Lagging strand encoding promotes adaptive evolution

AbstractCells may be able to promote adaptive evolution in a gene-specific and temporally-controlled manner. Genes encoded on the lagging strand have a higher mutation rate and evolve faster than genes on the leading strand. This effect is likely driven by head-on replication-transcription conflicts, which occur when lagging strand genes are transcribed during DNA replication. We previously suggested that the ability to selectively increase mutagenesis in a subset of genes may provide an adaptive advantage for cells. However, it is also possible that this effect could be neutral or even highly deleterious. Distinguishing between these models is important because, if the adaptive model is correct, it would indicate that 1) head-on conflicts, which are generally deleterious, can also provide a benefit to cells, and 2) cells possess the remarkable ability to fine-tune adaptive evolution. Furthermore, investigating these models may address the long-standing debate regarding whether accelerated evolution through conflicts can be adaptive. To distinguish between the adaptive and neutral models, we conducted single nucleotide polymorphism (SNP) analyses on wild strains of bacteria, from divergent phyla. To test the adaptive hypothesis, we analyzed convergent mutation patterns. As a simple test of the neutral hypothesis, we performedin silicomodeling. Our results show that convergent mutations are enriched in lagging strand genes and that these mutations are unlikely to have arisen by chance. Additionally, we observe that convergent mutation frequency has a stronger positive correlation with gene-length in lagging strand genes. This effect strongly suggests that head-on conflicts between the DNA replication and transcription machineries are a key mechanism driving the formation of convergent mutations. Together, our data indicate that head-on replication-transcription conflicts can promote adaptive evolution in a variety of bacterial species, and potentially other organisms.