Rooting the tree

Today’s post is another from regular guest blogger:

Ed Rybicki, Department of Molecular and Cell Biology, University of Cape Town, South Africa.

While fossilized viruses have never been found, we can often infer probable lines of evolutionary descent by analysis of extant genomic sequences. This sort of molecular phylogenetic approach has thrown up all sorts of interesting findings for a number of important viruses: for example, it is possible that the divergence of Herpes simplex types 1 and 2 (orofacial vs genital herpes viruses) occurred during the time humans separated from our nearest chimpanzee relatives - and possibly because the development of face-to-face sex in humans reduced the frequency of facial-genital contact, although this may only be idle speculation! Herpesviruses in general seem to have evolved along with us since before the separation of invertebrate and vertebrate lineages, and possibly since a prokaryote origin: herpesvirus capsid proteins are distantly related to tailed phage proteins, and thus possibly have a common - and very evolutionarily deep - origin. Human papillomaviruses have speciated along with us primates as well, and viruses preferring particular locations are often more closely related to viruses infecting the same locus on another primate than they are to other human viruses. I have speculated in print elsewhere on the “Gondwanan” distribution of geminivirus ancestors, and how their development can be traced back for at least several hundred million years.

But while this may be academically fascinating, how is it relevant to the real world? One special-case example - that of the re-awakening of a “fossil” endogenous human retrovirus - was covered on this site recently. This has all sorts of interesting implications for the evolution of new retroviruses by recombination, and for taking a good hard look at xenotransplants. However, a more recent development in the field of HIV evolution is even more important, inasmuch as it has very important implications for HIV vaccines.

An often-quoted factoid is that the extent of variation of HIV genotypes in one individual in a year is equivalent to the worldwide variation in influenza A viruses in the same time. If one considers that the rate of flu virus antigenic change requires annual redevelopment of flu vaccines, what does this mean for HIV? James Mullins’ group in Seattle and their co-workers in Pennsylvania have gone a ways towards answering this in their just-published paper in Journal of Virology entitled “Reconstruction and function of ancestral center-of-tree human immunodeficiency virus type 1 proteins”. They used phylogenetic analysis of many whole-virus sequences together with a new algorithm to computationally derive “centre of tree” (COT) ancestral protein sequences for HIV subtype B, as part of a strategy to design immunogens for use as vaccines that would elicit the broadest possible immune recognition of circulating HIV strains. The thinking behind this strategy is that use of ancestral sequences would minimise the genetic distance between a vaccine and circulating strains that could feasibly have speciated from it, compared to using any one of those strains, or even a consensus sequence.

Their Gag polyprotein sequence was capable of forming budded virus-like particles (VLPs); the ancestral Tat and Nef proteins retained appropriate function, and the proteins were immunogenic in terms of eliciting cytotoxic T-cell responses in mice. More importantly, the COT Gag elicited strong cross-subtype CTL responses when tested against peptide pools derived from subtypes A and M - and there was evidence that CD4 help was improved for these antigens. These are very important results when one considers the rate of change of HIV isolates: given that HIV phylogenetic variation can be depicted by starburst-like trees, and that the genetic distance between any two branch tips goes through a node, using the derived nodal sequences as vaccines could be a highly useful means of reducing the breadth of antigenic variation necessary to cover the current HIV spectrum - or even only part of it, given the geographic clustering of certain HIV subtypes. One could speculate that, when the correlates of protection for HIV vaccines are known (or in other words, when we finally know just what an HIV vaccine should look like), it could be possible to formulate COT-derived vaccine sequences on a yearly basis, as is routinely done for flu, based on up-to-date variation data. And given that subtype C - the target of the South African AIDS Vaccine Initiative and other vaccine development efforts - seems to be the prevalent form of HIV worldwide right now…it is possible that a single vaccine COULD protect people at risk in southern Africa, India and China. One lives in hope!

If you would like to be a guest blogger on MicrobiologyBytes, click the Guests tab above.

---

Possibly related posts: (automatically generated)

• Scientists Develop More Effective Form Of Bird Flu Vaccine
• New Bird Flu Vaccine Shows Promise