Michael Summers is the Professor of Chemistry and Biochemistry at the University of Maryland, Baltimore County, and an investigator with the Howard Hughes Medical Institute. In this interview, he tells us about his work investigating the structure of large RNAs using NMR spectroscopy.
When did you get involved in NMR spectroscopy?
I’ve been doing NMR since I was a graduate student at Emory University in the mid ‘80’s, and my real training in NMR was as a postdoc at NIH with Ad Bax. I’ve been using NMR as our main tool to study HIV structural biology since I took my faculty position in 1987.
Can you share some results of your recent work on RNA structures?
We’re really excited because we’re using NMR to look at really large RNAs. RNAs are challenging for a number of reasons, the main one being that with only four typical nucleotide types, the signals tend to come in very crowded regions of the NMR spectrum. So, historically, the sizes of RNAs that have been structurally characterized is only about 25 or 30 nucleotides. We’re developing some labelling techniques and some different kinds of strategies that have allowed us to look at much larger RNAs.
We’ve recently reported that we can get structural information from RNAs containing as many as 700 nucleotides, which is a huge molecule. We’re just now trying to develop tools so that we can actually develop three-dimensional models of RNAs that are that large.
Although we’ve been focusing most heavily on the labelling strategies and the approach that we would use, we’ve more recently started collaborating with Ad Bax using conventional NMR methods to try to take advantage of long range proton nitrogen scalar couplings to bring three and four-dimensional NMR to bear on these problems.
What impact could your work have on biomedical research?
It’s hard to say where the impact is going to come from. With some of our studies, we’ve been able to develop new ways of inhibiting HIV. When we solve the structure of one part of the virus, we realize that, just based on the structure, that there might be a way of inhibiting maturation. So, we’ve identified inhibitors, we’ve patented them, we’ve licensed them and there are companies that have been trying to get them to the point that they might be useful.
Using NMR within research to inhibit HIV
I don’t know where that will eventually go. What we’re most excited about is trying to understand how the virus works. I think that by getting information about how the key parts of the virus fit together, that may help other people who focus mainly on drug discovery. If we have some idea, we’re certainly going to try it out, but I hope that the basic principles of biology, of HIV virology, will be helpful more broadly.
Also, the things that we’ve learned about focusing on large RNAs and how you would try to determine the structure and dynamics of large RNAs, hopefully that could be broadly useful as well in the biomedical community.
How important is instrumentation in your work?
We couldn’t do anything without our instrumentation. NMR is the central tool that we use. We have attempted, in the past, to use other kinds of technologies. Recently, we’ve tried to grow crystals of some of our protein RNA complexes, of some of our RNAs, and there’s just no way that’s going to happen.
What we know now is that as we look at larger and larger biologically functional RNAs, they are flexible, they’re undergoing conformational changes, so the idea that they might crystallize is probably pretty farfetched. We’ve been able to learn a lot using NMR that I don’t think you’d be able to learn using any other technique.
What improvements in the technology would you like to see in the future, to help push your research forward?
The typical things that would help us are things like the improved signal-to-noise and resolution that we would get at higher fields. I’m not a physicist - I need to qualify that upfront, I was trained mainly as an inorganic chemist. But every time there’s been a big increase in field strength, we’ve seen major advantages. We’re collecting data that we can interpret at 800 megahertz that I don’t think would be possible on our 600 megahertz magnets. Higher field, higher signal-to-noise, and better resolution - those kinds of things are likely to really help us.