The infamous propensity of viruses to mutate their way to evolutionary success makes them a slippery target for antiviral drugs. But a new approach that exploits the effect of dominant–negative viral alleles promises to stem the growth of drug-resistant particles, essentially beating viruses at their own game.

All viruses mutate at high rates, but those with a single-stranded RNA genome, such as poliovirus, top them all and are therefore most adept at evolving resistance to conventional drugs. Scott Crowder and Karla Kirkegaard tried an alternative tactic, which involved identifying mutant lines of poliovirus that could interfere in a dominant–negative way with the growth of wild-type particles. Using reverse engineering, they produced a series of poliovirus genomes and then selected those that generated non-viable (that is, growth-defective) particles. Next, they co-transfected mutant and wild-type viruses into cells and investigated whether the yield of wild-type viruses had been affected.

The dominant mutations, which fell into four classes, were very effective. For example, dominant mutations in several capsid proteins inhibited the growth of wild-type viruses by 93%, on average, in just one round of co-infection. But how successful would such a virus-busting strategy be? One test shows that the predicted benefits translate well into practice: a dominant drug-sensitive virus was able to cut the yield of a co-transfected drug-resistant virus down to 3–7% of its normal titre — presumably because the dominant capsid protein destabilized the multi-subunit coat of the resistant virus.

The genomics screen described in this article has revealed some unexpected details of viral biology. But these results will perhaps be even more attractive to drug developers, whose eye will be on exploiting the dominant proteins as targets for anti-viral compounds.