Discovering Zika

Editorial

Rebecca Pool

Friday, November 18, 2016 - 12:30
Image: From enzymes to viruses, Michael Rossmann has unraveled some of the world's most complex structures.
 
In March this year, researchers from US-based Purdue University, led by physicist Professor Michael Rossmann and virologist Professor Richard Kuhn, were the first to reveal the structure of the Zika virus.
 
Using an FEI Titan cryo-electron microscope with Gatan K2 detector, Rossmann, Kuhn and colleagues mapped the virus atomic structures to 3.8 Å resolution, pinpointing regions that differed from the related dengue, West Nile and yellow fever viruses, they had previously determined.
 
The world is now closer to an effective vaccine, but for Rossmann, the breakthrough underlines his persistent passion for structure, symmetry and puzzles.
 
"I started working with plant viruses around 1971, and this eventually led me to Zika," he says. "Curiosity pushed me on through this and I've always found it's fun to solve problems; solving one problem usually gives you ideas for ten other problems."
 
The surface of the Zika virus provides insights critical to the development of an effective vaccine and antiviral treatments. The outside surface (also on right) shows the classic “herringbone” arrangement of the major glycoprotein (yellow), that allows the virus into a cell. [Purdue]
 
Born in Frankfurt, Germany, Rossmann studied Physics and Mathematics at the University of London where he received BSc and MSc degrees. Come the early 1950s, he moved to the Royal Technical College in Glasgow, to lecture electricity and magnetism to physics undergraduates, but in his words: 'I really wasn't very satisfied here'.
 
Keen to find something he actually wanted to do, he contacted Professor John Monteath Robertson, an early researcher into the structure of organic molecules using crystallography, at nearby University of Glasgow. He was admitted as a graduate student into Robertson’s laboratory, and as he puts it: "I've enjoyed myself ever since."
 
With his doctorate on 'A Study of Some Organic Crystal Structures' in hand, Rossmann left for the University of Minnesota in 1956, where he worked for two years as a postdoctoral fellow with Professor William Lipscomb. Here, Rossmann spent a lot of time writing computing programs to analyse crystal structures.
 
"I stayed with Lipscomb for two years and learnt how to use what was one of the first commercially available electronic computers," he says. "Here, I wrote most of the essential programmes for doing crystallography."
 
"In Glasgow we'd had to do all our enormous calculations by hand, and now having a computer to do this was a new, wonderful world," he adds. "This was a place where you could be very creative in how you approached things."
 
Lipscomb would later win a Nobel Prize in Chemistry for studying borane structures, but come 1958, Rossmann's US visa had expired so he travelled back to the UK. This time he went to work as a research associate at the University of Cambridge with Max Perutz, studying the structure of haemoglobin.
 
Rhinovirus: the first animal virus to have its structure published.
 
Perutz had set up the Molecular Biology Unit at the Cavendish Laboratory, which had already attracted Francis Crick and and James Watson, soon to win Nobel Prizes for their discoveries on the molecular structure of nucleic acids.
 
As Rossmann highlights: "Crick was in the office opposite and Watson was a frequent visitor; so with Perutz, we had many coffee conversations and this was such a highly stimulating environment with all these future Nobel Prize winners."
 
Whilst here, Rossmann spent a lot of his time writing programs for the custom-built EDSAC 2 computer. This was the successor to early British computer, the Electronic Delay Storage Automatic Calculator. 
 
"I had a wonderful time programming this computer and developing methods to solve the structure of haemoglobin," he says. But crucially, his patience and painstaking analyses had also sowed the seeds for a new method to solve protein crystal structures that would eventually prove instrumental to future crystallographers.
 
While, Rossmann had helped to develop 'multiple isomorphous replacement', a sometimes difficult method widely used to recover missing phase information from protein crystals, he now realised the non-crystallographic symmetry that existed within and between crystal structures could also be exploited.
 
From here, he went on to develop the “Molecular Replacement”, a relatively straightforward method to map the structures of protein complexes, now used to determine the majority of the hundreds of thousands of structures deposited with the Protein Data Bank.
 
Enzymes first
 
Come 1964, Rossmann joined the Department of Biological Sciences at Purdue University, as associate professor, where he still resides today as Hanley Distinguished Professor of Biological Sciences.
 
On leaving Cambridge, Rossmann had wanted to determine the structure of viruses, but explains: "No atomic resolution virus structures were known at the time, and the idea of working on a structure of a virus was mostly considered to be ridiculous."
 
So instead, the up and coming researcher looked at smaller protein complexes, choosing lactate dehydrogenase, found in nearly all living cells and vital to metabolism.
 
Computer illustrations of the immature (left) and mature dengue virus; this was the first flavivirus structure to be determined. [Purdue] 
 
At the time, researchers around the world were discovering enzyme structures, such as alcohol dehydrogenase, adenylate kinase and hexokinase. And as Rossmann explains, he had started to recognise certain properties in these structures that were repeated over and over again.
 
This insight led to his discovery of the 'Rossmann fold', a nucleotide binding motif associated with enzymes that is now described in most biology textbooks.
 
"We worked so hard to get these results but as soon as we had discovered and realised the significance of this common protein structure I thought, 'well maybe I can get some money to do some virus work'," says Rossmann. He did.
 
By now it was the early 1970s, and while computers were rapidly advancing, the relatively slow processing powers of the day meant determining the structure of human viruses was still unthinkable. So instead, Rossmann focused on the southern bean mosaic virus. In his words: "Purdue University had lots of greenhouses so we could go and infect bean plants with the virus and then collect grams of the virus very easily."
 
Come the end of this decade, the researcher and colleagues had constructed a 3D model of the Southern Bean Mosaic virus, making them the second research team in the world to map a virus.
 
Now keen to move onto human viruses, Rossmann contacted Professor Roland Rueckert from the University of Wisconsin, who was making great strides towards determining the structures of small RNA viruses, picornaviruses.
 
Rueckert persuaded Rossmann to take a look at the structure of a rhinovirus, the predominant cause of the common cold. And with funding from Purdue, Rossmann set up a tissue culture laboratory to propagate the virus.
 
Rossmann and colleagues team went onto produce crystal after crystal after crystal of the human rhinovirus-14, one of about 100 known common cold virus strains. Then, using the Cornell University High Energy Synchrotron Source, they bombarded their delicate samples with short, intense X-ray bursts, recording the virus structure across hundreds of diffraction patterns at various angles.
 
With patterns in hand, they reconstructed an atomic map of the virus using a then, state-of-the-art, Cyber 205 super computer. It was 1985, and at the time, the Rhinovirus narrowly beat the polio virus as the first animal virus to have its structure published.
 
Closer to Zika
 
While Rossmann had relied on X-ray diffraction to determine past viruses, he was well aware that cryo-electron microscopy was on the rise. Having taken a short sabbatical with molecular biologist Richard Henderson at the Laboratory of Molecular Biology in Cambridge, he began to combine cryo-electron microscopy with X-ray crystallography on his return to Purdue.
 
In the Spring of 1998, Rossmann's group had analysed in atomic detail the part of a cell's receptor that binds to a cold virus. He and his colleagues used the structure of the common cold virus to develop an anti-common cold agent, Pleconaril. Then in 1999, the researchers won the first of a series of multi-million dollar grants from the US National Institutes of Health to study a series of viral pathogens belonging to the flavivirus family.
 
Richard Kuhn (left) and Michael Rossmann; between them, the researchers have solved flaviviruses and more.
 
By now, Rossmann had joined forces with Purdue biologist, Professor Richard Kuhn, and over at least the next decade, the researchers used X-ray diffraction and cryo-electron microscopy to determine the dengue and West Nile virus structures, and more. However, Zika would be different; this time, the researchers relied on cryo-electron microscopy alone to determine its structure.
 
According to Rossmann, past flavivirus structure discoveries had been instrumental to Zika success but recent advances in cryo-electron microscopy and detector technology were imperative to further progress.
 
"Once Kuhn had produced a pure sample, it didn't take long [only months] to determine the structure," he says. "The Gatan K2 detector was critical. I knew we needed this, so I begged the Purdue University's president for the money; I guess I begged hard enough."
 
And at a time when the virus is still sweeping across South America, parts of Asia and Africa, resolving its structure has never been more important. Researchers worldwide now have a map of the virus, revealing regions that could be vulnerable to treatments. And as such, many are trying to develop effective vaccines.
 
A successful vaccine will be delivered soon, thanks to Rossmann and colleagues mapping the structure of Zika virus.
 
Rossmann doesn't believe he will be the first to develop such an anti-viral compound, but he reckons a successful vaccine will be developed within three years. And as he points out: "While studying ways to develop a vaccine we will also learn a lot more about the virus; the same happens with all studies - you make discoveries and they become useful in unexpected ways."
 
So what advice would the man, who has solved what can only be described as some of the world's greatest structural biology puzzles, give to the researchers of tomorrow? "Just be yourself and be interested," he says. "If you're not interested, forget it."
 
Website developed by S8080 Digital Media