WHY PENTOSE-NUCLEIC AND NOT HEXOSE-NUCLEIC ACIDS .7. PYRANOSYL-RNA (P-RNA)

Qualitative conformational analysis of the entirety of conceivable hex o- and pentopyranosyl oligonucleotide systems derived from the diaster eoisomeric aldohexoses (CH2O)6 and aldopentoses (CH2O)5 predicts the e xistence of a variety of pairing systems which have not been experimen tally investigated so far. In particular, the analysis foresees the ex istence of a ribopyranosyl isomer of RNA ('p-RNA'), containing the pho sphodiester linkage between the positions C(4') and C(2') of neighbori ng ribopyranosyl units. Double strands of p-RNA oligonucleotides are e xpected to have a linear structure and to show purine-pyrimidine and p urine-purine (Watson-Crick) pairing comparable in strength to that obs erved in homo-DNA. Experimentally, synthetic beta-D-ribopyranosyl (4'- 2')-oligonucleotides derived from adenine and uracil confirm this prog nosis: adenine-uracil pairing in p-RNA duplexes is stronger than in th e corresponding RNA duplexes. Importantly, adenine in p-Ribo(A8) does not show (reverse-Hoogsteen) self-pairing, in sharp contrast to its be havior in the homo-DNA series. The sheer existence of strong and selec tive pairing in a system that is constitutionally isomeric to RNA and can be predicted to have a linear structure has implications for the p roblem of RNA's origin. In this context, a comprehensive experimental study of the pairing properties of p-RNA, of its potential for constit utional assembly, self-replication, and intra-duplex isomerization to RNA seems mandatory.