Synthetic cADPR analogues may form only one of two possible conformational diastereoisomers

Abstract

Cyclic adenosine 5′-diphosphate ribose (cADPR) is an emerging Ca2+-mobilising second messenger. cADPR analogues have been generated as chemical biology tools via both chemo-enzymatic and total synthetic routes. Both routes rely on the cyclisation of a linear precursor to close an 18-membered macrocyclic ring. We show here that, after cyclisation, there are two possible macrocyclic product conformers that may be formed, depending on whether cyclisation occurs to the “right” or the “left” of the adenine base (as viewed along the H-8 → C-8 base axis). Molecular modelling demonstrates that these two conformers are distinct and cannot interconvert. The two conformers would present a different spatial layout of binding partners to the cADPR receptor/binding site. For chemo-enzymatically generated analogues Aplysia californica ADP-ribosyl cyclase acts as a template to generate solely the “right-handed” conformer and this corresponds to that of the natural messenger, as originally explored using crystallography. However, for a total synthetic analogue it is theoretically possible to generate either product, or a mixture, from a given linear precursor. Cyclisation on either face of the adenine base is broadly illustrated by the first chemical synthesis of the two enantiomers of a “southern” ribose-simplified cIDPR analogue 8-Br-N9-butyl-cIDPR, a cADPR analogue containing only one chiral sugar in the “northern” ribose, i.e. 8-Br-D- and its mirror image 8-Br-L-N9-butyl-cIDPR. By replacing the D-ribose with the unnatural L-ribose sugar, cyclisation of the linear precursor with pyrophosphate closure generates a cyclised product spectroscopically identical, but displaying equal and opposite specific rotation. These findings have implications for cADPR analogue design, synthesis and activity.