What Should We Be Worried About?

THAT WE WON’T MAKE USE OF THE ERROR CATASTROPHE THRESHOLD

WILLIAM MCEWAN

Postdoctoral researcher, MRC Laboratory of Molecular Biology, Cambridge, U.K.

Viruses replicate near the boundary of fidelity required to successfully pass information to the next generation. I worry that we will not devise a way to push them over that boundary.

We are at a strategic disadvantage in the fight against viral infection. Our genomes are large. Most of our copying errors will be deleterious to survival. We must therefore replicate our information faithfully—about 1 error in 1010 nucleotides per replication cycle is the mammalian rate. Meanwhile fleet-footed RNA viruses sample 4 orders of magnitude more sequence diversity with each generation. And a virus generation is not a long time.

But error-prone replication is not without limits. There comes a point, the error catastrophe threshold, where the replication of genetic information cannot be sustained. Beyond the threshold, the ability to replicate is lost. Gently tipping viruses over this threshold has been suggested as a therapeutic intervention. And what a beautiful idea! The error catastrophe threshold defines the boundary of the heritablility of information—the boundary of life. On the other side lies the abyss.

Error catastrophe has been modeled theoretically. In support of theory, tissue culture and mouse studies sustain accelerated mutation as an antiviral strategy. But unfortunately the same drugs that induce accelerated mutation in viruses are toxic to the host. Error-inducing drugs that are specific to particular viruses are feasible but, when compared to outright inhibitors, seem a risky game.

For now, we can delight in the recent knowledge that nature has already got there. The APOBEC3 family of genes systematically introduces errors into virus genomes, tipping them over the precipice. Viruses take this threat seriously. HIV has a gene whose function is to counteract APOBEC3. Without this gene, HIV cannot replicate. Other viruses seem to have lost the battle: The human genome is littered with the remnants of extinct viruses that bear the scars of APOBEC3 activity. Perhaps accelerated mutation was a contributing factor to these viruses’ demise. I fear that evolution’s ingenuity—accumulated at that geologically slow rate of 1010 base pairs per generation—may surpass our own, unless we can grasp this beautiful idea and devise some gentle, nontoxic tipping of our own.