A design principle for a single-stranded RNA genome that replicates with less double-strand formation

Usui, Kimihito; Ichihashi, Norikazu; Yomo, Tetsuya

In this study, we attempted to understand design principles of an ssRNA genome that replicates without dsRNA formation. We first proposed a structure-dependent dsRNA formation model at replication (Figure 1 ) and then obtained a set of supportive evidence for the model using RNAs with various structures along with Q␤ replicase. From these results, we extracted a critical rule for RNA sequences (less GC number in loops) to continuously replicate as a single strand, and then we designed an ssRNA that encoded an ␣domain of ␤-galactosidase and was continuously replicable, mainly as a single strand. Furthermore, we also found evidence that the ssRNA genomes of all bacterial and of most fungal viruses presently known obey this rule. In summary, this study revealed one of the structural design principles of an ssRNA genome, which is useful for developing an artificial ssRNA genome and contributing to our understanding of the structural constraints governing the ssRNA genome of bacterial and fungal viruses. A plausible mechanism for the dsRNA reduction by the 'less GC number in loops' rule is that there is a decrease in partial dsRNA formation between loops on the template and the newly synthesized RNA, as shown in the model (Figure 1 ). We do not deny the possibility that the mechanism of dsRNA formation depends on the host factors of Q␤ replicase (EF-Tu and Ts), one of which is known to mediate dsRNA separation during replication (35) . Nevertheless, if this dsRNA formation mechanism is independent of the types of replication enzymes, this rule might be applicable to other RNA replications catalyzed by different replication enzymes, including ribozymes. The formation of dsRNA is considered to be a serious problem in RNA replication by ribozymes in the hypothetical RNA world (3) and in in vitro RNA replication systems (39) . The results obtained in this study suggest that the dsRNA problem of primitive genomic RNAs can be solved by reducing the GC number in loops.