Phosphorothioate oligodeoxynucleotides: large-scale synthesis and analysis, impurity characterization, and the effects of phosphorus stereochemistry.

Abstract

Large-scale synthesis of phosphorothioate oligodeoxynucleotides on Tentagel using a 'batch mode' synthesizer and beta-cyanoethyl phosphoramidite coupling followed by sulfurization with bis(O,O-diisopropoxy phosphinothioyl) disulfide (S-tetra) provides stepwise yields of 98-99% and results in phosphorothioate oligodeoxynucleotides that are 93-97% pure, as determined by PAGE, after reverse-phase high performance liquid chromatography (RP-HPLC) and 'downstream' processing. The purity of phosphorothioate oligodeoxynucleotides synthesized on Tentagel is significantly higher than those synthesized on controlled pore glass. Electrospray ionization mass spectrometry of the n-1 impurity isolated by preparative PAGE was used to establish that the n-1 impurity is a heterogeneous mixture of all possible single-deletion sequences, relative to the parent phosphorothioate oligodeoxynucleotide, and results from minor, though repetitive, imperfections in the synthesis cycle. Acid-catalysed depurination was found to occur both during the synthesis and during the post-synthesis detritylation, following RP-HPLC. Studies of hybridization affinity and biological mechanism of action using independently synthesized n-1 phosphorothioate oligodeoxynucleotides relative to the 15 mer LR-3280 showed that, in this case, the majority of the n-1 sequences had more than a 10 degrees C decrease in melting temperature with sense RNA compared to the n-mer, and they did not cause detectable cleavage of RNA by RNase H in HL-60 human promyelocytic leukaemia cells. P stereoregular phosphorothioate oligodeoxynucleotides are not significantly more active than their stereorandom counterparts and thus their use in clinical studies seems unwarranted.