148 



J. BADDILEY 



derivatives which yield the corresponding 5'-iodo derivatives by reaction 

 with sodium iodide. These conversions, like those already discussed for the 

 trityl derivatives, are consistent with furanose ring structures. 



-0 



OH ^^ 



MeCMe 



/\ 

 



•-CH-CH-CH-CH-CH.-OH 



N, 



-0- 



2',3'-Isopropylidene inosLne 



MeCMe 



/\ 

 O 



LCH-CH-CH-CH-CH,-OTs 

 2',3'-Isopropylidene-5'-tosylinosine 



.N 



/. Configuration of the Glycosidic Linkage 



The behavior of ribonucleosides towards periodate has been used in a 

 chemical determination of the configuration of their glycosidic linkage and, 

 at the same time, provides confirmatory evidence for the location of the 

 sugar at position 9 in the purine nucleosides. 



The dialdehyde obtained by the action of periodate on adenosine is 

 identical with the dialdehyde from 9-D-mannopyranosyladenine.''« This 

 mannosyl compound was prepared by the synthetic route outlined on page 

 153 and so must be a 9-glycoside. It follows that adenosine is also a 9-gly- 

 coside. 



-0- 



OH 

 -CH-CH-CH-CH-CH-CH,OH 



,N. ^N, OH OH 



C 



— o- 



"1 



nQC > 



NH, ^ 



CH-CHO 0HC-CH-CH2-0H 





9-D-Mannopyranosyladenine 



NH, 



+ HCOoH 



Periodate oxidation of a glucosyladenine also gives this dialdehyde. "^ 

 This glucoside almost certainly has the jS-configuration since it was synthe- 

 sized from a-acetobromoglucose, and it is reasonable to expect inversion 

 during the formation of glucosides from this compound. The optical proper- 

 ties of the oxidation product are consistent with this view. It follows that 

 the natural nucleoside must also have the /3-configuration. 



In the same way, the dialdehydes obtained by periodate oxidation of 

 uridine and cytidine are identical with those from synthetic samples of 



" B. Lythgoe, H. Smith, and A. R. Todd, /. Chem. Soc. 1947, 355. 

 " J. Davoll, B. Lythgoe, and A. R. Todd, /. Chem. Soc. 1944, 833. 



