Robert Huber

Beauty and Usefulness of the Building Blocks of Life: The Architecture of Proteins

Category: Lectures

Date: 1 July 2008

Duration: 28 min

Quality: SD

Subtitles: EN

Robert Huber (2008) - Beauty and Usefulness of the Building Blocks of Life: The Architecture of Proteins

Atomic views of protein structures are determined with increasing pace in the last twenty years by a rapid development of methods and instrumentation of protein crystallography, electron microscopy, and nuclear magnetic resonance, allowing the determination of very large and complex protein structures

Good morning. As a chemist in the Lindau physics meeting I decided to speak about basic properties of proteins. Of course, with some apologies to the chemists in the audience. So my focus is different from Hans Deisenhofer's talk but there is a useful overlap and I'm grateful for the arrangement of these 2 lectures in sequence because we both study proteins and mainly use x-ray crystallography. Well the title of my lecture which you see in the program maybe seem problematic because fitness for purpose is objective but beauty lies in the eyes of the beholder and the early protein models which we construct in the late '60's, early 70's, made from wire and screws are barely regarded, can barely be regarded as beautiful. But today we use computer graphics and tricks offered by this technology. But we must be aware of the fact that these models are metaphors yet still they are useful to derive molecular properties and plan new experiments. And if you want then be pleased by their beauty. Now why do we study proteins. Now proteins are the product of a complex series of transcription and translation with enormous increase in complexity. The genome is simple, the proteome complex. Yet the proteome decides about life phenomena. A beautiful illustration of that you see here, the mature butterfly and its larva share the same genome but they are as different as you can imagine. So what are proteins. A few basic facts about their chemistry. They have a defined amino acid sequence. There are 20 natural amino acids of different stereochemistry and electrostatic properties. They are linked by amide peptide linkages. Now despite their construction from often ten thousands of atoms, they have defined structures. They are synthesised as unfolded structure-less chains and then in a sequence of hierarchical events of structure formation they form via secondary structures, tertiary structures, the final quarternary structure which is the functional protein. We now talk about models, then it is quite obvious that the structures became transparent only after very substantial simplification, this is an all atom model of a large protein with about 70,000 atoms, totally in-transparent. So we have to simplify it by just drawing out the polypeptide chain indicating the secondary structures, the helixes, the beta strands. So we can have surface representations to explore the binding potential of the protein and this is the most simplified representation of this 28 sub unit aggregate. Just indicating the arrangement of the sub unit, the architecture. Now we use metaphors as I said already to describe protein structures, so these are propeller structures, there are 6 bladed or 5 bladed propellers. There are barrels with a remarkable similarity to the Castel del Monte in Apulia which share both the molecule and the architect shares the 8 fold rotational symmetry. Strict of course in the case of the human construction and somewhat distorted in the real molecule. Now how do we know, Hans mentioned already there are 3 major techniques, crystallography, the working horse of structural biology. There is electron microscopy which is able to provide views usually at low resolution of very large molecular aggregates. Now what we often do now is disassemble these aggregates, separate the components, crystallise them, look at them at high resolution and then model back the large complex structure, the atomic resolution by fitting the atomic structures into the envelop given by electron microscopy and that is NMR which has the advantage of analysing protein structure resolution, all be it usually of small or medium sized proteins. Now Hans mentioned the major breakthroughs in structural biology but of course it began, crystallography began earlier with Rankin, with Van Laue, with Bragg, with Perutz, all of them Nobel laureates as you know. Now let me just briefly show these, illustrate these major break throughs which led to modern structural biology, crystallography. This is a museum Rankin's x-ray generator. And of course our modern in house machines look different but they use the same physical principle and generate relatively weak x-rays. The breakthrough came with the discovery of synchrotron radiation for use by crystallography, you know the principle of these x-rays are several orders of magnitude stronger than the in house apparatus. And they have different properties, so you can select variant and select wavelengths you are working with. We use recombinant proteins, so molecular biology essentially contributed to structural biology by making recombinant proteins in bacteria and in transformed higher cells and we use crystallisation robots. This is a real time representation of an experiment with a protein crystal on a strong synchrotron x-ray source. So we are able to collect x-rays of complete data set of a protein crystal within a few minutes. It took us when we started protein crystal, when I started protein crystallography in the late '60's, early '70's, months or even a year, now we have a data set in a few minutes. Let me show that again. So we rotate the crystal in front of this x-ray beam and have very fast and efficient detectors to evaluate within fractions of a second these diffraction patterns. We have fast computer algorithms, fast computers to store the processed data. And last, not least the development of graphic system in the middle '70's. You saw the first metal wire model, now we use computer graphics and even can make an automatic interpretation if the electron density is very good. This is clearly a tyrosyl residue side chain. So we know usually the sequence and then let the computer and the algorithm fit the modelling the electron density. Now back to proteins and how are they synthesised. So there is messenger RNA transcribed from the DNA, transferred into the cytosol and then bound at the ribosome which is the machinery which assembles the amino acids to the polypeptide chain. The protein folds up but if damaged or incorrectly synthesised has to be removed and there are proteases around, which degrade this incorrectly folded proteins down to amino acids which are then reused for synthesis. This is a great EM picture which I found in literature, unfortunately I forgot the reference, with ribosome's sitting on the messenger RNA and synthesising the polypeptide chain. So the analogy, the similarity to an assembly line of a car factory is really astonishing. Now this team of workers of course has to move, they have to assemble the parts and also proteins have to move. Now this is a protein as an example from a degradation machine, not a synthesis machine and you see how extensive these movements may be of a protein. So this protein is made up from domains joined by linkers which allow extensive movement, how can we do that. Now what we do is we stabilise certain conformations of the proteins and make snap shots and then make a movie. Now new born proteins are sometimes rather sensitive and they need protection. They are not yet completely folded, they are sticky and they need protection. And for that purpose there are so called Chaperonin and Chaperonines. These Chaperonines are big protein containers which in their interior take up this freshly born protein so that it has some time to find its proper fold. This is a eukaryotic version, this is a bacterial version, they have similarity in the sub unit structure, if you compare this with this, but the architecture is rather different, this is made up of 16 sub units, now this is made up of 14 sub units. So nature plays around with symmetry properties. Now proteins that are not properly folded and cannot be rescued in these chaperonines have to be removed and for that purpose we have intercellular proteases. This is one example, it's a big protein, a hexameric protein, the sub units of size 1,200 amino acids, about and it acts like a bread slicer. So we can make a movie again by taking snap shots of certain stages of this degradation process carried out by this protease and we see that an unfolded peptide then moves into the centre of the protein to the scissors, it has to be completely unfolded because this is just a narrow tunnel when dipeptides are cleaved off and leave the protein through the other side. When we learn from protein structures about evolution, you are quite familiar with the fact that we use skeletons to compare species. This is a quite funny example, I took a photograph in Monterey in Mexico in the museum where they showed the skeleton of a dead 200 times enlarged, really quite similar to a human skeleton. Now using protein structures we can look much further back in time. This is an oxygen storage protein from an insect and this from a sperm whale and they obviously have the same structure. So this again is protein structure from the early times, 1968 or so, where we cut electron densities from woods and glued them together but it's quite clear that these two molecules are closely related. Another example, protein degradation again an essential process in bacteria and in eukaryotes, this is the bacterial version, the human or the eukaryotic version, except with the same architecture but more simple chemistry consisting of just 2 different sub units while there are 14 different sub units in the eukaryotic version. Clearly again showing at the molecular level evolution. Well time rushes and I would like, I think just a few words about basic physical processes about which we learn from protein structures. This is photosynthesis. The only biological process that we can observe from space when we look down on earth in winter and in summer, this is the colour of chlorophyll, a sign of intense photosynthesis going on. Now as in a technical process, you have to collect, in photosynthesis one has to collect the light and then convert light into an electrical current. So you have no lenses or mirrors in biology, you have proteins and cofactors. This is one example of a quite efficient light collecting element and sign of bacteria, these are called phycobilisomes, huge organils sitting on the photosynthetic membrane, close to the photocell, the photosys, the 2 erection centre. This is derived from an EM photograph at low resolution. So we can dissect, isolate these components, crystallise them and look at their crystal structures. We can calculate spectral properties from the arrangement of the protein and the chromophores. And also the energy, the excitation energy transfer properties from here to here but I think that you immediately see is that we have 15 chromophores in the outer components, Now this is the reaction centre, the bacterial reaction centre, here in comparison with a sino-bacterial, a plant-like photosystem too. Again showing evolution because the heart of the reaction centre, this is where charge separation occurs, is identical in both, in bacteria and in plants. I would like to bring forward one example of an interesting chemistry going on in biology, now unusual chemistry, namely a chemistry on carbon monoxide. Now you know carbon monoxide is a poison for higher organisms because it binds to heme groups, to cytochrome. Now bacteria can use carbon monoxide for their energy needs and they use carbon monoxide as a carbon source. There are many sources of carbon monoxide in technology and by natural processes. But these bacteria are able to make use of them. Now the process that is carried out in this bacteria is of enormous importance in technology, it's the so called water gas shift reaction, the major source of technical hydrogen. So it needs big plants, high pressure, high temperature, it needs metal catalysts to carry out the reaction. Now bacteria do it at room temperature and ambient pressure. You find one kind of these bacteria in crater lakes, they live under anaerobic conditions, have a big protein to carry out the reaction and the protein alone can't do it, you need metal cofactors. This is a rather complex metal cofactor consisting of Ni-4Fe-5S. Nature makes this with ease, spontaneously. Now inorganic biochemists were yet unable to synthesise this. This is an anaerobic species carrying out the same reaction, totally different protein, you find this bacteria in charcoal piles, they use quite exotic copper, sulphur, moliptinum cluster for this reaction, you see here at high resolution. Well at the end a few words about application of protein structures. So far I spoke about the basic science for an understanding of chemistry and physics, of life functions but protein structures also allow rational design and development of drugs for medicine and of new compounds for plant protection. So of course begins with a diagnoses, now obviously there is a problem with blood coagulation with this patient. We know the molecular components involved in the coagulation cascade very well. They have been analysed by (inaudible 21.07) sequence function. Now at the end of this cascade there is a component called thrombin, now thrombin cleaves fibrinogen and the resulting product, fibrin polymerises and forms the meshwork, matrix for the thrombus. So if we want to interfere with this process, like we would like to do with this patient, we may inhibit thrombin and this was actually the first structure in this system of components that we had been determining. Now how do we do this ligand planning, this design, we follow 100 year old hypothesis of Emil Fischer, the Schlüssel Schloss principle. So this is the empty enzyme, this is the designed ligand and this is the ligand protein structure. And this is the real example, this was a designed, the first designed lead structure for thrombin. It fits quite nicely but not perfectly and then was developed further to fit these individual pockets perfectly with an improvement in binding constants of several orders of magnitude. This is another example. Highlighting different problems in drug design and development. Now this furin is an enzyme which is essential for activating hormones from their pro forms by cleaving off n-terminal segments. But it is misused by toxic, lethal toxic bacteria and viruses which also need furin for processing their toxins or their code proteins. So is there a therapeutic window where we can kill viruses or the bacteria without killing the patient. Now what we can do is design a ligand to the active cite of the enzyme inhibiting it. And then in fact there are already mouse experiments with the simple compound that protected mice from pseudomonas exotoxin which otherwise die within 2 days. Well time is running out and I skip this third example of using structures for understanding a genetic disease variegate porphyria causing skin lesions due to mutations in an enzyme protoporphyrinogenoxidase involved in heme and in plants and chlorophyll biosynthesis. So in plant that same enzyme is a target for an important class of herbicides. So structures help us to understand the cause of this disease, we can't help, but here we can help by helping to design better herbicides and this is actually quite important because there is the appearance of a resistance to this class of herbicides in an agronomically important wheat in the United States, there is a mutation in the structure. We do understand that the herbicides no longer can bind to these mutant enzymes while it is still enzymatically active but as we know the mutation, we can model the structure and design variant or different, changes in these herbicides. Now basic research of protein structures in function and structure based rational design, development of ligands of target proteins, has found its way to application and even led to the foundation of medium sized companies with intensive research activities, I can't resist to show that, this is a spin off from my department 10 years ago which flourishes. And what they do is offering in a very professional way the techniques which I described. So protein expression from genes, protein structure determination and then protein ligand structure determination for which the customers pay. So this is what they offer, protein production, structure analysis and protein ligand structure information. Now back to the beginnings of protein crystallography which was of course driven by scientific curiosity, how life functions at the molecular level and protein crystallography was founded and developed as a science at the interface of chemistry, physics and biology. So it was purely basic science at the beginning and it is still an essential technique for understanding biological phenomena. Protein structures have become an important aid now in the development of new strategies in medicine, for developing novel drugs and for efficient plant protection. Clearly a mature field within not more than perhaps 50 years of development. Thank you.


Atomic views of protein structures are determined with increasing pace in the last twenty years by a rapid development of methods and instrumentation of protein crystallography, electron microscopy, and nuclear magnetic resonance, allowing the determination of very large and complex protein structures. These structures document the beauty and unlimited versatility of the proteins´ architecture, but reveal also unexpected relationships, allowing views on biological evolution far back in time. The structures are a basis for understanding the proteins’ binding specificities and catalytic properties (chemistry), their spectral and electron transfer properties (physics), and their roles in physiological systems (biology and medicine). They allow design and development of specific ligands of target proteins opening novel strategies for therapeutic intervention and development of new medicaments and for plant protection.