Figure 3, Jern et al
Simulation of hA3G-mediated HIV-1 evolution.

One of the host defenses against HIV1 is “APOBEC3G” and related proteins.  These proteins force HIV to hypermutate, killing its ability to replicate in the next cell it infects.  On the other hand, it looks as if low-level mutation by APOBEC3G over the decades has driven HIV evolution:

We have found that hA3G activity acting on prior generations of virus has left detectable footprints in the HIV-1 genome. 2

HIV can only infect human cells because it has a defense against APOBECs: the viral protein “vif” causes APOBECs to be destroyed, and the virus is able to replicate without being hypermutated. 

So let’s say we develop an antiviral drug that blocks vif.  APOBECs would drive hypermutation of the virus.  This hypermutation would include the gene encoding vif.  Would this mutation, in a kind of molecular judo, drive rapid evolution of vif, so that it becomes resistant to the drug?

According to some experiments 2  both in cells and in silico, perhaps not:

However, since the predicted effect on resistance to standard antiviral drugs is likely to be small, we propose that concerns over increased resistance mutations should not impede development of HIV-1 Vif as a candidate drug target. 2

I’m not quite convinced this paper was actually modeling the phenomenon they say they’re modeling — is vif that’s shut off by deliberate mutations the same as vif that’s blocked by a (hypothetical) drug? But it’s a useful start, anyway.

  1. and other lentiviruses; as well as quite a few other virus types[]
  2. Patric Jern, Rebecca A. Russell, Vinay K. Pathak, & John M. Coffin (2009). Likely Role of APOBEC3G-Mediated G-to-A Mutations in HIV-1 Evolution and Drug Resistance PLoS Pathogens, 5 (4) DOI: 10.1371/journal.ppat.1000367[][][]