Talk summary: human vs. viral genetic arms race

The speaker at yesterday's microbial pathogenesis series (UCSF) was exceptional.   Harmit Malik spoke about viral co-evolution with (primate) host defense responses, casting the process as a genetic arms race. Using signature gene sequences and mining available genomes, he traced HIV, hepatitis C, etc. back to various stages in primate evolution.  Because retroviruses insert their DNA into ours (sometimes becoming part of our germ line), by looking for signature viral genes one can type the viral burdens of our ancesters back in some inferred cases by hundreds of millions of years.  Harmit then went into case studies showing how a virus acquired functions to subvert host defenses and described the host response to that - all at the amino acid level of key proteins over evolutionary time.  His lab further showed that if you reversed a host mutation back to its "ancestral" state, it lost the ability to block viral activity.  The research article is unfortunately closed access, so I won't link to it.

Anyway this led to a discussion with down-the-hall colleague Suzanne Noble in which she pointed out that viruses are thought to be most virulent when they hop to a new host species, decimating it because no host-pathogen equilibrium has yet been established.  Over time, the host evolves a response and the virus becomes more stably integrated within the population.  So is the end "goal" of a (retro)virus to not be a virus at all, but to be a heritable genetic element in the host genome?  Another colleague in our lab, and always-fun-discussion partner, Mark Voorhies, pointed out that the best of all worlds is to retain the ability to synthesize virions and induce lysis, such that if the host runs into trouble the latent virus can jump ship.  Such a strategy is well-characterized in bacteriophages and is called lysogeny.  However in this case I suspect it's more complicated.  For instance, the less frequently a virus escapes its host, over evolutionary time, the less capable it will be to successfully escape.  The requisite genes, having been isolated from selective pressure while dormant and transmitted, will not necessarily be competent to escape the host. 

Anyway it was a fun morning.  If you ever get a chance to hear Harmit speak I strongly recommend taking it; we see many speakers at UCSF and he was a stand out example of how to give a talk.

If you have any comments or want to set me straight on retroviral evolutiion and its consequences, please comment away.

update: here's the PLoS Biology comment introducing Harmit's publication: