Interview - Magnetic Resonance

NMR Methods to Study RNA Protein Complexes

With NMR allowing you to see only certain NMR-active nuclei, you can basically highlight those regions that you’ve labeled with this NMR-active nuclei and look at those parts of the molecule in the context of the whole molecular machine, at atomic resolution, but without having the complexity of all the rest.

This is the first installment of a two-part interview with Prof. Dr. Teresa Carlomagno, of Leibniz University of Hannover and the Helmholtz Centre for Infection Research in Braunschweig. Here she discusses her work in structural chemistry and developing NMR methods to study RNA protein complexes.

Please can you give an introduction to your research?

Since the early 2000s, when I started my group at the Max Planck Institute, we have been following two main lines of research. We study RNA protein complexes, the ribonucleoprotein (RNP) complexes, especially those involved in regulatory processes, such as regulatory RNAs, and RNA metabolism. We develop methods that use NMR, but also combine it with other techniques in solution, in order to access the structure of very large RNP complexes, the so called molecular machines.

This had always been a dream of mine since I moved to the lab of James Williamson as a post-doc. I immediately thought that NMR is the perfect tool for zooming in on particular regions of the molecular machine and look in great detail at those regions that are important.

I also thought this would be possible at high molecular weights, for which a complete structure determination is not possible. I always had this idea in the back of my mind that combining NMR and other techniques would make it possible to get both the structure and the functionalities of these big molecular machines. This is what I started doing in Goettingen. This is also what we have been doing all these years and what we are still doing now.

The second line of research deals with drug design. We are developing methods to facilitate structural-based drug design. Also, here, our methodology tackles those systems that are not easily studied by crystallography or by other more classical methods. It can tackle big systems, so proteins or molecular machines,  that do not crystallize. Our methodology gives us information about how drug leads bind to these big molecular machines, without needing to solve the structure of the whole molecular machine or of the whole complex with the small molecules.

The mission of my group is to try and really push the methodology to its limits in terms of the molecular weight and the complexity of the system that it can look at.

Our philosophy is to combine different methods in solution; to combine NMR with other methodologies like small-angle scattering, EPR, or FRET, and to really have a pool of complementary structural information that we can then combine with clever structure calculation protocols to get information about the structure and the structural changes of molecules while they function.

What are the main advantages of using NMR methods to study the structures of proteins, nucleic acids and small molecules?

With other technologies such as electron microscopy or X-ray crystallography, you either get it all or you don’t get it. In crystallography, you need a crystal to get a structure. For that, you need the protein to crystallize and to diffract at a decent resolution. Electron microscopy has seen huge progresses in the past couple of years and is now able to reach a resolution of up to 3.5 ångström or even better. Also, you can get a complete structure of the molecule, but it’s difficult to zoom in and see what specific parts of your complex are doing, so the active sites or side chains of specific sites of the molecule.

With NMR, this is very easy. You selectively label one part of your molecule. There are very clever ways to do that, both for RNAs and for proteins, including segmental labeling of only one part of your sequence, amino acid or nucleotide-specific labeling, or even atom-specific labeling of different parts.

With NMR allowing you to see only certain NMR-active nuclei, you can basically highlight those regions that you’ve labeled with this NMR-active nuclei and look at those parts of the molecule in the context of the whole molecular machine, at atomic resolution, but without having the complexity of all the rest.

You can basically look at how an active site changes conformation during catalysis or the mechanism of how catalysis is carried out by an enzyme, by just zooming in on the interesting part and without worrying about what the rest is doing and all the complication of that. You can also study the dynamics of these functionally interesting parts very effectively.

In what ways has NMR allowed you to elucidate the structural basis for the function of RNP complexes with catalytic activity in RNA metabolism and gene regulation?

The complex we were interested in methylates ribosomal RNA (rRNA). rRNA is modified at different positions; actually, RNA, in general, is modified at different positions. This is used by the cells to increase the chemical complexity of relatively simple molecules and modulate the function of the RNA in different ways.

One of the most widespread of these chemical modifications is methylation at the 2’-O position of the ribose and the cell goes through a lot of trouble to achieve this modification.

It puts in place a very complicated machinery that comprises both RNA and proteins. The machinery consists of a guide RNA that recognizes the rRNA sequences that need to be methylated. Then, several proteins come into play. Some are structural proteins and so serve the purpose of keeping the structure of the molecular machine, and some are the enzymes that really do the catalysis.

Now, each guide RNA recognizes two rRNA sites. If there were a hundred RNA sequences that needed to be methylated, there would be 50 guide RNAs, each of those pairs with two different sequences, and methylates them.

The question that was always in our minds was why these two different rRNA sites coupled. We asked: why are they recognized by the same guide RNA and is there any functional significance behind this coupling?

We thought that if there is any functional significance behind this coupling and if the two rRNA sites that are methylated have to be controlling each other, then, in the structure, we should see some sort of asymmetry between the two sites.

We thought the two sites could not be structurally exactly the same; there needed to be a difference in terms of one side controlling the methylation of the other side or one side being more easily methylated than the other side.

Now, since we started this project, two structures were published, one by electron microscopy and one by X-ray crystallography. The structures were different. In the electron microscopy structure the RNA was not visible; the crystal structure was absolutely symmetric when it came to comparing the two rRNA binding sites. This always puzzled us. We could not understand how it could be that the two sites somehow talk to each other if they look exactly the same and could be reached by the protein that actually does the methylation, in exactly the same way and at exactly the same time.

We were not put off by the fact that these two structures came out while we were doing this work. We continued working on it. Using a combination of NMR spectroscopy and small-angle neutron scattering (both techniques were very key to our structural investigation) we actually arrived to a structure that differed from both the electron microscopy and the X-ray crystallography structure.

The complex is 400 kilodalton in size and we are not able to do a complete de novo structure determination of all single proteins in the complex at this molecular size. We had to rely on building blocks that we know the structure of, so, basically, the sub-parts of the complex for which we have high-resolution structures.

Then, we collect a series of structural restraints in solution that put together all these different pieces. We integrate this NMR information with the information from the small-angle neutron scattering and put everything together in a structure calculation protocol that then gives us the structure of the full complex. The high-resolution structures of the pieces of the complex, which could be solved either by NMR or by X-ray crystallography actually helped us a lot, both in interpreting our NMR results and in building up our structure calculation protocol, which uses these high-resolution building blocks. As always in science, we build on the work of others.

Our structure shows, for the first time, that the two rRNA recognition sites are not in the same environment, but in two different environments and that actually, they can only be reached by the enzyme that performs methylation (fibrillarin), at two different time points, so one after the other.

In order to verify that this is really the case and that the two sequences are not methylated independently of each other, we used another big advantage of NMR, which is the ability to do functional assays in the NMR tube, in the same conditions you saw the structure in. We ran activity assays in the NMR tube. This is the beauty of NMR.

NMR is the structural biology technique that is run in conditions closest to physiological conditions. The conditions we use are not physiological, of course, but they are the closest to physiological. Being in solution, we are able to run activity assays in the same conditions as we do collection of our structure restraints.

We conducted these functional assays in the NMR tube by using a carbon 13-labeled cofactor. This can be connected to what I said before. By having only one single label, you can follow exactly with NMR, what is happening at the specific site. By using this labeled cofactor, we could demonstrate that in solution, the methylations of these two rRNA sequences are dependent on each other. One of the two sequences is, so to say, the leader sequence; it needs to be methylated before the other sequence can be methylated efficiently. In this way, we determined that there is temporal regulation in the methylation of these two sequences that are paired.

Now, the question was what does this temporal regulation, this temporal dependence of one methylation on the other, mean?  What we think is that this is a process the cell uses to efficiently and quickly allow for ribosomal RNA folding . In fact, if you eliminate methylation from the cells, the cells are not viable anymore. This emphasises the fact that this methylation is essential for the life cycle of the cells.

We are now collecting cell biology data that would support this hypothesis and going deeper into dissecting exactly what the function of this methylation is. We do this in collaboration because our lab is not an expert in cell biology

All these functional experiments could now be designed in, owing to the information that we got from our structure and to the discovery of this sequential methylation of the two sites. This is something that neither the X-ray structure nor the electron microscopy structure were able to reveal.

Since NMR is the closest you can get to the physiological state, were you able to see what was really happening?

Yes, that is one aspect. The other techniques only take a static picture of the complex, whereas we can follow the whole cycle. The other aspect is that this complex is extremely flexible because it needs to assemble and disassemble in the nucleolus, while the ribosomes are being made.

It’s also an enzyme, so it needs to perform turnover; it needs to allow disassociation of the products and association of a new substrate. The electron microscopy and X-ray crystallography studies were probably just trapping the complex in one conformation, which might or might not be one of the states along the reaction cycle of this enzyme.

There is also the possibility that because the crystallography had to use a modified construct of the RNA in order to obtain the crystals, that induced a conformation that is different from the native one. We don’t actually know why the structures look different, but they definitely look different in solution and with these other methods.

So you weren’t put off by the fact that they got to the structure first, but as a result of their findings, you saw that something didn’t seem quite right?

It didn’t seem to explain a lot of the biology. I also had a very talented and really very smart student who was willing to take up the challenge. It is extremely important in science, that you have the right team, with smart, motivated people who have the understanding to question whether something is quite right and then continue investigating. That’s the key. I don’t want to take the merit of the whole story.

I think this is the most exciting and most successful project that we have ever done. We are continuing with this.

Read part two of the interview with Prof. Dr. Teresa Carlomagno about the research being done to uncover how drug leads bind to certain receptors.

About Prof. Dr. Teresa Carlomagno

Prof. Dr Teresa Carlomagno studied chemistry in Naples, Italy. She did her PhD partly in Naples and partly in Germany, in the group led by Christian Griesinger in Frankfurt, where she also stayed on for a post-doc. Then, she moved to The Scripps Research Institute in the US to work together with James Williamson.

With Christian, Prof. Dr. Carlomagno learned all the NMR and the methodology development side of NMR. With James Williamson, she wanted to learn the wet-lab part and she was particularly interested in RNA structure at the time. RNA structure was just coming up. People mostly liked to focus on proteins, but she wanted to learn how to produce RNA and how to look at RNA interactions with proteins, so she moved to the Williamson lab where she spent two years as a post-doc.

Then, she took up a group leader position at the Max Planck Institute in Goettingen. From there, she moved to the EMBL, where she also had a group leader position for about seven years. Since summer 2015, Prof. Dr Carlomagno has been in Hannover, where she holds a professorship. She’s also associated with the Helmholtz Centre for Infection Research in Braunschweig.