Towards the Synthesis of 1-Deoxy-1-nitropiperidinoses

In the course of the first of several attempts to elaborate methods for the synthesis of 1-nitropiperidinoses, lincosamine was transformed into lactam 6 via hemiacetal 1, lactone 2, amide 3, oxo amide 4, and its cyclic tautomer 5. Treatment of the N-Boc-protected lactam oxime 9, obtained from lactam 6, with brominating agents failed to provide the bromonitroso carbamate 10. The N-Boc-protected lactam 13 derived from 6 was reduced to hemiacetal 14, but the corresponding N-Boc-aminooxime did not tautomerise to the C(1)-hydroxylamine, and nitrone 17, a potential precursor of the nitropiperidine 12, was not formed. Oxidation of the anomeric azide 20 with HOF center dot MeCN failed to provide the expected nitropiperidine 21. The phosphinimines 22 derived from 20 did not react with O-3. In the next approach to 1-nitropiperidinoses, we treated the N-Boc-protected hemiacetal 25, obtained from the known gluconolactam 23 with N-benzylhydroxylamine. The resulting nitrone 26 exits in equilibrium with the anomeric N-benzyl-glycosylhydroxylamine that was oxidized to the anomeric nitrone 28. Ozonolysis of 28 led to the hemiacetal 25, resulting from the desired, highly reactive protected nitropiperidinose 29, that was evidenced by an IR band at 1561 cm(-1). Similarly to the synthesis of nitrone 26, reaction of the N-tosyl-protected hemiacetal 31 with N-benzylhydroxylamine and oxidation provided the anomeric N-benzylhydroxylamines 33 via the p-toluenesulfonamido nitrone 32. Their oxidation with MnO2 led to the anomeric nitrone 34. Ozonolysis of 34 as evidenced by H-1-NMR and ReactIR spectroscopy led to the highly reactive nitropiperidinose 35. Like 29, 35 was transformed during workup, and only the hemiacetal 31 was isolated. The similarly prepared lincosamine-derived nitrone 17 was subjected to ReactIR-monitored ozonolysis that evidenced the formation of the protected nitropiperidinose 12, but only led to the isolation of 14. The facile transformation of the nitropiperidinoses to hemiacetals is rationalised by heterolysis of the anomeric C,N bond, recombination of the ion pair, and denitrosation of the resulting anomeric nitrite by a nucleophile. Attempts to convert the 1-deoxy-1-nitropiperidinose 35 to uloses 43 by base-catalysed Michael additions or Henry reactions were unsuccessful.