The Proximity of the 5' and 3' Ends of Single-Stranded Polynucleotides

Effective circularization of the RNA molecule is important for the genome replication of many RNA viruses, as well as for recruiting cellular cofactors to initiate translation of messenger RNAs. The proximity of the two ends of single-stranded nucleic acids has recently been argued to be a direct consequence of their large degree of self-complementarity, independent of nucleotide sequence and length. This conclusion has been numerically demonstrated by applying folding algorithms for the calculation of RNA secondary structures to many biological and random sequences. Here, we test experimentally the related hypothesis that linear and circular (covalently-closed) versions of a ssRNA or ssDNA are physically indistinguishable because they have essentially the same secondary and tertiary structures, hence the same relative positions of their ends. In principle, then, linear and circular forms of a long ssDNA or ssRNA genome should be indistinguishable by native gel electrophoresis. We have tested this hypothesis by comparing the behavior of linear and circular forms of M13 virus DNA. We found that under both native and denaturing agarose gel conditions the linear and circular molecules exhibit different electrophoretic conformers, suggesting that, contrary to expectations, they do not have the same secondary and tertiary structures. However, what is consistent with theory is that linear ssDNA has only one structural conformation in a native agarose gel, independent of where single restriction cuts have been made in the circular molecule. Experiments are undergoing to measure directly the distance between the ends of the linear molecule by cryo electron microscopy.