Seminaire 2016

Séminaire 1 :Fluorescence lifetime imaging across the scales – from automated microscopy to tomography and endoscopy

Paul M. W. French

Photonics Group, Physics Department, Imperial College London
We develop and apply FLIM technology across the scales from 3-D STED FLIM microscopy1, incorporating adaptive optics to correct for aberrations, through high-speed FLIM of fixed and live cells2, e.g. for high content analysis (HCA) including automated multiwell plate assays of protein interactions3,4 or cellular metabolism5, to in vivo preclinical and clinical applications, including autofluorescence-based lifetime readouts of cancer6, heart disease7 and osteoarthritis8. For in vivo preclinical studies, we are developing optical projection tomography (OPT) with FLIM, particularly applied to zebrafish, and FLIM endoscopy9. For clinical studies we are developing multiphoton multispectral FLIM to diagnose skin cancer and single point fibre-optic probes for time-resolved and spectrally resolved fluorometry of autofluorescence, including an ex vivo and in vivo studies of cancer in skin and colon.
This talk will focus on our development of FLIM instrumentation for HCA to study signalling networks and mechanisms of disease in 2-D and 3-D cell-based assays, including the intracellular measurement of KD to study signalling networks10, on the development of a preclinical imaging platform based on in vivo OPT and FLIM11 of zebrafish (larvae to adult) for studies of cancer12 and apoptosis13, on FLIM endoscopy applied to FRET
SEMINAIRES
We aim to implement all of our technology using open source software tools for instrument control, data acquisition, analysis and management and to provide lists of equipment components to enable other users to replicate our instrumentation for application to their own biological questions. Our open source FLIM analysis software, FLIMfit15, which provides rapid global fitting capabilities and is available as an OMERO client from: http://flimfit.org/downloads/. Our FLIM HCA and OPT instrumentation are controlled by μManager. The software for FLIM HCA is available at: https://github.com/imperial-photonics/openFLIM-HCA/wiki.

References
1 Lenz et al., J. Biophotonics 7 (2013) 29-36, doi: 10.1002/jbio.201300041
2 Grant et al., Opt. Expr.15 (2007) 15656 – 15673; doi: 10.1364/OE.15.015656
3 Alibhai et al., J. Biophotonics 6 (2012) 398-408, doi: 10.1002/jbio.201200185
4 Kelly et al., Anal. Methods, 7 (2015), 4071-4089, doi: 10.1039/C5AY00244C
5 J. Innov. Opt. Health Sci. 7 (2014) 1450025-15 pages, doi: 10.1142/S1793545814500254
6 Patalay et al., PLoS ONE 7 (9) e43460-, 2012, doi:10.1371/journal.pone.0043460
7 Lagarto et al., Biomed Opt Express 2015, 6 (2015) 324-346; doi: 10.1364/BOE.6.000324
8 Manning et al., Matrix Biology 32 (2013) 32–38; doi: 10.1016/j.matbio.2012.11.012
9 Kennedy et al., Journal of Biophotonics, 3 (2010) 103-107; doi: 10.1002/jbio.200910065
10 Margineanu et al., Sci. Rep. (early view) doi: 10.1038/srep28186 (early view)
11 McGinty et al., Biomed. Opt. Expr. 2 (2011) 1340-1350; doi: 10.1364/BOE.2.001340
12 Kumar et al., Oncotarget.7 (2016) 43939-43948; doi: 10.18632/oncotarget.9756
13 Andrews et al., J. Biophotonics 9 (2016) 414–424. doi:10.1002/jbio.201500258
14 Sparks et al., J Biophotonics 8 (2015) 168–178; doi: 10.1002/jbio.201300203
15 Warren et al., PLoS ONE 8(2013) e70687; doi:10.1371/journal.pone.0070687

Séminaire 2 :Receptor-scaffold interactions, a thermodynamic approach using super-resolution microscopy: toward “chemistry in cellulo

Antoine Triller
IBENS Institut de Biologie de l’École Normale Supérieure

The efficiency and accuracy of neurotransmission strongly depends on two apparently antagonist properties of synaptic membrane: the stability of its organization and its ability to adapt to plasticity events. In addition, the structural stability of synapses has to be reconciled with the notion that cell membranes are fluid. Membrane molecules are compelled to move within the membrane surface due to thermal Brownian agitation, which favors the homogeneous distribution of the molecules. As a result, neurons spend energy to stop or reduce these movements, and maintain molecules in certain locations via mechanisms that decrease this fluidity. We investigate the regulation of synaptic receptors dynamics by the different (structural and functional) elements that make the synapse. We have approached these conceptual paradoxes by developing new technological and analytical tools that allow the monitoring of the behavior of synaptic components at the molecular level and change of the scale of analysis. We demonstrated rapid exchanges between synaptic and extra-synaptic receptors and we showed that transient stabilization of receptors at synapses occurs by interaction with partners, such as scaffold proteins. Novel super-resolution imaging methods (PALM, STORM) gave us a precise insight on the organization of these postsynaptic structures. Thus combination of single particle tracking and super-resolution methods, open access to molecular counting and energy involved in receptor-scaffold interactions as well as on and off rate of molecular interactions. Thus beyond super-resolution methods is chemistry “in cellulo” accounting for the regulation of receptor number and consecutively that of synaptic strength. Ultimately, the dynamic regulations of receptor-scaffold and scaffold-scaffold interactions appear as a central tenet for the maintenance and plasticity-related changes of receptor numbers at synapses. These processes are likely to be deregulated in pathological situations such as in neurodegenerative diseases.

Séminaire 3 :Multi-scale multi-dimensional and multi-modal analysis of cell-matrix adhesion

Benny Geiger1, Ariel Livne1, Melanie Horev1, Or-Yam Revach1, Ohad Medalia2, Joachim Spatz3
1Weizmann Institute of Science, Rehovot, Israel;
2Zurich University, Switzerland;
3MPI for Intelligent Systems, Stuttgart, Germany

Integrin adhesions, through which cells stably interact with the extracellular matrix (ECM) are structurally stable and molecularly dynamic cellular structures, located at the interface between the ECM, outside the cell, and the cytoskeleton, at the cell’s interior. These adhesions are extensively investigated for nearly 60 years, yet despite the vast effort invested in their characterization, their structure, diverse functions and molecular dynamics are still highly challenging. These adhesions are quite robust, yet they are highly dynamic structures at the molecular level. Integrin adhesions consist of a largely uniform set of components (termed, collectively: “the integrin adhesome”, yet the diverse forms of integrin adhesions are highly complex, and molecularly diverse. Moreover, these adhesions are mechanically robust, acting as cell and tissue scaffolds. Beyond their scaffolding activity, they function as key transmembrane sensory and cell signaling sites. The main issue to be addressed in this lecture is the mechanism whereby the molecular structure and functions of integrin adhesions are regulated, enabling them to function as cellular scaffolds, environmental sensors and singaling organelles. I will address this issue using a broad range of complementary multidisciplinary approaches, including correlated light and electron microscopy analyses, quantitative live cell imaging and a multitude of physical approaches. I intend to address here recent insights into the molecular hierarchical structure of the different forms of integrin adhesions and the mechanisms underlying their activity as chemical and mechanical sensors of the extracellular environment.

Séminaire 4 :The Extra Microscope

Alberto Diaspro
Nanoscopy and NIC@IIT, Istituto Italiano di Tecnologia and department of Physics, University of Genova, Italy

In the last 40 years, optical fluorescence microscopy, due to its inherent ability of imaging living systems during their temporal evolution, had a continuous update on three main tracks, namely: three-dimensional (3D) imaging, penetration depth at low perturbation and resolution improvements. However, computational optical sectioning, confocal laser scanning, two-photon excitation and super-resolved methods can be considered as milestones in optical microscopy. With the recent Nobel Prize in Chemistry in 2014 for the development of super-resolved fluorescence microscopy, it has been revealed how the optical microscope can offer unimaginable performances in terms of spatial resolution. This fact pushed a myriad of variations on the theme and of new approaches aiming to increase the budget of information that one can get using an optical microscope. The crumbling of the diffraction limit has been demonstrated in ‘‘super-resolved ensemble fluorophore microscopy’’ and ‘’superresolved single fluorophore microscopy’’ also known as targeted and stochastic read out methods, respectively [Diaspro A. 2014. Il Nuovo Saggiatore]. Starting from this point, considering important revolutions like the ones of confocal and two-photon excitation microscopy coupled to the advent of green fluorescent proteins [Diaspro A. and van Zandvoort M.A.M.J. (eds) 2016. Super-resolution Imaging in Biomedicine. CRC press], I will discuss about some converging and correlative techniques that can be used for what I like to name “the extra microscope”. The meaning is related to series of advances and new methods that allow getting information at the nanoscale or preferable at the molecular scale referring to both spatial resolution or structural information. Attention will be given to those imaging techniques that permit direct measurements of the live-cell molecular dynamics at the nanometer scale including the study of thick biological samples. An important list of converging technologies is currently under development, for example: recent advances in camera technology and real-time image processing have led to substantially improved time resolution, RESOLFT nanoscopy and the utilization of temporal information for decoding spatial information allowed super resolution at reduced beam intensities, the advent of new fluorescent molecules are providing better quantitative abilities, label free approaches in the IR, original image deconvolution processing for noise removal, in vivo imaging, new lenses and beam shapers for fast imaging. Moreover, phase methods are gaining relevance in a real multimodal context. The Extra Microscope has tunable and flexible performances depending on the biological question. It is worth noting that new correlative approaches coupling optical super resolved methods with scanning probe microscopes are providing interesting developments that will be outlined.

References
1 Chacko, J.V. et al.2013. Cytoskeleton;
2 Monserrate, et al. 2013. ChemPhysChem
3 Viero G. et al. 2015 J.Cell.Biol

Séminaire 5 :Using various imaging techniques to establish a new regeneration and aging model

Eric Rottinger
Institute for Research on Cancer and Aging, INSERM U1081 – CNRS UMR 7284

The sea anemone Nematostella vectensis is a well-known model to study embryonic development and metazoan evolution. This marine invertebrate possesses also extraordinary regeneration capacities, as it is able to regrow missing body parts in only seven days. However, quite little is known about the morphological, cellular and molecular mechanisms underlying this fascinating biological process.
In the present study, we lay down the basic framework to study oral regeneration in Nematostella vectensis. Using various imaging techniques (macrophotography, epi-fluorescence and confocal microscopy, cytometry etc…), we characterized in detail the morphological, cellular, and global molecular events that define specific landmarks of this process. Furthermore, we describe in vivo assays to evaluate wound healing success and the initiation of pharynx reformation. These specific landmarks as well as the tools we developed, are currently used to characterize precisely the effects of perturbation experiments on the regenerative process in Nematostella vectensis.

Séminaire 6 :Séminaire historique sur Marvin Minsky et Roger Tsien

Marvin Lee Minsky, né le 9 août 1927 à New York et mort le 24 janvier 2016 à Boston1, est un scientifique américain. Il a travaillé dans le domaine des sciences cognitives et de l’intelligence artificielle. Il est également cofondateur, avec l’informaticien John McCarthy du Groupe d’intelligence artificielle du Massachusetts Institute of Technology (MIT) et auteur de nombreuses publications aussi bien en intelligence artificielle qu’en philosophie comme La Société de l’Esprit (1986)
Roger Y. Tsien, né le 1er février 1952 à New York, et mort le 24 août 2016, est un biochimiste et biophysicien américain d’origine chinoise, corécipiendaire du prix Nobel de chimie de 2008 avec Osamu Shimomura et Martin Chalfie

Séminaire 7 :Imaging across scales using electron and light microscopy: from protein complexes to model organisms

Jan Ellenberg
Cell Biology and Biophysics Unit, European Molecular Biology Laboratory (EMBL)

Recent developments in imaging technologies have given us an unprecedented ability to visualize the molecular mechanisms of life from the scale of single protein complexes to whole model organisms. In this lecture I will use the examples of the nuclear pore complex and the protein network controlling cell division to illustrate how we can seamlessly cross between these scales and what we can learn from new imaging technologies including super-resolution, correlative light electron, high-throughput and light-sheet microscopy.
Essential functions of life, such as cell division, require several hundreds of proteins and the molecular machines they form to work together. Super-resolution microscopy, now allows us to understand the 3D molecular architecture of single molecular machines, such as the nuclear pore complex in situ inside the human cell. To understand how such machines are assembled, we can employ correlative light electron microscopy to stage assembly steps in a living cell and then analyze their ultrastructure. To understand how many molecular machines are orchestrated in space and time requires mapping their dynamic interactions inside cells. To this end, we have established an integrated systems microscopy workflow, combining genome editing, imaging and computational modeling to map the protein network that drives cell division in live dividing human cells. After homozygous genome editing using CRISPR/Cas9 to tag all endogenous copies of a given protein fluorescently, the subcellular concentration distribution of each protein is measured by high-throughput single molecule calibrated imaging. Computational image analysis is then used to generate a standard mitotic cell model into which the data of all imaged proteins is integrated. Machine learning-based mining of the data-driven model can map dynamic protein networks. Finally, to understand how complex essential functions such as cell division occur in the multicellular context of a whole organisms, we have developed new light-sheet microscopy technology, which for the first time allows us to follow the molecular machinery of cell division in real time in a developing mouse embryo.

Séminaire 8 :How statistics can help improve microscopy image analysis

Ivo F. Sbalzarini
MOSAIC Group, Center for Systems Biology Dresden
Chair of Scientific Computing for Systems Biology, Faculty of Computer Science, TU Dresden
& Max Planck Institute of Molecular Cell Biology and Genetics, Dresden, Germany

Microscopy image data are different from the photographs or satellite imagery traditionally considered in machine vision and image processing. Biological images are often three- or even higher-dimensional, and not standardized. Moreover, they frequently only become interpretable when considering additional prior knowledge, e.g., about the labeled structures or the optics of the microscope. Here is where statistics and knowledge about the human visual system can help. We show how to formulate image segmentation and object detection as statistical inference problems, and how this allows us to systematically include prior knowledge, or models, about the images. This enables more flexible algorithms that can be customized to specific tasks by adjusting the models. It also provides us with a principled way of quantifying the accuracy and probability of the result, as well as designing models based on physics. This interdisciplinary community effort ultimately results in novel algorithms and user-friendly software. We show this in the example of fluorescence microscopy, where this approach has enabled versatile and statistically optimal segmentation algorithms, physics-based prior models, and popular open-source projects.

Séminaire 9 Les défis de l’innovation : introduction à la conception innovante

Benoit Weil
MINES ParisTech, Chair Design Theory and Methods for Innovation