2017 Nobel Lectures in Chemistry

welcome everyone and could you please take a seat so we could start [Music] okay the dear laureates colleagues friends students ladies and gentlemen when we listen to music or look at the world around us or smell the air when the first snow is falling there are molecules in our cells that move change their shapes bind to each other imagine if we could see all this and it became possible through the work of this year's laureates so the Nobel Prize for chemistry 2017 has been awarded to Jacques DeBusschere he work in Frank and Richard Henderson for developing cryo-electron microscopy for the high resolution structure determination of biomolecules in solution if this sounds complicated on the next slide cool microscopy captures life in atomic detail this is what it is about the electron microscope was invented almost a hundred years ago it would take many decades before the instrument could be used to study biological molecules then many more decades passed and during that time we could only see molecules at atomic at relatively low resolution as seen on the left side we could see the outer shapes or blobs of those caused and that field was referred to as blue Balaji but some people then in a relatively short time the information content in the pictures were was increased by a factor of thousand and we could see their atoms here of the molecules and this is where we are today about 25 years ago Richard Henderson demonstrated for the first time that cryo-electron microscopy could be used to visualize biological objects at atomic resolution already at that time he predicted that the technique and this is what he wrote in his article could become a routine and quick method to be used for obtaining these structures this is where we are today and many of the more recent improvements have also been done by Richard but during the years we have not only been able to harvest the fruits of Richard's work he has also generously shared the seeds with his colleagues and co-workers so that they could grow their own research environments and directions and I think that has been very important in the field so now it's my privilege to introduce the first speaker Richard Anderson please [Applause] [Music] so thank you very much indeed Peter for that kind introduction and of course it's also a great honor to be selected by the Nobel Committee for chemistry and the Royal Swedish Academy to to share with Jacques and Yocum this year's 2017 Nobel Prize in Chemistry but I wanted to begin by saying that the three of us are really representatives of what's now become quite closely knit and well organized filled with a phenomenal growth rate so I'd like to start by showing the first slide which is from the first three dimensional electron microscopy annual Gordon conference that was organized in 1985 by the two people in the yellow circles nigel Anwen and watch ooh and so they had the idea this field was going to grow and inaugurated this meeting which is now held every year and I think you can see in the red circles that Jacques Yocum and I are clustered around being as close as possible to Nigel and wah but there are other people in in the and this is about a hundred people who are where the field was starting there's a few more who weren't at the meeting but I particularly wanted to point out because the group of Bob Glaser Ken Taylor and Ken downing they were doing low-temperature electron microscopy cryo microscopy before those of us who are here today so they were early adopters of that idea and then because I'm the first speaker I wanted to show also a kind of overview of so that you know what is electron cry microscopy which we call cry wee M for short and it comes in many different flavors because once you have a good electron cry microscope you can put all sorts of different specimens into it and this slide sort of summarizes a variety of the things that have been studied and have been quite informative so on the left is a field of view showing amorphous ice with ribosomes embedded in it this is single particles without symmetry and this is from work that you will hear more about from yoke and Frank and this was their state of play in roughly the year 2000 from a review that Tim Baker and I wrote so single particles without symmetry single particles with symmetry this is hepatitis B virus caused from Bettina butchers work in 1997 this was the first structure that went below nanometer below 10 angstrom resolution and then the third panel is helical arrangements of biomolecules so this is the thin filaments in muscle decorated with subunits of myosin from the work of Ron Milliken the blue is actin the green is tropomyosin and the different colors pink orange and yellow are the different domains of the myosin that's involved in muscle contractility this is relatively low-resolution all of these are fairly low-resolution but on the right is the 2d crystals that I'll talk about it at the beginning of my own part of the talk this was the second high-resolution structure determined by cryo a.m. by Verner cool Brandt and with collaboration with fujiyoshi in Japan in in 1994 and shows a 2d crystal at higher resolution because the 2d crystals were easier to start with which is how I'll start so that is a range of specimens giving you a kind of overview the structures here are about six times bigger magnification than the images at the top but you can see and it's this is so this is much easier to understand than the gravitational waves and the black holes you just take images of the molecules you can see them and then in the computer you calculate the structure so you know anyone can do this kind of work so then I want to jump forward and this is this year's Gordon conference so this is now 32 years later it's a it's a bigger group more highly selected you see some of the faces Bob Glaser and and Yocum myself but I've outlined in yellow some of the people who more recently have made very important contributions in terms of developing all the computer programs it has make it so easy and so popular amongst newcomers who don't know so much about it so Fred Singh was one of the original maximum maximum likelihood sure share as an equal Grigory F Marion van heel who'd worked with Yocum at one point and and correct so from Spain and then to other people in the pale blue Bridget Carey Kerr and Helen Sabol who in the USA and in Europe have been very important in running courses and educating people although they also did individual work in the past so that gives you a little bit of an overview of the community and how the closely knit sort of brother and sisterhood of cryo-em has adopted its own sort of culture and identity and that's very important and so we are only the tip of the iceberg if you like in this crime area and then to give you an idea of where we are now this is the growth in the deep positions of high-resolution atomic coordinates in the protein databank which traditionally had only coordinates deposited from x-ray crystallography and nuclear magnetic resonance spectroscopy and you see back in in the 1990s and so on there were invisible numbers of structures but in the last five years has been a great growth and it looks like this year we learned that with about 700 new structures deposited in this protein databank more maps and and we then have a we have the Gordon conferences there's a 3d en mailing list which didn't exist until 1995 started by Ross Smith but when that first Gordon conference started we were probably about a hundred or 150 people now there are 3,000 people who subscribe to that mailing list so that gives you an idea of the field and the importance of the community so now I just like to I'm going to do three things in this talk so tell you how I got involved in being converted by meeting Nigel and went from an x-ray crystallographer to an electron microscopy enthusiast and then how we work together to get a low resolution structure of a membrane protein bacteria Dobson then a high resolution structure and then how our own efforts switched from being electron crystallography into single particle e/m and which has been the methodology that's made the method really popular so going back now to about in my case 1971 or 72 I heard a talk by water stickiness in San Francisco this is the first time he talked about it he had in a bacterium halo bacteria these membrane areas the his creations these are little two-dimensional crystals one molecule thick and with backing and support from Don angleman and Tom Stites at Yale we phoned up zucchinis and got some samples of these bacteria grew the map and then purified these membranes and each one of these is a little single crystal maybe a hundred molecules or two hundred across so 20 or 30,000 of these identical protein molecules in this 2d crystal and my original plan had been to dissolve them in detergent make 3d crystals or analyze the 2d pattern this is an x-ray powder pattern from stacks pellets of those same membranes that you saw in single sheets and so there are lots of spots you can see they go out this would be the one the first spot second third and so on and this is about seven angstrom resolution they're all very strong but you can see them all going out and you can index them right oh right the way out to three or four angstrom resolution so that showed that these 2d crystals were really highly ordered and very suitable for structural analysis and at that point I had the great good fortune to meet Nigel Unwin who had been giving a talk about electron microscopy using negative stain but in that top it was clear he was thinking about how the images would also contain inside the stain the images of the proteins that were being encased in it so we work together for two or three years doing on these little patches that would be one of these crystals electron diffraction and these spots are the diffraction spots from the lattice and molecules in the crystal and images that could then be analyzed either in the computer or by optical diffraction to give you the amplitudes and phases of the furry components that describe the distribution in the matter in the crystals and when we we only saw the spots and we had numbers we wrote them a sheet of paper and then to our great delight when we submitted these averages and phases to the computer and then went down to the big computer in Cambridge University we got this map which was surprisingly informative and very beautiful we thought the the plotter which was the only accurate plotter in all of Cambridge at a time had three color pens the red green and the blue so the red green and the black and it showed in projection three of these protein molecules with big Peaks 10 angstroms apart and and we'd already suspected from the x-ray pattern that there were alpha helixes this is the Pauling quarry alpha helix from the model building that needed in the 1950s and which had been discovered in myoglobin and hemoglobin but here we now have these alpha helix YZ pointing towards us in a top view of the membranes so then we continued tilted the specimens and went into 3d got these 3d maps that are plotted on sheets of perspex with hand drawn the the computer plotter couldn't do floating on per spec sheet so these had to be hand drawn from computer plots and then looking at it sideways we built this balsa wood model in 3d which showed that not only the four after Gila C's that could be seen and on in the projection but with slight tilts there were seven transmembrane helix season this was the first real strong and hard knowledge about the structure of membrane proteins 1975 we were at that point stop in terms of knowing that the crystals were better ordered but not knowing how to to proceed and then we tried various things but in the interim Hartmann would Mikkel who had been also working on bhakti Robson managed to crystallize another membrane protein so in the end the first high-resolution membrane structure was the reaction centers that Heartland Michael and his colleagues from artist Reed were given the chemistry Nobel Prize in 1988 which was before we managed to do any better with this so then we went on and tried to go from seven angstroms to a higher resolution where you could see the chemistry the amino acids in the side chains and that needed a couple of things it needed higher resolution electron diffraction patterns and all the work I've shown you up to now was done with specimens at room temperature where the water that normally surrounds the the the protein that the membranes had been replaced by a sugar solution by glucose but now we had cold stages in Cambridge that we could take electron diffraction patterns on film initially and this one was later with the development of electronic detectors but our microscopes in Cambridge were not capable of doing high-resolution imaging of cryo specimens and and what encouraged us was this paper from watch you the same person who chaired the very first program of 3d M Gordon conference he had worked with Fritz Sam Lynn in Berlin who had a very high-performance prototype microscope that had been made by the siemens company liquid helium lens a liquid helium stage and he took pictures of this two dimensional three slightly thin 3d crystal of pro toxin a structure we still don't know to this day but he showed images of crystals with diffraction spots from the bridges that went beyond 4 angstroms 23.9 just the resolution that we were looking for so this and the existence of the electron diffraction patterns with cryo convinced us that we should make a major effort then to get into high resolution electron cryo microscopy with the idea of determining the stretch of this membrane protein but on in doing that we did try other things for many years and so on this slide in red is what worked and in these other three panels are other ideas that we tried so tom Seska David a guard and Joyce Baldwin tried this was our idea to bring the methods of x-ray crystallography into electron crystallography but they weren't they just weren't powerful enough the heavy atoms don't scatter as much any electrons and so on this was the idea of using heavy atom derivatives we found the heavy atoms but the phasing power was not enough we tried model building we knew where the HeLa C's were so in principle you could search for the location wasn't perfect after we tried molecular replacement with more than one crystal form it worked slightly but again not powerful enough what worked in the end was to take images just like you saw for Crowe toxin from watches work in Berlin and then also we began actually with a one-week trip to Jacques Dubois Shay's lab who was an EMBL at that time he'll tell you about this later and Zhang de Paul worked for a week on a really very difficult to use microscope I think which they dismantled after that we did get one image the very first high-resolution image which then gave us further encouragement and then visiting Burton's 1984-1985 following in watch his footsteps we went to to Berlin Fritz Emmaline eric Becklund worked for many years 5 or 6 years taking images making the microscope better and then in parallel Bob Glaser who had been visiting us persuaded Ken downing also to take pictures so in the end we had contributions from three different labs travel and we also tried in Cambridge but we never got image as good as these these were all prototype noncommercial not available and of course after it worked we then try to put pressure on the microscope manufacturers to develop the commercial versions which mean it is what we now use in parallel with the practical problem of getting really good high resolution images there were a number of technical things that I won't go into but particularly finding out that beam tilt was one of the critical problems that we hadn't realized and I'll figure out a way in tilted specimens to correct for the different height the different D focus but after doing that we ended up with improved maps so eventually from 1975 to 1950 years worth of various problems I don't think it cost nearly so much as the gravity wave project but we went from this is the same map I showed you with the balsa wood model 7 angstrom resolution in 1975 slightly better by molecular replacement these features turned out to be real although we weren't sure at the time and then the first 50 images gave us a map with more better resolve features and eventually with about 70 we got a map where you can see key features sticking out from the alpha helixes and at higher magnification they looked like this and when we then tried to interpret that map using the amino acid sequence from ovchinnikov and caranas lab which by then had been determined by biochemistry we found that you could interpret the map and see phenylalanine tyrosine tryptophan and other tyrosine and then the retinal the color of the membranes they were purple we could see all of these structures and built a complete atomic model of the battery ops and that was about 1990 and it was the first high-resolution structure where you could use electron microscopy to determine the structures so then we went on and did other work in the early 1990s nickel grigoriev refined the structure Sri Ram came we did lots of kinetic strapping intermediates and so on but at that time I also got involved in a review for an x-ray microscope and ended up writing a review comparing the radiation damage from neutrons x-rays electrons and I then became completely convinced that the e/m method was definitely the way to go and had an enormous untapped potential so we switched our effort from working on electron crystallography with crystals you still have to make crystals whether it's in 2d or 3d to working on single particles and and trying to assist the method in going from the relatively low resolution to try to go to a higher resolution and of course in this respect Jacques will tell you about this in due course Jaques group at the MBL had been exploring the behavior of water when he froze it and they had a variety of of gravity-driven or elastic band driven freezing apparatuses and I thought I'd just since I'm speaking first I should give you a little outline of it and this is the the cryo e/m grid preparation method that mark Adrien Alastair McDowell who's here and Jacques developed in the early 1980s the idea is you take grids that have got wholly carbon films on it this says wholly carbon film you you take them one at a time in a pair of forceps you apply a drop of your solution with your protein of choice or you arrive in some or whatever you blot it on a piece of filter paper and then you plunge it one or two meters per second into liquid ethane which was allister's idea occasionally when we were working on bacteria we also flashed it with light but when you did that you get images beautiful images where you see the amorphous ice this was a specimen that Richard param gave us Peter Rosenthal made the specimen veena of Kumar took this image quite recently with the new detectors you can see the particles you could see their orientation it's clear without doing any further computation that it's really powerful and then this is the very first hepatitis V virus core work that came from Tony Grider's group done by Bettina butcher and Nikolay kiselev had brought from Riga in Latvia a specimen early on in 1994 and using older microscopes not such bright sources not very Korean they got a first 30 angstrom resolution cryo-em structure in 1994 and then we bought a a higher voltage field emission gun brighter source microscope which Bettina used with specimens that Sam when had made in Tony's computer program and they got this then first structure from single-party work to go below 10 angstroms we called it sub nanometer structure and of course now we're now 20 years further on this is a recent paper published from Hong group at UCLA higher voltage brighter microscope much more stable but still all of these are pictures taken on film and they got a 3.5 angstrom resolution structure you can see these are after he received a bundle of four alpha helixes you can see the side chains and so on so the microscopes due to pressure from the academic researchers but response by the microscope countries produce much better data so then the thing that's kicked the field off in the last few years these new detectors came in previously we were using film there are three companies that now make detectors there now better than this they're now up here this would be a perfect detector at the top using these detectors I'm going to show you just two structures one of them is beating like societies which I think was the structure that Peter had in his introductory slide this is a half a mega Dalton structure it's the lack see the very first operon discovered in the early stages of understanding gene expression and we had adopted this in 1997 when Sriram Subramanian and Jacqueline mills were there learning and deciding to get into cryo a.m. and this was the first we didn't believe this structure in 2005 low resolution this is the first time we believed it this was done on film and we hadn't by then we had new analytical methods till pairs and so on that proved that this was correct but as soon as the new detectors came in here we had higher resolution and then with new programs such as rely on from Shores Charon the same data gave us much higher resolution you could resolve then the strands of the polypeptide in the beta sheets and then Sriram who had by then an independent group at NIH worked much harder on this turn the magnet collected better data and has now a 2.2 angstrom data structure and I've believed now this is now better than 2 angstroms from subsequent work and then a second one I wanted to show you is mitochondrial complex one which again nickel grigoriev had worked on back in the late 1990s with film and got to about 20 angstrom resolution recently Judy Hearst from another MRC unit brought a sample across and vinoth Kumar started to take it this is one of his pictures this is a one mega Dalton complex with 45 different proteins in it and it has no symmetry you can see it has also two different views when you record images and process them you get a map like this in 3d contour at one level or at a higher level and you can see the eight iron sulfur complexes where electrons are fed in from nadh come down into the membrane into quinone and involved in proton transport and then in the membrane there are 76 transmembrane helix sees this is about 5 angstrom resolution so you rather similar to the bacteria adopts an early map at room temperature so you can see the general structure of the of the protein but not the detailed chemistry for which you need higher resolution and then at a at a lower con to a level still you see the band of lipid and detergent that sounds surrounds this membrane and the protein but V North JPEG Jew who had prepared the protein in Judy hers group went on and improved it until they eventually got a model with all 45 polypeptides identified and largely built although there were some poly alanine giving you a complete atomic model for the structure and in blue in the background are the core subunits that a bacterial homologue of this mitochondrial membrane protein involved in proton transport coupling to oxidative phosphorylation the blue is a bacterial homologue that leo satin offs group had determined earlier by x-ray crystallography so that's a kind of complete summary from staged bacterial Dopson to higher resolution and then switching the the approach if cry we am into the single particle area I just want to finish that with one slide and this is from recent work that in the last year Chris Russo and I have been doing and Chris's is in the audience here and the idea is where are we now so we are now and so what this is this tells you the amount of information in images that you get with the new microscopes and the new detectors and this is the exposure that you're giving it five electrons per square angstrom then 10 and 15 and radiation damage eventually kills the structure you're turning it into a cinder a carbon carbonaceous fragment of what was once a beautiful protein so this is what we what we observe when we analyze the data but what we know we ought to observe if all the remaining problems that exist which are not nearly so profound as the problems in getting gravitational waves to go forward the red is where we are hoping to be and then there are a number of factors that explain we think why we don't do as well as we should in theory and the idea now for the next year or two is to try to address all of these problems but the main one being in this blue area is dementia specimen motion from the radiation damage or from releasing stresses that are is a result of the freezing in the first place and then when we do that we think you'll be able to do all the things I showed you lots more more than a thousand structures per year and so on and it will be less work and more high-resolution as on so with that I wanted to finish by saying that of course quite a number of people have been involved and I've worked with them personally but the main one at the beginning which which basically converted me from an x-ray crystallographer to an electron microscopist and I think there was a slight action and reaction because after we had worked together on bacterial officer Nigel also changed still doing electron microscopy but doing it on membrane proteins as well so it was a important branching of our potential career paths for both of us and then we had this 15-year gap where we had to work on a lot of different methods and in Cambridge Joyce Baldwin Tom Seska and David a guard were the ones who were pushing it but Bob Glaser and Ken downing were important to taking pictures but particularly Fitz and semolina and Eric worked very hard for many years and then we had this one week with with Jacques and Zhang Nepal in the early stages that so that was the high resolution group and then recently we've been involved in a lot of the attempts to improve the methodology through detective development and microscope technical developments and these are the people huazi Greg and Xiao Jie who is here from Cambridge and the two people from Rutherford Appleton science technology facility Carson near Oxford who have spent 15 years or so developing one of the three types of detector that you can now buy from the market and then more recently when we switched into single particle work I mentioned most of these in John Rubinstein who were on one of the different type of ATPase rotary motor and Peter Rosenthal who prepared one of the specimens that I showed you before and then the early the most recent work in all of the biological structures like complex one came from V North Kumar who is now just opened a new crime facility in Bangalore in India and then finally in the last year Chris ruse and I have been trying to address and solve the remaining problems for the future and then I should just finish by saying that most of the work obviously the people from other labs were funded by other agencies but most of this was funded by the Medical Research Council which has always had since the time of brutes and Kendrew a clear mandate to do long term support for projects that take longer than typical project state so with that thank you very much [Applause] [Music] [Music] we are composed of 70% water ourselves all the molecules inside are surrounded by water as pointed out by Jacques de Lucia in 1995 he wrote that during the first 50 years after the invention of the electron microscope the most abundant constituent of living things water had been excluded from studies using the electron microscope Jacques develop a preparation method that solve this problem it is based on cooling water so rapidly that the water molecules are trapped in a disordered state which is called glass or liquefied this phenomenon was considered to be impossible and we will soon learn how the impossible became reality but the road was slippery as we will here but on on this ride Jacques has also looked beyond the molecules and unfortunately my French is not that good but from what I understand by reading the texts that rock has presented that he he writes with a warm and open heart of a humanist on topics on opinions on topics covering problems of our society politics and of course science of which we will hear more about in his presentation I now introduce it's a great privilege to introduce dr. Boucher 70% water 100% warm curiosity [Music] [Applause] [Music] [Applause] well in this very special moment I am overwhelmed with gratitude gratitude for my parents for my family my friends my collaborators for the Royal Swedish Academy of Science and the Noble Foundation this can be well understood but I want to address special thanks to those two person it was kallenberg was my what so what is called dr. fat he brought me in to science it taught me to be a scientist and he explained to me that science is something that should happen in a society Sir John Kendrew was the first general director of embl and he gave me this tremendous chance to work on this quite strange project of dealing with water in the electron microscope so why cryo-em because as you said we are made out we are bags of water and bags of water made of billions of cells which are each one bags of water and if you remove the water everything floating in the cellar in the body will will condensate will aggregate and if you are good in arranging the specimen when removing the water you will get fine aggregates if you are not so good you get coarse aggregates but aggregates you get because fish doesn't fly so even a very tough object like this default bacteriophage when you dry them on a supporting film they are in a very bad shape oh so people develop remarkable method to preserve the specimen though the water is removed one eye one certainly one fine method is to dry the specimen in a solution of heavy salt and the more the water go the more the soul concentrated at the end you have only the particle with salt around it heavy salt heavy metal salt which hold the structure more or less correctly another possibility is freeze right you freeze the specimen and instead of reivew removing the water from the liquid you remove it from the solid and it has great advantage but if you look carefully the head of this poor bacteriophage doesn't look very healthy so at that time my hero is still my hero was is Nigel hanwen and we were working in very parallel direction each one trying all kind of trick to have a better preservation of the specimen he had a funny idea he thought instead of drying the specimen in in heavy metal salt heavy metal salt is not healthy he dry them in sugar because sugar is much better for the material and he joined his force as you know with this guy with with fish shark and together they got this remarkable result 99 with this then 90 degrees and 75 what am marking history another person was very important a boost I moved too much yes you have already heard all that but Glaser perhaps you should stay here bug laser during my PhD and so I did a lot of work just following his steps but the the most impressive message I got off from him was this photograph and this is this is these are bacteria quite broken with all kind of sub structure of the bacteria but what is special we were working with this kind of thing with all all our trick for trying to better preserve the specimen in sugar or in any other thing but here he had it in ice in frozen ice we see the marks of the ice but it's beautifully preserved and having seen that I thought oh this is than anything I've seen before and this is the future so two years later I was invited to be group group leader at Yale by John Kendrew with this project how to deal with water in cryo electron microscopes so there is a problem though frozen water is ice and ice is not liquid water and it's as bad for a biological specimen to be in ice than to be without water but we were trying all kind of tricks this is our high technology device and it goes like that we will try many many things but this is one of the things we had a nebulizer coming from the right here high pressure nebulized throwing a stream of micro droplets of water of any kind of suspension and going through this all slit here the slit and here you have a tweezer and at the top of the tweezer you have one of this supporting film this grid that you use a specimen order in the electron microscope and while the stream of droplets is going you let this fall down in the liquid nitrogen and we got like that things like that a nice droplet of ice and this is characteristic ice now we have the MacDowell I can't see him but he is here thought I will do better and it put here and here just in the liquid nitrogen here a very small beaker with very condensed liquid ethane and instead of letting the specimen fall down in liquid nitrogen in let it fall down in liquid ethane and so he went to the microscope look at the object in the electron microscope and called me hey if there is something here and this was what this is what we see and this is Beatrice then this is not crystallized so we didn't knew what this was the first idea is its liquid if they iterate it's Ethan which has been condensed here no it can't be the temperature that temperature Ethan should have been gone should they have evaporated so we thought ok we will let it evaporate and we will see what it is so we turn off the cooling of the specimen holder and we'll let it warm I do not remember how long it last part of an hour warm slowly and when we came at minus 135 certainly but in really teller rally in one second the drop transform into a milky crystal that we recognize it immediately this was Cuba guys you because is a different form of ice quite conventional we had it studied before in detail in our previous experiment and electron diffraction told confirm it it's ice so if you have a bit if you have a vigorous droplet which turn into ice you know that you had bitterest water ha so what do you do with that ah this is Alistair McDowell vitrification man great this was the first haha I remember I said at that moment oh we have something great Juli and the trouble with vitrification with which which water is that vitrification should be impossible what are this mean yes I will not go into the detail I will just mention the founder of cryobiology father basil Riya who work is you is a Catholic priest in a very severe Catholic congregation and he worked for 40 years in trying to produce suspended life this means immobilizing life or in order to obtain typically for breeding typically any make a lot of money with this not for him but for his congregation and and it was working with very good people and those people with him demonstrate that vitrification was impossible at the end for thermodynamic already complicated reason that I am NOT going to explain in detail so I want to speak about this man for another reason he was born in savea's in valise Switzerland and I start school just very close to him in a place which is would be here in the small village just on the other side of the valley here and at present we have still a nice chalet here nice Mountain huh here and for those who like love mountains this is the so called vanilla this is the blush beautiful mountain okay so so our beautification we send it for publication and we got rapidly this answer no you it was reject that you can't spend nature and they were wrong they were all wrong for two reasons the first one if that while they were rejecting our paper they had on the print press the paper by burglar and Meyer who was who was proving that obtaining complete vitrification in pure liquid water and dilute aqueous solution that means they made the discovered the discovery that nitrification is possible and then it was they were doubly wrong because vitrification is very easy that's and with our method it was reproducible rapid and no problem so so what do you do with that first conclusion Mitra SWAT is not simple it's not just liquid water immobilized in the same shape no no it's something else but doesn't matter it's it works so well for cryo-electron microscopy that we don't dare to go into the detail though when we will go into the detail we will certainly understand very important things about water that we yet don't know and perhaps it will be important for biology also ii haha came can be explained like that we work with these grids and on the grids we have a thin thin film and we spread the specimen on the grid but this is difficult because water has high surface tension and what won't what want water is to form a sphere minimize the surface and trying to maximize the surface on a grid is just technically difficult you have to arrange the surface so that too high number just come equally in negative one to Eve and the other it's a complicated process and my collaborator Macario at that time he died unfortunately very badly after they were stupidly after his retirement and this is a very remarkable person he is very stubborn extraordinary cultivated and it has an enormous culture and very stubborn and it was not at all interested in spreading the water water on the grid so he wanted to have no support and he was worried he was working trying to get the specimen on the holes of the grid without support I was quite opposed to that because there is no chance but because I knew the physics but he did he got that and it turns out that it was not so it was quite easy very easy in fact and this is the typical example on whole of the on the whole of the grid which is 20 micrometer broad he get this dance from my sense of micrometer thick layer spreading over the whole grid so the film is 200 times thinner than broad technically in term of energy it's quite strange but it works well later we understood why there aren't interesting energy and thermodynamic reason which explain all that but at that time we were not aware of that so the method and now comes the most difficult part of the Samina well I hope this is how it should go so you have a grid a grid on the tweezer have a beaker with liquid nitrogen you have liquid ethane and dr. o we try again before you came with the blotting paper as you explained in one second to grid the liquid disappear in the oh yes what water disappear goes in the blotting paper and but the last tenth of a micron takes another second to go away and you have time to let the grid fall down and get the sample get me through fight this was 89 1984 this was rapidly published whether the the general director of embl was a new person and he was not leather philipson and he was not very happy with those physicists working with water and but so we worked with his bath virus the adenovirus and when he got this nice photograph he was convinced and we became the best friend in the world and and so was it and you notice but of course you have heard more and you will heard hear much more from drug him in a moment when you have the particle floating in water in all with all orientation you can make 3d reconstruction and so we had this 1986 this model of virus at 35 angstrom resolution 35 action resolution that's that was that and so came the beautiful word blob all orgy blob barrage yes this is blah blah and then a lot of people continue to work we're working before we were also working but those those you understand were very well what are they more important than the other I guess yes but they were playing to your father and so we had 30 years we had ten times better resolution 10 times better resolution this means thousand thousand times smaller volume this means thousand times more density of information and this is not a small smaller achievement and and three point five thirty five to three point five both these are just number but three point five mean seeing atoms and when you see atoms then you are a chemist and and therefore you have from cryo-em all the power of chemistry which means that you see how filaments aggregates together how you the brain becomes senile perhaps because of that you say perhaps you will see how you will find a drug which will prevent them to aggregates perhaps now I come to the onion to my reading now you perhaps we will understand how to do medicine with that there are plenty of people were interested with results there and science will continue it will take time to explore the brain we may understand understand how we think perhaps conscience will emerge this is knowledge without limit its yes but knowledge has its consequence knowledge makes what what our lives are when the year we discovered we we publish the 15 vitrification method I got my first personal computer now now billion 1 billion of people are our city is sitting in front of a computer screen for most part of their day and communication between people is fundamentally change my the father of my grandfather was living in scarcity he was never sure to be able to bring home the minimum required for decent life for his family we are now submerge with access and the world climate is getting robbed and the glacier just above our mountain hut is collapsing we have a problem Shas sarkozy on sneaker wind alarm this is not very new roughly 500 years ago but now it's getting very actual and very urgent well yes and oh sorry the region what can we do it's very urgent what can we do one thing is certain we scientist must come out of our ivory tower and be involved in the Society for which we produce knowledge that knowledge can it can be equally can have equally good or bad concept and we must become more aware and responsible for these this is the reason why more than 20 years ago we introduced in our university a compulsory curriculum biology and society because we want our students to be as good citizen as they are good biologist this is good but this is not enough how can we be as good in using our knowledge for the well-being of all as we are in producing it I don't know you see we are used to speak freely but this part is in some ways too important to be said like that therefore I want to write it - - I wrote it down and it's make it very complicated yes I don't know but we must fight for our most precious good namely we must protect and fight for the common the for knowledge knowledge is our common good and which must make out of it the best for the well-being of all now if a future generation imagine even imagine imagine if it's easy if you try imagine for example we think about else imagine that we take the w-h-o the World Health Organization or the organism of the you know and imagine we and we we give to this organism all the knowledge about health and about medicine about how to help people have and we and we give this organism the right and the power to deal to do the best with this common knowledge of course we will give those who produce the knowledge the right where a war reward for the effort it would not be hard to do [Music] [Music] thank you [Music] [Applause] [Music] [Applause] Thank You versa for leaving us with these important souls electron microscopy pictures of Biola male biological molecules in water have very low contrast in 1975 yucking Frank realized that the challenge in using electron microscopy for studies of unstained biological molecules he wrote there is the assignment of features that are only faintly visible on noisy background he solved the problem then he spent more than 10 years on developing methods to calculate three-dimensional structures from these faint molecular shadows we will hear more about this in his lecture but perhaps what you know less about is that yuck him is also published author of poems and short stories and he is a former president of the Hudson Valley Writers Guild who meets to discuss poetry and fiction and as a colleague of your team recently described your team's contributions he has been exploring the intricacies of the human heart with the same keen insight and perception that he brings to his scientific work we are now looking forward to hear more many more details of this scientific work it is my privilege to introduce in three dimensions that high resolution you work in Frank [Applause] [Music] [Applause] members of the Academy distinguished guests ladies and gentlemen allow me do you speak about science rather than live we open heard it's an enormous honor to be selected as one of this year's Nobel laureates in chemistry I would like to thank the Royal Swedish Academy of Sciences for bestowing the Nobel Prize on me along with my co laureates Richard Henderson and dr. Boucher the thought of being considered a peer of Ernest Rutherford Marie Curie and Linus Pauling is truly daunting I developed an interest in electron microscopy and when I worked with against Hinda on loops I developed an interest in electron microscopy now who worked with Ernst kinder on my master thesis and physics at the University of Munich my thesis had something to do with Beck scattering of electrons in 1943 working with the electron microscope he had studied butterfly wings which as he realized gain their brilliant colors from interference of light on gratings formed by tiny scales arranged in a regular order I signed on II to a graduate project with Walter hopper at the Max Planck Institute in Munich an x-ray crystallographer who had developed an interest in electron microscopy as a means to study biomolecules he viewed the electron microscope as just the diffractometer that unlike the one employed an x-ray crystallography could record not just amplitudes of diffracted electrons but their faces as well this was just a fancy way of saying electron microscopes were able to form images for the beginning years from 1930s to the 1950s the contributions of from microscopy to biology were confined mainly to the investigation of tissue serious forays into a quantitative visualization of molecular structure did not start until the 1960s and were concentrated in three groups aaron klug at the lap of molecular biology the MRC and Cambridge my mentor Walter hopper at the Max Planck Institute in Munich and Edward kallenberger as you heard at the beer central in Basel now unless symmetries are present the three-dimensional reconstruction of an object requires the combination of its projections from a wide angular range first pioneering achievements in molecular structure in molecular structure research with the electron microscope with a three-dimensional reconstruction of the barrack of bacteriophage tail with helical symmetry in 1968 by the ricean Kruk and the first reconstruction of an icosahedral virus in 1970 by tony Crowder in the same lab at that time biological molecules could not be imaged in a close to native state negative staining which amounts to embedding the molecule in the layer of heavy metal salt was the only means available to produce contrast on the other hand biological molecules are quite fragile and it was known that maintenance of the integrity would require a fully hydrated environment when I started my work as a graduate student of under the water hopper I was exposed to discussions in small workshops in the thurible Alps co-organized by Walter hopper and Max Perutz in Alba and once in here check these were the first meetings that brought together protein and crystallographers and people working in electron microscopy the many problems faced in attempting to image biological molecules the electron microscope where discussed at a workshop in he organized by edward carnby and guys in the swiss alps in 1973 the state of the art at the time was reflected in the title of a proceedings paper since high resolution was equated with any results better than 30 angstroms her amount at the meeting was the search for a method that would keep the molecule fully hydrated while exposed to the electron beam following pioneering studies by Pope Glazer in 1971 just at the time I visited his lab as a Harkness fellow radiation damage was recognized as a major obstacle in the strife toward higher resolution averaging of a large number of repeats of a structure exposed to very low dose was seen as a general solution to this problem thus this meeting set the stage for a groundbreaking study by Richard Henderson a nodule on wind that you've already heard about in 1975 the reconstruction of parabolic period opsin from the purple meander a mean of halibut are embedded in glucose under quasi native conditions the confluence of novel approaches to sample preparation data collection at extremely low electron dose and merging a tilt series of these noisy images made this work a towering achievement the time after that was marked by general excitement in the community and many attempts to use the same or similar methods in a study of other proteins amenable to a two-dimensional crystallization however the intrinsically low contrast between proteins and glucose made these attempts difficult and embedment and eyes remained the agreed general goal at the time most researchers in the field that was to be called structural biology were united in the belief that serious structure determination required highly ordered samples such as 2d crystals here little fibers or viruses with high symmetry attempts to extract structural information from freestanding single a symmetric molecules were not taken seriously one such alternative approach pursued by Walter hopper he was to tilt the am grid - which isolated molecules were attached in the multiple angles while collecting projections from which the molecules could be reconstructed while the angles are exactly known in Hoppus tomographic approached the accumulation of radiation damage over multiple exposures to more than a thousand electrons per angstrom square at the time rendered the end result essentially meaningless the other approach was the one I formulated in a concept paper in 1975 which I will elaborate in the following so this approach takes advantage of the fact that in biological molecules purified from cell extract and suspended in solution typically exist in thousands or even millions of copies that is separate realizations of the same structure hence if agreed to which such a sample has been applied it put into the electron microscope a large number of projection image of the same molecules lying in random orientations are encountered the important point is that instead of collecting multiple images from one molecule by tilting it in the electron microscope one merely has to take a snapshot of multiple copies of that molecule with a very small radiation dose the advantage in the latter case is of course that the 3d image obtained relates to a molecule that has seen only one single low exposure and has remained practically and damaged the term determines single particle averaging and single party reconstruction which were later coined refer to the fact that the molecules in the sample are freestanding and not attached to one another as in the crystal so at this point it is important to reiterate that no symmetries are assumed in this approach rather the idea was that entirely a symmetric molecules could be reconstructed in this way symmetries obviously simplify the problem as in that case the image contains multiple projections of the repeating unit in different known orientations allowing data from different molecules to be readily merged now simple as the concept sounds the realization of simple particular construction required several problems to be solved all in the area of image processing first is it feasible at all to align nauseum molecule images with one another with sufficient accuracy second how do we obtain how do we estimate the resolution of a reconstruction obtained by combining all images third how do we sort molecule images by appearance related to view angle and confirmation fourth how do we find the viewing angles of projections our posterior fifth how do we reconstruct molecules peeler molecule from projections at randomly spaced new angles so it was in the nature of these problems and their novelty that progress was slow and piecewise one step at a time during the work on my PhD thesis which was 1968 through 1971 explored the use of cross correlation function for aligning electron micrographs I discovered there two successive images of the same area of carbon could be aligned with a position of better than 3 angstroms actually the process of 2d alignment is a bit more complicated since molecules picked from the micrograph differ both in shift and rotation the solution I found to determine both relative rotation angles and shifts simultaneously makes use of the fact that the autocorrelation function of a molecule is a shift invariant it does not depend on its position within the image frame then working with Bourne Sexton at the old Cavendish lab in the group headed by Vernon crosslet I was able to show that alignment by a cross correlation would be accurate enough for the purpose of aligning noisy images of single molecules if the dose exceeded a certain level which depended on the molecule size and contrast against the background using the ribosome as an example it became clear from the formula we obtained that single particle approach to structural research was indeed feasible for molecules of sufficient size my appointment in 1975 a senior research scientist at the division of labs and research later named Portsmouth Sunday and Albany offered me the opportunity to explore the idea with the help of samples provided by collaborators I was able to demonstrate the feasibility of obtaining two-dimensional averages showing enhanced features of molecules with images of glutamine synthetase as a dual colony receptor and 40s ribosomal subunits from HeLa cells among these the 40s subunit averages were most striking in showing the potential of the single particle technique and proof instrumental for gaining funding by the National Institute of Health nonetheless presentations of single particle averaging results at the meeting it were given in 1975 1709 by myself and my collaborators martin castle and peter zingsheim were greeted with a great deal of skepticism one issue to be addressed which I had already mentioned before was the fact that due to the absence of crystal order the average of aligned molecule images shows no diffraction spots in its Fourier transform in the air for lexan inherent measure of resolution without such measure progress and quality cannot be tracked and compared among different groups following an earlier study during my dissertation on the effects of drift on electron micrograph I realized that signal bandwidth is reflected by the extent of reproducible information in Fourier space this extent of reproducible information is apparent from Young's fringes that show up in the optical diffraction pattern went to a successive micrograph of the same specimen field are superposed with a slight shift applied how can this idea be translated into computation computational procedure reproducibility can be quantified by dividing the data randomly in half then comparing the Fourier transforms of half averages over rings in Fourier space resolution is then defined as the ring radius we are a measure in where measure of comparison for instance the phase residual or R factor or cross correlation passes a critical threshold the same measures computed over shells would later prove important an estimated resolution of 3d reconstructions as well these first studies of image averaging immediately brought up the problem of heterogeneity only those mala to molecule images could be combined in an average if they are originated from molecules of identical structure and showed the same you at that time one of Aaron's from Brooklyn students Marine Fern Hill visited my lab with images of Limulus Polyphemus hemocyanin which is an oligomer with the distinct architecture producing multiple prefer to use when negatively stained an image in the electron microscope these images therefore presented a perfect example of heterogeneity the solution to this problem came from the inside that images once aligned with one another may be regarded as vectors in a space of n dimensions n is the number of pixels groups of images that are similar will show up as clusters and of vectors in that high dimensional space equivalent problems of finding clusters in high dimensional space were encountered in many fields of science and gave rise to multivariate statistical analysis a procedure which determines a compact low dimensional subspace tailored to the problem with the help of jean-pierre Britta J Watts was sent a scientist working in Laboratory Medicine we were able to use a program meant to sort blood samples to sort images instead application to hemocyanin proved the great success early on as a set out on this new approach to recovering structure it became also clear to me that in order to make systematic progress in the development of algorithms and computer programs with ever-changing in expanding goals required a workbench with a large set of tools to this end I developed a modular image processing system called spider which made it possible to design complex programs from pre crowded building blocks using a simple script language for example the command WI would invoke a routine for extracting a length a portion of an image and AC would compute the autocorrelation of an image all programs were coded in Fortran the most advanced language at the time most of the initial programming was done by a student of computer science and one undergraduate the script language of spider then became literally the lingua franca in my lab and as the suit was disseminated to other labs within a growing community of users I attribute the idea underlying a spider system and it's modern and it's modular design to my stay at the Jet Propulsion Laboratory in 1970 under the Hartness fellowship during this one year visit I learned the right program modules that interfaced with Vic are the image processing system in use by NASA since 1966 for analyzing images from planetary missions such as the unmanned Jupiter flyby mission now to the subject of three-dimensional construction for computing the three-dimensional structure of an object from its projections one requires the angles of each projection to be known determination of the angles of the randomly oriented molecules was the most important yet most difficult problem to resolve the solution came from the inside at two micrographs one of a field of untilled particles one of the same field he tilted contained all the information required to assign Eulerian angles to each tilted particle in this geometry the angles of the tilted particle loop projections lie on a cone with a random azimuth giving rise later to the term random conical reconstruction in 1982 I was joined by Michael Rademacher also the student of Walter hopper who had worked in his dissertation on algorithms for 3d reconstruction from projections arranged in a regular chronicled geometry thus he had exactly the skills needed to develop computer programs that implemented the random Chronicle reconstruction one important step was still missing the generalization of the algorithm that assumed irregularly spaced conical tilting to the general case of random angles once this was accomplished by Radha Maha in 1986 we obtained the first single party reconstruction using the random Chronicle method the 50s subunit of the equal eras ohm this reconstruction was limited in quality by two factors one was the missing cone of information in the three dimensional Fourier transform a source of unidirectional artifacts in the three-dimensional density map and the other way the artifact stood at the preparation of the sample by air drying with negative stain both were ethically overcome at the time within a short period the missing cone problem was solved by merging datasets obtained with three or more different 0 degree views and the preparation of the sample with negative staining was replaced by cryo embedding in vitreous eyes following the spectacular success of jerk du Burgess verification method in the application to viruses in 1984 yet another important problem affecting the quality of all recruiter constructions from am data is the distortion of information by contrast transfer function or a CTF caused by the lens operations of the electron microscope this problem and its effect on the reconstructive density maps was eventually overcome by my group and by others through merging data obtained with different D focus settings using a wiener filter algorithm the first molecules visualized by a-crying m and reconstructed in three dimensions with a single particle method where the e coli ribosome hemocyanin and calcium release channel a cranium reconstruction of the hollow ocular Mauresmo TOA ribosome proved to be helpful in the facing of the first x-ray structure of the large subunit all final reconstructions benefit from an iterative in angular refinement that starts out with the first rough reconstruction in any project dealing with a molecule whose structures unknown it is practical to distinguish between two phases the bootstrap face and a refinement phase and bootstrap phase a first rough reconstruction is obtained either by the random chronicle method or by an alternative method developed by marathon Hill and others in which common lines in Fourier space are employed in the refinement Phase II an existing reconstruction is used to generate a library of even spaced projections with which the experimental projections are compared to assign refined angles to them for the next round of reconstruction now I would like to show a few examples for how the new technology well before it's reached the present state of perfection gave us important insights into pivotal processes of translation by the ribosome in the late nineties one of my postdocs rajendra arguable prepared a sample containing allegation fact G bound to the ribosome as it catalyzes translocation of mRNA and tRNAs comparison of the crying M reconstruction with that of the unbound ribosome showed a dramatic change the small subunit head Road headed by almost 10 degrees this finding provided first clues on the mechanism of Emma and a tRNA translocation in this important molecular machine in 2002 another postdoc Michael voila I'm sorry in 2002 another post of Michael valor found that in the decoding process tRNA entered the ribosome in complex with protein in fact the ef-tu in a strongly distorted form in the so called 18 state this observation indicated that the tRNA acts as a molecular spring apparently setting the threshold for discrimination between cognate and near cognate codon and decode on pairing as a third example in 2001 we teamed up with jennifer doudna then at yale to visualize the mRNA from hepatitis C virus in the process of hijacking the human ribosome my postdoc Christian span working alongside with Jeff chief from the Doudna team visualized the so called iris element of the virus attacking the small subunit of the ribosome even though before 2013 when the camera the cameras came out in a crime reconstruction fell short of resolutions in the 3d angstrom range it was still possible to interpret them the basis of existing structures by means of flexible fittings such as mdff deriving interpretation through atomic models I'm showing two examples among many a six point I on strim cry a map presenting a snapshot of the decoding process and its interpretation by Elizabeth Villa and the atomic model of the ribosome from t Brucie built from 5.5 angstrom map by Yasser Hashem still from data recorded on film I finally come to a distinct advantage of single particle I am that is brought out in many applications now here unlike in x-ray crystallography molecules are unconstrained by crystal packing and are able to assume the full range of confirmations present present in solution for a process of molecular machine like the ribosome unless it is stopped by a chemical intervention many different states coexist in solution the full potential of single particle IEM has been realized for the introduction of maximum-likelihood classification by co-chairs who work as a student in the lab of Jose Maria Caruso and later independently the LMB MRC and Cambridge these algorithms perform the seemingly impossible task of disentangling changes of view angles from changes in structure a large well characterized ribosome data said that we as supplied was correctly classified in the initial study working with his to it set multiple structures are now routinely recovered from single sample in the next slide in the next slide shown I show two examples several state of the translation a negation cycle are fished out from images of ribosome purified from a cell extract of Plasmodium falciparum the malaria parasite the second example on the right is another study by my group where part of the elongation cycle is visualized in an example containing a mutant of a negation factor G literally where cryo-em is now able to tell us a story from a single sample showing how the molecule changes its shape and binds smaller molecules as it performs its work ultimately even the recovery of a continuum of structures reflecting the states of a molecule at work and the mapping of the free energy landscape are no longer a distant goal even though I have illustrated the progress achieved by using my favorite molecule the ribosome which now has the best resolution of 2.5 angstrom the range of applications in biologists is virtually unlimited except for some lower bound on molecule size and of course the necessity of having the molecule suspended in solution my own recent collaborations with the groups with the groups of and remarks Wayne Hendrickson Alexander Sokolov Sookie and Philippa Marcia demonstrate the gain and knowledge achievable now for membrane bound channels and receptors to conclude this account of 40 year journey I must confess that even though I was always a firm believer and the technique I dreamt up many years ago I never thought he would see it and come to a fruition in this spectacular way let alone that I would earn that it would earn me a share of the highest prize coveted by scientists all over the world it has been a truly fantastic journey of all the knowledge that human medicine as a strength the benefit from the new technology is ultimately the best reward I would at this point I would should at this point you give tribute to old people who have made this development possible over the years it goes without saying that without the pivotal contributions by Mirko laureates Jack to Boucher and Richard Henderson the field would not have developed and prospered in this way my things go to many students postdocs and collaborators who shared my vision and put it foot in so much hard work to make it all happen but grass uchi has been with my group for more than 30 years in management and support of instruments with ever increasing complexity my very special thanks goes to my wife Carol Saginaw who has steadily supported me for close to 40 years and never lost faith in me in my family and friends who have cheered me on at every turn what are the early developments of the technique took place at the Watts Watson day in Albany which provided a sheltered supportive environment with generous funding by the inner state of New York particularly in the starting phase in my expression of cattitude to the Watts words and I would like to single out Carmen Mandela who supported me throughout those years as a colleague collaborator and friend my affiliation with the department of biomedical science at the University of Albany since 85 allowed me to foster academic interactions with SUNY faculty at large and since 2008 when I joined the faculty of Columbia department of biomedical sciences and biological sciences I enjoyed multifaceted intellectual environment foster and maintained per one of the count country's greatest universities my workmates been generously supported by the National Institute of General Medical Sciences of NIH almost without interruption since 1981 funding for two inspiring sabbatical states in England and Germany was provided respectively by the four got a foundation at the humored Foundation I received sizeable awards from the National Science Foundation for several instrument purchases and support by the Howard Hughes Medical Institute for almost twenty years has allowed me to build up a state-of-the-art professionally supported cry um facility firstly the Watts word Center and later at Columbia University thank you [Applause] [Music] could I please ask colore it's thank you [Applause] [Music] [Applause] [Music] [Applause] you

Channel: Nobel Prize

Views: 10,905

Rating: 4.9370079 out of 5

Keywords: nobel prize, nobel prize in chemistry, Jacques Dubochet, Joachim Frank, Richard Henderson, cryoelectron microscopy

Id: p98JYaHaDC4

Length: 96min 5sec (5765 seconds)

Published: Fri Dec 08 2017