Fig. 1. Comparison of the symmetry elements of a virus particle (right) with the seam structure of a soccer ball.
The celebration of the International Year of Crystallography (IYCr2014) has prompted me to remember the celebration of a remarkable achievement of the early years of macromolecular crystallography, just twenty years after the first protein structures (the two globins: myogoblin and hemoglogin) were solved. I was the fortunate participant in that singular event and inspired by the scientific achievement, the spirit of the group and the delight of the discovery, I was stirred to write a ballad to commemorate the event. Sadly, the passing of Pete Seeger (1919-2014), the inspirer of the music for the ballad, has put a sorrowful taint in these lines. May these brief words, written by a scientist who greatly admired him, serve as a small homage so that he can appreciate how immense the influence of his songs and actions has been in our times.
Most people would associate the term “ballad” with past achievements that could go back as far as the origins of story telling. It is certainly unusual to read or even hear this word associated with contemporary scientific events. However, in the early nineteen seventies this modern research project in structural biology reminded me of the mighty feats of medieval heroes. For nine long years, several generations of valiant postdoctoral investigators, led by Prof. M.G. Rossmann, struggled to adapt the methodology of protein crystallography to the solution of the atomic structure of the first spherical virus particles. Viruses (from Latin virus, poison) are cellular parasites unable to reproduce by themselves. The class of spherical (or isometric ) viruses was established to differentiate them from other plant viruses such as tobacco mosaic virus (TMV) that were more elongated, or fiber-like and generally helical. There were two other groups working on a similar endeavor. One at Harvard led by Prof. Stephen Harrison, focused on tomato bushy stunt virus (TBSV), at the time the best structurally characterized spherical virus. A second in Uppsala, Sweden, trying to solve the structure of satellite tobacco necrosis virus (SNTV), a very small satellite virus, under the aegis of Prof. Bror Strandberg.
The achievement established new methodology that is now routinely used, and the results obtained opened intriguing lines of research on the structure, function and evolution of viruses. As I worked on the project, the stanzas of a ballad came to my mind as the most natural way to express my admiration for the feat of all the participants. It was a way of documenting the main milestones of a journey of scientific discovery of which I was a lucky participant. If you do not hear from the other groups, it is not because their achievement was less significant. Absolutely not; they simply could not find their balladeer.
Southern Bean Mosaic Virus (SBMV) . The virus of our story is Southern Bean Mosaic Virus (SBMV), a humble RNA-containing plant virus that infects bean plants in the South of the United States. Neither SMBV nor its relative TBSV were ever as famous as the animal viruses that are fashionable today as human pathogens (for instance, AIDS virus or common cold -rhino virus-). Small as viruses go (approx. 300 Å in diameter), non-enveloped, single-stranded, RNA plant viruses like them were easy to obtain in gram quantities from a few infected plants. In addition, they were easy to crystallize and consequently they were the objects of a concerted effort to obtain their atomic structure by X-ray diffraction methods with conventional in-house X-ray sources.
Viruses had to be constructed from a few identical subunit. The icosahedral symmetry of small spherical viruses had been proposed by Watson and Crick in the early fifties[1]. They hypothesized that the limited size of the coding material in a virus (either DNA or RNA), could only code for a few proteins and therefore the virus particle had to be assembled by the repetition of a single –or a few- protein chains. Therefore, they predicted that the virus envelopes would be highly symmetrical, and most likely icosahedral, containing at least twenty copies of the coat protein in the shell, or capsid. The detailed arrangement of the proteins in the capsid on the surface, and a preliminary classification of icosahedral viruses was presented by Caspar and Klug in their classic 1962 paper [2].
The symmetry of the particle. It is instructive to look at the geometrical arrangement of protein chains on the surface of a virus particle from the viewpoint of more familiar objects. Think of a mosaic tile of hexagonal entities covering a planar surface. A moment’s thought would make you realize that in order to curve a planar surface covered by hexagons you have to introduce a certain number of pentagonal links or tiles; otherwise the surface will always be flat (Fig. 1, lower part and right). A close look at a soccer ball composed of pentagonal and almost perfectly hexagonal parts (quasi-hexagonal in the jargon of virus crystallography, Q6) sawn together illustrates the concept (Fig. 1, left). Similar geometrical principles were used to build the celebrated geodesic domes of Richard Buckminster Fuller (1895-1983), and have now been seen again in a special form of carbon containing clusters of sixty atoms (referred to as fullerenes or buckyballs). In fact, it was recognized by Caspar and Klug that the inspiration for their insight into virus structure sprung from the arrangement of components in the geodesic domes. Based on this conceptual framework, the geometric lattices underlying the construction of certain spherical virus permits their classification as T=1 (icosahedron: 60 protein subunits, 20 x 3), T=3 (180, larger virus containing 180 subunits, 60 x 3), T=4 (still more complex lattices containing 240 (60 x 4) protein subunits) and others (Figs. 2). This is a key parameter in understanding the three-dimensional structure and external appearance of icosahedral viruses(2).
The atomic Structure of Virus Particles. Nevertheless, there was yet no atomic model for an icosahedral virus particle well into the1970’s. As initially proposed by Michael Rossmann., the crucial factor in the determination of the structure was the presence of several identical copies of the polypeptide chain in the asymmetric unit: i.e. the presence of non-crystallographic symmetry. In those days, the major hurdle was to devise algorithms and programs, which would allow averaging of enormous electron density maps, containing many millions of grid points, over the redundant copies in the asymmetric unit. These methods were also employed to solve the structure of the common cold virus (rhinovirus) and are today standard in the structure determination of viruses and large protein complexes made up of several identical copies. The structure of the polypeptide chain of the different subunits of SBMV eventually emerged from the electronic density maps and I had the priviledge of building the atomic model (Fig.2).
Fig. 2. Atomic structure of the entire virus particle of SBMV shown with computer graphics. Each subunit is shown as the Carbon alpha tracing in different colors to depict the different geometric environment within the icosahedral shell. Red: 5-fold; Blue: 3-fold, quasi-6-fold; Green: 2-fold. Compare with the schematic depiction of Fig.1 (right); each triangle-shaped unit corresponds to one polypeptide chain. Copyright CAZ.
One of the most striking results of the structure determination of SBMV was the similarity of folds between the protein making the capsids of TBSV[3]and SBMV[4], and later STNV[5]. This unifying principle had an enormous impact in understanding the structure, function, evolution and diversity of a wide spectrum of viruses. The X-ray structure of SBMV was soon followed in 1982 by the complete amino acid sequence of the capsid protein and then the different amino acids could be related to their function in the assembly and disassembly of the virion particle. The impact that this early work on virus has had in the virus of structural virology can be appreciated in Fig. 3, reproduced from a recent publication by the Protein Data Bank, attempting to extract trends in data deposition[6]. There are now near 400 viral structures deposited in the PDB [6]. This remarkable figure shows how the different groups that have published crystallographic structures of viruses since the early structures appeared in the late 1970s and early 1980s, are all interconnected and related to the pioneer achievements of the structures of the three viruses related above: TBSV, SBMV, STNV (Fig. 3).
The detailed structural analysis of virus structures is being used to understand how virus particles infect their host cells and to design drugs against the common rhinovirus and other viral pathogens[7]. The epic feat of the determination of the first atomic structures of viruses should be part of the folklore of macromolecular crystallography.
Fig. 3. Inter-connection among the different structural groups that have solved and deposited virus structures at the PDB. For more details consult Fig. 11 for reference [6]. This figure reproduced with permission from Herman et al. Copyright by Elsevier, FESB and Oxford University Press. All rights reserved.
The entire original text of the ballad in English has been published in the official journal of Spanish Society of Virology (
SEV, Revista de la Sociedad Española de Virología, 2013; Volume 13(3), 66-69)
(www.cbm.uam.es/sev/revista.html) and is appended below. You can follow the entire text of the Ballad with some explanatory notes in Spanish. I am extracting here the most relevant ones as published on Chapter 22 of
Crystals and Life (https://
www.uic.edu/labs/caz/crystals/index.html) with some clarifications. You can consult this chapter also to fully appreciate the structural relationships among the different viruses as they were unveiled in the early 1980s.
Notes to some stanzas of the Ballad of the 2.8 Å Structure of SBMV
M.G.R. began to work on small plant viruses soon after the structure of lactic dehydrogenase (LDH) had been published. He went to Upssala, Sweden for a sabbatical leave in the laboratory of Bror Strandberg.
LDH has now been solved
I must find something to do
Rossmann fold has been proposed
I’ll take a sabbatical leave (repeat)
And I’ll look at the STNV.
Soon after the return from the sabbatical leave he started to work on SMBV. Afternoon tea was a ritual at that time in the laboratory with a distinctly international flavor, and where the work progress was discussed.
Shall we start by growing some crystals?
It’s only a matter of weeks
After that we can write some proposals
For the future of SBMV (repeat)
You should drink your afternoon tea.
Although M.G.R.’s dream was to solve viruses ab initio, he soon realized that heavy atom derivatives were a safer route at the time. Of course, he continued to sail in Lake Freeman, Indiana.
Heavy-atoms must now be found
Playing chemists is all we must do
One alone will be safer ground
For the structure of SBMV (repeat)
I’m sailing the Indiana sea.
After eight years of work, the atomic model of SBMV slowly grew as a metallic sculpture made up of Kendrew parts in a forest of metal rods within a Richard’s Box. M.G.R. kept bumping his head against the top of the box, so he purchased a hard hat, which he rigorously wore for his building sessions with me.
Eight years have already passed
Many people have done their best
I won’t say the struggle has finished
For the structure of SBMV (repeat)
I’ll buy a helmet for me.
The fold of SBMV turned out to be a beta-barrel (or jelly-roll) almost identical to the one already described by Steve Harrison and colleagues for the structure of TBSV.
After so many years of labor
All we have is a barrel of sheet
Old Steve did us a favor
With the structure of TBSV (repeat)
We can trace our SBMV.
———————————————————
The entire original text of the ballad was sung at a party at the Rossmann’s residence on Sunday, November 4, 1979 to celebrate the solution of the three-dimensional structure of SBMV. The melody was adapted from a song by Pete Seeger (the grandfather of American folk music) that I heard on the radio one beautiful Autumn morning on my way to the lab. The entire ballad was meant to be an homage to all participants in the project of the three-dimensional structure of SBMV. Many of them I have met through the years at meetings and conferences. I shared with others hours of effort, frustration and excitement in the basement of the Lilly Hall of Life Sciences at Purdue University, West Lafayette, Indiana. The years spent in that laboratory still bring cherished memories. A unique laboratory whose day-to-day routine was masterly orchestrated by Sharon Wilder. To all of them (the unsung heroes of this ballad), and to many others who participated in less visible ways, I want to express my deep appreciation: Sherin Abdel-Meguid, Toshio Akimoto, J.E. (Jack) Johnson, Andrew G.W. Leslie, Ivan Rayment, Michael G. Rossmann, Ira Smiley, Dietrich Suck, Tomitake Tsukihara and Mary Ann Wagner. I am just a modest minstrel, the troubadour of this epic feat.
Notes
(1). Crick, F. H. C. & Watson, J. D. Nature (1956), 177, 473-475.
(2). Caspar, D. L. D. & Klug, A. Cold Spring Harbor Symp. Quant. Biology (1962) 27, 1-24;
(3). Harrison, S. C., Olson, A. J., Schutt, C. E. Winkler, F. K. & Bricogne, G. Nature (1978) 276, 368-373.
(4). Abad-Zapatero, C., Abdel-Meguid, S.S., Johnson, J. E., Leslie, A. G. W., Rayment, I., Rossmann, M. G., Suck, D., & T. Tsukihara. Nature (1980) 286, 33-39.
(5). Liljas, A. et al. J. Mol. Biol. (1982), 195, 93-108.
(6) Berman, H. et al. FEBS Letters (2013), 587, 1036-1045.
(7) Perutz, M. From a Tomato Virus to Tumor and Influenza Viruses. Chapter 8. In Protein Structure. New Approaches to Disease and Therapy. W.H. Freeman and Company. 1992. New York.
The Ballad of the 2.8 Å Structure of SBMV (Complete text)
LDH has now been solved [1]
I must find something to do
Rossmann fold has been proposed
I’ll take a sabbatical leave.
I’ll take a sabbatical leave
And I’ll look at the STNV.
Would you like to get some experience? [2]
It will help your future a lot
It’s your turn to do some big science
Come to work on the SBMV
If you work on the SBMV
You’ll have your afternoon tea.
Shall we start by growing some crystals? [3]
It’s only a matter of weeks
After that we can make some proposals
For the future of SBMV
If you work on the SBMV
You should have your afternoon tea.
Would you like to collect some data? [4]
It will take only several months
I am sure we will have some errata
In the work on the SBMV
If you work on the SBMV
You must have your afternoon tea.
We should now process some native [5]
My programs will make it go fast
After that we’ll have new perspective
On the structure of SBMV
For the structure of SBMV
We still drink our afternoon tea.
Heavy-atoms must now be found [6]
Playing chemists is all we must do
One alone will be safer ground
For the structure of SBMV
For the structure of SBMV
I’ve sailed the Indiana Sea.
Other films must now be processed [7]
Small changes is all that we need
More files will have to be accessed
For the structure of SBMV
For the structure of SBMV
I’ve sailed the Caribbean Sea.
Tsukihara computes day and night [8]
MR-map must be calculated
Might longer that I’d have liked
For the structure of SBMV
For the structure of SBMV
I’ll have the cell constants at least.
Eight years have already passed [9]
Many people have done their best
I won’t say the fight has finished
For the structure of SBMV
For the structure of SBMV
I’ll buy a helmet for me.
After so many years of labor [10]
All we have is a barrel of sheet
Old Steve did us a favor
With the structure of TBSV
With the structure of TBSV
We can trace our SBMV.
Michael Rossmann explains evolution [11]
For the virus it’s easy to do
Old Steve reacts with emotion
To the Structure of SBMV
Since the structure of SBMV
Is like the one of TBSV.
Pat will do some more prediction [12]
Only thing he can do without labor
It will give bad reputation
To the structure of TYMV
For the structure of TYMV
Older brother of SBMV.
With the virus structure in my hand [13]
New ideas and projects arise
This is only the start of the fun
With the structure of SBMV
With the structure of SBMV
I’ll take a sabbatical leave.
If I can reduce my sabbatical [14]
I could certainly go to Israel
On the way I’ll make some proposal
For the structure of STNV
For the structure of STNV
Little brother of SBMV.
What comes after sabbatical [15]
Ribosomes, proteosomes, you name it.
This man is a structural radical
Since the structure of SBMV
With the structure of SBMV
We have had our pizza free.