluorescence microscopy, however, have exploited the ability to systematically switch the emission of fluorescent molecules on and off in order to break the "diffraction limit" and achieve resolutions down to 10's of nm - almost the scale of the typical biological proteins that are involved in signalling processes that determine the fate of cells and organisms and so are essential to disease processes.
This project builds on the partnership between the MRC Clinical Sciences Centre (CSC) and Imperial College London to develop a range of SRM instruments to advance our understanding of epigenetics, regulation of genes and cells, immunology and metabolism by resolving fundamental biological mechanisms inside the nucleus, inside live cells and at the synapses of interacting cells. It also aims to establish super-resolved microscopy as a widely accessible tool for the research community at the CSC and Imperial.
The first SRM technique would be structured illumination microscopy (SIM) that entails illuminating samples with patterns of light and observing the interference patterns between these patterns and the structure of the sample. This effectively doubles the resolution of a normal microscope. The second approach is an advance on laser scanning confocal microscopy, in which as sample is imaged by scanning a focused laser beam across a sample and collecting the fluorescence pixel by pixel through a pinhole that blocks out of focus or scattered light. The size of this focussed laser beam spot defines the resolution of this microscope. By using a second, collinear, specially shaped laser beam that is scanned synchronously and can switch off the fluorescence from molecules around the outside of the first beam focus, the fluorescence excitation is restricted to a smaller spot than the diffraction limit and thus resolutions below 50 nm can be realised. Because this technique can block scattered or out of focus light, it is particularly suitable for imaging in live organisms. The third SRM modality relies on having only a small fraction of the fluorescent labels being switched on at any time such that they can be treated like arrays of single molecules. Their individual positions can be determined by locating the centres of the "blobs" that represent each molecule in an image - which is possible to a precision of a few nm. By sequentially switching on different subsets of the fluorescent labels and determining their individual positions, one can build up a super-resolved "image" that is simply the map of the locations of all the fluorescent molecules.
In this proposal, we will develop and apply these new imaging techniques to visualise biological structures and processes that are simply too small or inaccessible for conventional microscopes. Starting at the surface of cells, researchers will be able to observe how immune cells communicate and nerve cells interact with each other, and how tumour cells relate to normal cells. Looking inside cells, we will visualise signals that communicate the energy status of cells and how the genome is organised. It will be possible to resolve how chromosomes behave in developing germ cells, which is relevant to the most frequent causes of infertility and birth defects. The research will deliver new insights into both normal biological processes and disease states
Progress to date
SIM FLIM microscope platform
We have developed a unique instrument for super-resovled studies of biology that combines SIM (on a Zeiss ELYRA) with optically section FLIM implemented using a Nipkow spinning disc scanner and a time-gated image intensifier to provide fluorescence lifetime imaging (FLIM). FLIM may be used to read out protein interactions and biosensors reporting cell signalling events using Förster resonant energy transfer (FRET) and also to read out changes in cell metabolism via autofluorescence from NAD(P)H and/or flavoproteins. This instrument will therefore enable cell signalling processes or metabolic changes to be correlated with morphological information below the resolution of standard confocal or wide-field microscopes. The FLIM modality is controlled by µManager open source software and the FLIM data analysis is undertaken using FLIMfit, our open source client for OMERO.
We are interested to explore the potential applications of this novel instrument, particularly to correlate the response of biosensors with morphological changes, e.g. during the cell cycle. If you would like to work with us and explore the potential of this instrument, please email Paul French.
Low cost PALM/STORM microscope platform: “easySTORM”
We have developed a new approach to localisation microscopy (PALM/STORM) and to TIRF that is more robust and significantly lower cost than standard instruments. Essentially we have demonstrated that we can realise TIRF and STORM using multimode fibres, which greatly relaxes the optical alignment tolerances and the efficiency with which the excitation radiation is coupled to the sample. This enables TIRF and STORM to be realised with much cheaper excitation lasers, including multimode broad-stripe diode lasers. Our simplest “easySTORM” configuration can utilise a £30 laser diode and provide STORM images with sub-50 nm resolution. To make this technique more widely accessible, we have also implemented simpler sample preparation protocols based on Vectashield rather than more complex buffers. We have also written the control and data acquisition software in the open source µManager software and used open source localisation software (mainly Thunderstorm) for analysis. We have also developed software tools to enable the localisation and the visualisation steps in the data processing to be separated. We are also extending the functionality of this instrument to 3-D STORM.
This easySTORM platform has been extended to 3-D STORM using an axially varying PSF and we are currently working on approaches to correct for spherical aberrations in order to image deeper into biological samples and on ways to accelerate the data processing for localisation microscopy.
New STED microscope technology
We have previously developed our first STED microscope, which introduced the Ti:Sapphire/supercontinuum configuration for depletion/excitation beams as well as the incorporation of time-gated detection, including for FLIM, and a spatial light modulator for aberration correction in both microscope and sample. We subsequently extended this instrument for 3-D depletion and a priori correction of aberrations. In this project we have upgraded this instrument by replacing the stage scanner with a rapid galvanometric scanner to provide faster imaging and eliminate sample motion.
We have also worked to develop a new STED microscope for application to GFP-labelled samples, for which we have demonstrated a novel ultrafast fibre laser to be used as the STED depletion laser. This offers advantages in terms of cost compared to current approaches based on Ti:Sapphire lasers and optical parametric amplifiers and provides pulsed depletion, which is an advantage compared to existing c.w. fibre lasers that are often used for STED microscopy.
In parallel work the MRC Clinical Sciences Centre has purchased a commercial Leica TCS SP8 STED 3X microscope that provides a useful benchmark for our home-built SRM systems.
Applications of super resolved microscopy
We are using super resolution microscopy to investigate how chromosomes become structurally organized around proteinaceous axial elements at the onset of meiosis, and how these structures are remodelled at later meiotic stages. We are particularly interested in structural changes that occur in the context of fully paired homologues, when homologous axial elements are separated by a distance of about 100 nm, and therefore cannot be resolved by conventional microscopy. The laboratory of Enrique Martinez-Perez has successfully developed methods to visualize C. elegans meiotic chromosomes using the Zeiss ELYRA SIM and the Leica TCS SP8 STED 3X systems at the CSC. Using different approaches to label meiotic proteins (tagging with fluorochromes by CRISPR or single-copy transgenes, and protein visualization with primary antibodies), we have been able to resolve individual axial elements from three-dimensionally intact germ lines with the SIM and STED systems. For example, we have used the SIM system to investigate how WAPL, a protein that regulates the association of cohesin with chromosomes, affects the removal of different cohesin complexes during meiosis.
We are also working to apply SRM to the immunological synapse (IS). Initially applying STED microscopy to image the IS between two interacting cells and progressing to apply two-colour STORM to image actin and tubulin at the IS and also to implement SRM distinguishing the actin of two interacting cells.
Experimental proof of concept of nanoparticle-assisted STED, Y. Sonnefraud, H. G. Sinclair, Y. Sivan , M. R. Foreman , C. W. Dunsby, M. A. A. Neil, P. M. French, and S. A. Maier,Nano Lett., 14 (2014) 4449–4453, DOI: 10.1021/nl5014103
Duration-tunable picosecond source at 560 nm with watt-level average power, T. H. Runcorn, R. T. Murray, E. J. R. Kelleher, S. V. Popov, and J. R. Taylor, Opt. Lett. 40 (2015) 3085-3088; doi: 10.1364/OL.40.003085
Fiber-integrated frequency-doubling of a picosecond Raman laser to 560 nm, T H Runcorn, R T Murray, E J R Kelleher, S V Popov, J R Taylor; Opt. Express 23 (2015) 15728-15733; doi: 10.1364/OE.23.015728
Cohesin-interacting protein WAPL-1 regulates meiotic chromosome structure and cohesion by antagonizing specific cohesin complexes, Crawley O, Barroso C, Testori S, Ferrandiz N, Silva N, Castellano-Pozo M, Jaso-Tamame A,
Martinez-Perez E, eLife 5 (2015) e10851, DOI: 10.7554/eLife.10851
easySTORM: a robust, lower-cost approach to localisation and TIRF microscopy, K. Kwakwa, A. Savell, T. Davies, I. Munro1 S. Parrinello, M.A. Purbhoo, C. Dunsby, M.A.A. Neil and P.M.W. French, To be published in Journal of
Biophotonics, DOI 10.1002/jbio.201500324
Nanoparticle-assisted STED, theory, and experimental demonstration
Yannick Sonnefraud, Yonatan Sivan, Hugo G. Sinclair, Christopher W. Dunsby, Mark A. Neil, Paul M. French, Stefan A. Maier, BiOS Photonics West, San Francisco, 2014
Proc. SPIE 9169, Nanoimaging and Nanospectroscopy II, 916903 (19 November 2014); doi: 10.1117/12.2060993