Serial Crystallography using x-ray Free Electron Lasers Francesco - - PowerPoint PPT Presentation
Serial Crystallography using x-ray Free Electron Lasers Francesco Stellato I.N.F.N. Sezione di Roma Tor Vergata Milan, July 11 th 2014 Summary Structural biology and X-rays From synchrotrons to Free Electron Lasers
Milan, July 11th 2014
Francesco Stellato I.N.F.N. – Sezione di Roma ‘Tor Vergata’
Structural biology and X-rays
From synchrotrons to Free Electron Lasers
Diffract-and-destroy measurements
Serial Crystallography at FELs
Sample Preparation and Charcterization
Sample delivery
Data analysis
The Cathepsin B experiment
Serial Crystallography at synchrotrons Applications & Future perspectives
1E+00 1E+03 1E+06 1E+09 1E+12 1E+15 1E+18 1E+21 1E+24 1E+27 1E+30 1E+33 1E+36 1880 1910 1940 1970 2000 2030 Year Source Peak Brilliance
Röntgen Bragg & Bragg reflections von Laue crystal diffraction Hodgkin penicillin, B12 Perutz & Kendrew myoglobin Franklin, Crick, Watson DNA MacKinnon Potassium channel Kornberg RNA polymerase Jacobsen Holography Kirz & Schmahl Microscopy
Free Electron Lasers (FELs) Radiation is generated by an undulator
Electrons are bunched up by interaction with x-rays
courtesy: Thomas Tschentscher (XFEL)
LINAC Coherent Light Source
FLASH
FLASH Hamburg, Germany λ > 4.2 nm LCLS Stanford USA λ > 0.12 nm
FELs around the world Soft x-rays
FERMI Trieste Italy SACLA Rikken Japan
Under construction
Hard x-rays
very different peak brilliance 1012 photons in ~0.05 μm2 FEL Pulse-rate FEL: 100 Hz (so far…)
10 -100 fs FEL 10-100 ps sinchrotrons
Up to 10 keV FELs (first harmonic) Up to 100 keV sinchrotrons
Diffraction pattern FEL puse Particle injection
One pulse,
A detectable signal must be recorded before the sample is destroyed
Diffract and destroy proof-of-principle First pulse
Second pulse
1 micron SEM picture of a FIB-bed pattern etched on a Si3N4membrane
Chapman et al. Nature Physics (2006)
Reconstructed image at 32 nm resolution
1 micron
FLASH Diffraction pattern
Diffraction pattern
The diffract-and-destroy principle can be exported for in principle all synchrotron x-ray techniques
200 nm
2D reconstruction of a mimi-virus from a single 200 fs LCLS pulse Seibert et al. Nature 470, p.78 (2011)
single crystal diffraction patterns
phasing methods
10000 20000 30000 40000 50000 60000 70000 1972 1976 1980 1984 1988 1992 1996 2000 2004 2008
Electron NMR X-ray
Standard crystallography is the election technique for structural biology
Experimental setup
beam
Pilot experiment
Chapman et al. Nature 470, (2011)
Photosystem I Sun-catcher Membrane protein 36 proteins 381 cofactors AMO beamline @ LCLS
Single shot at LCLS E = 1.8 keV 80 fs pulse 2 mJ pulse energy Upper front CCD Lower front CCD Resolution at corner = 8.6Å
beam center
Pilot experiment
Chapman et al. Nature 470, (2011)
LCLS data allowed solving PSI structure at 7 Å resolution (wavelength and geometry limit) The electron density is compatible with the known structure one Virtual powder data show that there is no damage up to 70 fs pulses Molecular replacement method is used starting from the known structure
First FEL based pdb structure Pdb ID: 3PCQ - www.pdb.org
Needle- shaped Cathepsin B Nanocrystals A Proteinase K Nanocrystal
SEM images
Standard techniques can be
and/or nano-crystals:
Several techniques are used to detect nanocrystals:
Several techniques are used to characterize nanocrystals in terms of quality, concentration and size distribution
A good sample delivery system should:
to native conditions
pulse
Systems used so far at FELs:
Hitrate (fraction of FEL pulses that hit a sample) is determined by
Examples of hitrate at LCLS
Liquid line Gas line Liquid jet
100 m/s 0.5-5 μm diameter 1-10 μl/min 10% hitrate
Gas line
Gas bottle Sample reservoir
De Ponte D et al. J. Phys. D 2008
A drop-on-demand system can be used to generate 20-40 μm diameter droplets An electrospray source can generate small droplets and an associated Differential Mobility Analyzer can size- select particles Cone-Jet Mode
Si3N4 membranes Ideal for 2D crystallography
Frank M. et al., IUCrJ 2014
Good to keep samples hydrated
Zarrine-Asfar A. et al., Acta D 2012 10 μm
Diffraction pattern acquisition Hit-finding Background subtraction Peak finding
Only ‘hits’ are processed Sparse patterns: average of many frames Peaks are identified in the bkg subtracted patterns
Indexing Intensities Intensities merging Structure factors
White T. et al. J. Appl. Cryst 2012 White T. et al. Acta D 2013
Standard programs (DirAx, MOSFLM, …) called by dedicated softwares (CrystFEL, Cctbx) The (partial) intensity is evaluated as a locally background subtracted sum of pixels close to the detected (or predicted) peak position
Ring-scheme Background Empty region Bragg peak
Cathepsin B Cysteine protease expressed by T.brucei, organism that causes Human African Trypanosomiasis
Luci di Sincrotrone CNR – Roma, 22 Aprile 2014 The structure of the protein in the non-native form is known, the glycosylated one not Baculovirus infection of insect cells is commonly used for the expression
post-translational modifications.
Needle-shaped crystals were observed in the cells over-expressing the protein They were purified and concentrated to reach about 109 #/ml 10 ml of concentrated solution were
Luci di Sincrotrone CNR – Roma, 22 Aprile 2014
SEM picture of a purified Cathepsin B crystal
Synchrotron data 60s exposure pattern have been collected at DORIS, Hamburg 1010 photons/s in 200x200 μm2 There is a clearly visible ring at 60 Å Faint rings at higher (20-40 Å) resolution. 1s exposure pattern have been collected at SLS, Switzerland 1011 photons/s in 20x20 μm2 Bragg spots are visible up to 8 Å Why such a low resolution? Essentially, because of damage
Single crystal diffraction pattern
Measurements at the CXI beamline - LCLS 9.4 keV 40 fs pulse-length 1011 photons/pulse 293,000 hits 175,000 indexed patterns
A virtual powder pattern
thousand single crystals patterns
Projection of the measured intensities on two planes in the reciprocal space
3D structure of the fully glycosylated protein
Redecke et al., Science 2013
Motivations
Warkentin et al. Acta Cryst. D D67 (2011)
The serial approach can be used at synchrotrons Peak brilliance is lower than FEL‘s one Exposure time must be longer Rotation during the exposure helps integrating the Bragg peak
Beamline P11 @ PETRA III – DESY Hamburg Photon energy: 10 keV Beam size: <10x10 µm2 Flux: 1012 photons/s Detector: PILATUS 6M – 172x172 µm2 pixels
Lysozyme microcrystals grown in batch in high-salt and high viscosity medium Crystal suspension flowing in a thin-walled SAXS capillary at 2.5 l/min Exposure time: 10 ms
Hit-finding
Indexing
2.1 Å Bragg spots visible up to 2 Å resolution
Lysozyme structure solved at 2.1Å resolution by molecular replacement merging intensities from 40,000 single crystal diffraction patterns Pdb ID: 4O34
Stellato F. et al., IUCrJ 2014
Unit cell parameters are in excellent agreement with known values
a = (79.50.3) Å b = (79.40.3) Å c = (38.40.2) Å
10,000 single crystal diffraction patterns would probably be enough
Stellato F. et al., IUCrJ 2014
500 patterns from Cathepsin B microcrystals at cryogenic temperature Structure solved at 3Å resolution
Gati et al., IUCrJ 2014
Luci di Sincrotrone CNR – Roma, 22 Aprile 2014
Aquila et al. Optics Express 470 (2011) Kupitz et al. Nature (2014)
Changes observed in the putative S3 state in the Photosystem II complex
GPCR in Lipidic Cubic Phase
Liu et al. Science (2013) Johanssonn et al. Nature Methods (2012)
Photosyntetic reaction centers in Sponge phase
X-ray emission (XES) X-ray absorption (XAS) Small angle scattering (SAXS) Wide angle scattering (WAXS)
Higher and higher brilliance will enable approaching the limit of high-resolution single molecule imaging
Luci di Sincrotrone CNR – Roma, 22 Aprile 2014
The Biophysics Group in Tor Vergata biophys.roma2.infn.it
Silvia Morante Giancarlo Rossi Velia Minicozzi Francesco Stellato Marco Pascucci Claudia Narcisi Emiliano De Santis CFEL-DESY
Nass, F. Stellato, H. Fleckenstein, L. Galli, R. Kirian, K. Beyerlein Arizona State Univeristy
Kupitz SLAC
Uppsala Univeristy
Svenda, T. Ekeberg, J. Andreasson, A. Rocker, O. Jonsson, D. Westphal University of Tübingen, Hamburg and Lübeck
Duszenko, T. Stehle Max Planck Heidelberg, LBNL, LLNL European XFEL Massimo Altarellii
Contacts Francesco Stellato I.N.F.N. Sezione di Roma Tor Vergata Via della Ricerca Scientifica, 1 Tel: 0039 06 7259 4284 francesco.stellato@roma2.infn.it