Transcript side view
Architecture of the photosynthetic apparatus by electron microscopy Egbert Boekema Leiden March 2009 Dear keynote speakers in our Solar Biofuels of Microorganisms Workshop <http://www.lorentzcenter.nl/lc/web/2009/333/info.php3?wsid=333>, The workshop is embedded in the Leiden University Honours programme, and there will be 20 of our best bachelor students participating. We have comfortable slots for the talks and the discussion, and with this email I would like to ask you not to hesitate to include an educational dimension in your lecture, it will be appreciated, both by our students and by the participant out side your own field in this multidisciplinary workshop. Thanking you for your efforts, and looking forward to seeing you soon in Leiden. Kind regards, -on behalf of the organizersCorrie Kuster Electron microscopy is possible at atomic resolution Unfiltered image of a copper phtalocyanin crystal removal of noise by averaging of many images Photosystem I trimer + 18 antenna proteins antenna protein = IsiA, the iron stress induced protein A of 37 kDa “single particle averaging” main steps in single particle averaging • EM + selection of particle projections • alignment of randomly oriented projections rotational + translational shifts • sorting of projections statistical analysis + classification • calculation of two-dimensional projection maps summing of projections into “classes” • calculation of 3D structures symmetrical class tilted class tilted class Resolution in single particle cryo-3D reconstructions object mass symmetry number projections resolution (Å) none 4,000 20,000 75,000 25 17 10 worm hemoglobin 4000 kDa Transferrin receptor 290 kDa transferrin complex 12-fold 1,000 13 2-fold 36,000 7.5 2300 kDa 4-fold 22,000 14 823 kDa 14-fold >5000 kDa >20-fold 10,000 40,000 8,400 8 6 4 70 s ribosome Ca release channel GroEL protein 6 rotavirus 3000 kDa First protein at atomic resolution viral protein 6 in rotavirus DLP: 8,400 particles 8,400 x 60 x 13 = 6.6 million copies Zhang et al. PNAS 2008, 105, 1867 Cryo-EM image Cryo-EM picture showing Virus particles in a thin layer of ice of a holey carbon film lower resolution presentation of virus reconstruction Electron density map plus Assignment of amino acid amino acid side chain side chains in the 3D fit map (blue wires) A test object: worm hemoglobin 12 x 12 proteins 18 linker proteins 18 linker proteins 100 Å Most complicated step in single particle averaging sorting of projections statistical analysis + classification symmetrical class tilted class tilted class Gallery of aligned top- and side views Classification map after statistical analysis Factor 2 Factor 1 Each dot is a particle in side-view position close in space = high similarity Classification of aligned side-view projections A E C B F G D H I partition of data set into 9 classes Position of classes in the classification map D I H C G B F E A Relationship between side-views and top views beam predominant position 1 “broad type” side view Support film “narrow type” side view predominant position 2 Support film Sinograms of individual hemoglobin classes to find searching common lines Worm Hemoglobin 3D Model EM X-ray 1988 2009 sum of 1024 particles 11 Å resolution in negative stain > 2010 Photographic emulsion 5000 particles 1 minute / particle Handcraft CCD cameras (200,000 €) 50,000 particles 1000 particles / minute Semi-automation Atomic resolution electron counters (800,000 €) 500,000 particles 10000 particles / minute remote-control Seeing is believing The skull from Dali Seeing is believing Complex III (Cytochrome reductase) EM (18 Å resolution) The skull from Dali X-ray Example of combining EM and X-ray diffraction Cytochrome reductase – and cytochrome oxidase supercomplex (Heinemeyer et al. 2007 J. Biol. Chem. 282, 12240 maps of the supercomplex and a fragment (left) show enough fine structure to dock the complex III and IV crystal structures accurately into the EM density maps Conclusion: from 15 Å EM data + X-ray structures we get a pseudo-atomic model, which has enough resolution to predict interaction of alpha helices of different subunits Scheme of the cyanobacterial membrane Phycobilisome ATPase PSI PSII Cytb6f NDH-1 Cyanobacteria do not have a membrane-bound antenna with LHCH2 Rows of PSII are a scaffold for the phycobilisomes but nobody knows how Phycobilisome (PBS) PBS components known at high resolution Phycobilisomes are floppy: Structure work on truncated PBSs Need for solving interaction with PSII-PSI, FNR, quenching proteins Single particle analysis of PBSs Single particle electron microscopy digitonin-solubilized cyanobacterial membranes Photosystem 1 Complex I ATPase Photosystem 2 50 nm Selected gallery of projection maps from 15,000 projections Performed on Synechocystis 6803 / Thermocynechococcus elongatus 733 351 304 312 512 512 218 1230 ~300 50 291 7 Seeing is believing Some ’’’assignments’’’ Phycobilisome Glutamine synthase ATP synthase from the PDB site “molecule of the month displays” T-shaped particle fragment GroEl-GroES Small Photosystem II arrays in solubilized membranes from Synechocystis 6803 Analysis of Photosystem II arrays and double dimers 16.7 nm 12.5 nm Phycobilisome model Analysis of Photosystem II double dimers Double dimer model Is there a specific subunit involved in double dimer formation? Models for the photosynthetic membrane