The Structure and Composition of Viruses

Viruses are generally made up of two parts, the outer protein shell (called a capsid) and the genetic infomation inside. Generally the morphology of a virus can be one of two structures, that of a sphere or that of a tube. A single infectious unit is called a virion and is made up of from ten to fifteen percent nucleic acid and from fifty to ninety percent protein. The general purpose of the proteins is to protect the genetic information. And that about does it for basic information here.

The first major concept to grasp when dealing with viral morphology is that viruses are composed of subunits. They ARE NOT nucleic acids wrapped inside of one gigantic protein. We know this from a variety of means. First, we know that proteins are never ever regular in shape, while virons are (and are symmetrical in most cases). Secondly, a coding triplet of DNA has a molecular weight of about 1000 daltons. The molecular weight of the amino acid coded for by that codon has a molecular weight of about 100 daltons. This 10:1 weight difference is not seen in experimental observation of viral composition (where no more than fifty percent of viral mass in nucleic acid). This means that to have a higher percent of molecular weight from proteins than DNA there must be more proteins than their are genes; eg each gene gets transcribed and translated numerous times.

There are also a number of reasons why being made of subunits is beneficial to virus and its chance of replication. As noted in the argument above, having one large protein encapsulate the entire genome is not genetically space efficient. Using subunits allows more economic packaging for DNA. Also, the shorter the gene is, the less the chance that a function-altering mutation will occur within that gene. This allows for two things: greater genetic stability and a rejection method. The rejection method is based on the example that a cell is often invaded by a number of viral particles at once. If one of these has a mutation in a subunit protein, the mutated product could, if the single-protein capsid hypothesis were true, improperly encapsulate the viral nucleic acid and lead to the virus's destruction. With multiple subunits, however, there is the possibility for a higher degree of specificity in subunit binding, and it is possible that a deleterious mutation would be unable to form a protein capable of integrating into a viral capsid. Although the mutated genetic information may still be packaged and passed on, no healthy information is destroyed in this process due to poor protection.

But what exactly is poor viral protection? What exactly must the capsid be capable of? Well, there are several things. First and foremost, the capsid must be able to in some way allow the viral nucleic acid to enter the cell. This could be a result of enzyme action, phagocytosis, or other means, but it must be done or the virus cannot replicate. Secondly, the capsid must be able to keep the virus safe while in the host body before entering a cell. Also the capsid must be able to resist environmental stresses incurred when outside of a host, often for long periods of times. In short the viral capsid must be very, very stable in order to successfully protect the genetic information.

The next question of course, then, is what makes a stable capsid? There are usually a few key ways to do this. First and foremost, free energy needs to be minimized. This needs to be done while allowing intermolecular binding to a high degree. And the best way to do both of these is through symmetry, in particular through helical and spherical symmetry.

Helical symmetry works (and spherical symmetry as well) because it maximizes both protein-protein interactions as well as protein-nucleic acid ones. That is why this is helical. DNA is in a helix (a double one to be precise), and it turns out that it is possible for proteins to align themselves in the major and minor grooves of the double helix in order to form the helical structure seen in tubular (or filamentous) virii. The only thing left for the virus to worry about are the ends of the thing, which are taken care of with special "head" and "tail" proteins. Spherical symmetry, exhibited by most animal viruses, is not quite spherical. However, it is a good approximation. And it is an approximation that uses cubic symmetry. Review the chart below:

Faces Subunits per face Total Subunits


cube 6 4 24


tetrahedron 4 3 12

octahedron 8 3 24

icosahedron 20 3 60


dodecahedron 12 5 60

What does this chart mean and what can we draw from it? Well, from observation we can determine that an icosahedral arrangement is the most common by far of all viral arrangements. This is due to the large number of subunits present; it allows small proteins (with small genes) to effectively act in large numbers creating high symmetry and solid protection. What's that, you say, symmetry? Why yes, take a look. There are three axes of symmetry on the icosahedron, ones that are 2-fold (along an edge in the middle), ones that are 3-fold (in the center of the triangle), and ones that are 5-fold (at the top of the durn thing). This high level of symmetry, as noted, is a very positive thing in that it allows fewer genes to do the job. But it turns out that icosahedrons also have other benefits.

To begin with, an icosahedron is not the say all/end all of virus capsids. There are more complex forms of icosahedral virii, ones with more than 60 subunits. In fact, icosahedral subunits may range up to around 1500 in number. How is this possible? Well, the icosahedral viruses know something that took Buckminster Fuller years to figure out: the geodesic dome designed through a triangulation of spheres. That is the next major topic.

Triangluation of spheres involves subdividing a big triangle into smaller ones. This creates different arrangements of both proteins and faces, creating some pentagons, some hexagons, and so on on the icosahedron. But it also creates non-equivalent positions; in other words the molecular forces may not be truely symmetrical and equal for each protein. Instead a close approximation is made, creating QUASIEQUIVALENCE.

We know all this through a combination of experimentation, observation, and mathematation (okay, so that's not a word. It sounds good). Keep in mind the following equations:

P = h^2 + hk + k^2

where h and k are any integers without common factors, and

T = Pf^2

where T is the triangulation number (or the number of smaller, equilateral triangles), f is any integer, and P is from the last equation. Once this answer is gotten, then:

60 * T = the number of subunits in the viral particle. (ranging from 60 to 1500, with 180 being the most common).

These subunits could be identical polypeptides, such as in the 180 subunit tomato bushy stunt virus, or ivolving different polypeptides, as seen in the picornovirus, where a series of pentameres lead to an icosahedron. But again another question is raised... what features exist on the proteins making up these subunits?

Fortunately smarter minds then mine have tackled this problem. As it turns out, antiparallel Beta-barrel structures are a very common feature in such virii as listed above. These features are assembled to make the generally wedge-shaped protein subunits, and are themselves made up a Beta-hairpin formed into a "jelly-roll" type configuration. As of yet no scanned images of such features are available, although hopefully they shortly will be.

This page will be updated again on Thursday, Sept 3rd.