A prebiotically plausible synthesis of ribonucleotides is required to support the "RNA world" hypothesis. 1 The conversion of D-ribose-5-phosphate 1 to the aminooxazoline 2, and thence to R-Dcytidine-5′-phosphate 3, by sequential treatment with cyanamide and cyanoacetylene was demonstrated over 30 years ago (Scheme 1), 2 but no prebiotic synthesis of 1 has been reported.Not only is ribose notoriously difficult to synthesize as anything other than a component of a complex mixture, 3 but it is unstable, 4 and prebiotic phosphorylation furnishes the 1-, 2-, and 3-phosphates and not the 5-phosphate. 5 D-Pentose-5-phosphates, including 1, are theoretically accessible by aldol reaction of glycolaldehyde 4 and D-glyceraldehyde-3-phosphate 5, but such an approach is likely to be thwarted experimentally by the instability of 5 under the alkaline conditions required to enolize 4 (Scheme 2).The enolate of 5 is easily formed by intramolecular general base catalysis by the phosphate dianion in mild alkali and eliminates phosphate rapidly. 6 Expectations that this E1cb irreversible elimination behavior of 5 would prevent an aldol synthesis of 1 were realized at the outset of this study. It was found that incubation of a D 2 O solution of 4 and 5 for a day at pD ) 9.4 resulted in significant elimination of phosphate from 5 even before 4 had undergone significant enolization (as evidenced by the lack of exchange of the CH 2 (and CH) protons of 4 for deuterons). Incubation of a solution of 4 and 5 at neutral pD did not result in elimination of phosphate from 5, but neither was aldolization evident.We have recently found that 2-aminooxazole 6, a condensation product of glycolaldehyde and cyanamide, 7 reacts with glyceraldehyde to give pentose aminooxazolines in a remarkable process that is essentially quantitative and is highly stereoselective for the ribo-and arabino-products. 8 We thus wondered if it might be possible to produce 2 from 6 and 5 (Scheme 3). However, in addition to the possibility that elimination from 5 might still occur faster than the addition of 6, there was also the chance that the stereoselectivity found in the earlier reaction of 6 with glyceraldehyde might be altered by the presence of the phosphate group in 5. 9 To investigate this potential reaction, we incubated a solution of 5 and 6 (both 69 mM) in H 2 O at pH ) 7 and room temperature. After 2 days, an aliquot was removed and lyophilized, and the residue was dissolved in D 2 O for 1 H NMR analysis. A series of doublets (J ) 5-6 Hz) in the range of δ ) 5.4-6.0 ppm suggested that pentose aminooxazoline-5′-phosphates had been formed in good yield ( Figure S1, Supporting Information). 10 One stereoisomer was dominant to the extent that it exceeded the sum of the other stereoisomers. By sample spiking with an authentic standard of 2 prepared from 1, we were able to determine that this predominant stereoisomer was 2. To characterize the other products, the mixture was treated with alkaline phosphatase, divided into aliquots, and then separately spiked wit...