Lethal mutagenesis: targeting the mutator phenotype in cancer

Abstract

The evolution of cancer and RNA viruses share many similarities. Both exploit high levels of genotypic diversity to enable extensive phenotypic plasticity and thereby facilitate rapid adaptation. In order to accumulate large numbers of mutations, we have proposed that cancers express a mutator phenotype. Similar to cancer cells, many viral populations, by replicating their genomes with low fidelity, carry a substantial mutational load. As high levels of mutation are potentially deleterious, the viral mutation frequency is thresholded at a level below which viral populations equilibrate in a traditional mutation-selection balance, and above which the population is no longer viable, i.e., the population undergoes an error catastrophe. Because their mutation frequencies are fine-tuned just below this error threshold, viral populations are susceptible to further increases in mutational load and, recently this phenomenon has been exploited therapeutically by a concept that has been termed lethal mutagenesis. Here we review the application of lethal mutagenesis to the treatment of HIV and discuss how lethal mutagenesis may represent a novel therapeutic approach for the treatment of solid cancers.

Figure 1. Lethal mutagenesis by 0.1 µM 5-aza-5,6,-dihydro-2'-deoxycytidine (KP-1212) ablates HIV infection

Supernatants from HIV infected human lymphoblastoid CEM cells, cultured in the absence (blue) or presence (red) of 0.1µM of the mutagenic nucleoside analog KP-1212, were serially transferred to uninfected CEM cells. Viral production was quantified by the detection of p24 antigen (histogram; left scale) and by viral infectivity, measured by TCID50 (lines; right scale). In cells incubated with 0.1µM KP-1212, the amount of p24 was permanently reduced to less than the limit of detection (4 ng/ml) by passage 8, and no infectious HIV was recovered after passage 12.

Figure 2

Lethal mutagenesis of cancer. Due to the high fidelity of DNA replication and multiple mechanisms for the repair of DNA damage, normal human cells accumulate few mutations (on average approximately 70 per sexual generation). Cancer cells, however, accumulate large numbers of mutations (see Table 2). We propose that genetic instability in cancer cells is limited and, analogous to the situation for RNA viruses, a threshold of mutations exists above which cancer cells are no longer viable. Given their pre-existing mutational load, cancers can therefore be selectively ablated by the incorporation of mutagenic nucleosides.