Oligonucleotides are synthesised almost exclusively via the solid-supported phosphoramidite method. However popular this method may be, the expensive reagents used in large excess during the synthesis as well as the large amounts of organic and aqueous solvents and purification steps makes the scale-up of oligonucleotide synthesis costly and possibly harmful to the environment. The therapeutic use of anti-sense oligonucleotides (ASOs) is hindered by their susceptibility to nuclease catalysed hydrolysis and to overcome this problem ASOs have been modified commonly by the introduction of a phosphorothioate backbone. This research aims to provide a better understanding of some of the more problematic stages of the synthesis cycle, the formation of the sulfurizing agent and sulfurisation of inter-nucleotide phosphite linkages, in order to make this method more sustainable and efficient.
The investigation of the activation, alcoholysis and hydrolysis of the phosphoramidites 2´-methoxy-5´-O-DMT-uridine 3´-CE phosphoramidite (UAm) and di-tert-butyl N,N-di-isopropyl phosphoramidite (DBAm) using several tetrazole activators found that complete conversion of the phosphoramidite UAm to products required an excess of activator and that this was due to the generation of di-isopropyl amine during coupling. Conductivity measurements show that the amine deprotonates the acidic activator and that the ammonium and tetrazolide ions that are subsequently formed strongly ion pair (Kip = 6540 M-1) removing free activator from solution. The tetrazole-catalysed reaction of phosphoramidites with oxygen nucleophiles was found to be first order with respect to phosphoramidite and activator and the nucleophilic displacement of the di-isopropyl amine group by the tetrazoyl group at phosphorus is rate-limiting.
Investigation into the 3-picoline-catalysed ageing of the sulfur transfer reagent phenylacetyl disulfide (PADS) has shown that the process is overall second order and is proportional to the concentration of PADS and 3-picoline. Deuterium exchange experiments show that ageing proceeds via abstraction of the methylene CH2 protons of PADS via an E1cB-type decomposition of the PADS molecule generating a disulfide anion and a ketene by-product which was trapped using an intra-molecular [2+2]-cycloaddition reaction. Mass spectrometry data shows that disulfide anions act as nucleophiles with PADS molecules to generate polysulfides which are the active sulfur transfer reagents in aged PADS solutions. Using pyridines that are less basic than 3-picoline causes the rate of degradation of PADS to become slower, indicating the possibility that the rate-limiting step of this process is the generation of the disulfide anion.
The rate of sulfurisation of phosphites by both ‘fresh’ and ‘aged’ PADS in the presence of 3-picoline was found to be first order with respect to phosphite, PADS and 3-picoline at low concentrations of each. However, the rate of the reaction becomes independent of base when using aged PADS in the presence of high 3-picoline concentration. Brönsted correlations for the sulfurisation of alkyl phosphites using fresh PADS give a βnuc value of 0.51, consistent with a mechanism involving nucleophilic attack by the phosphite on the disulfide bond of PADS to generate a phosphonium ion intermediate. This degrades to the phosphorothioate product via a base-catalysed mechanism which has been confirmed by removal of the methylene protons from the PADS molecule. Comparison of the βnuc values seen when altering the pKa of the pyridine catalyst used shows that the rate of the reaction of fresh PADS is much more sensitive to the pKa of the pyridine than is aged PADS (βnuc = 0.43 and 0.26 for fresh and aged PADS respectively). This suggests that in the case of aged PADS, the phosphite attacks the sulfur atom adjacent to the carboxylate group in the polysulfide chain. This generates a phosphonium intermediate which can be broken down via a much more facile S-S bond fission, as opposed to the C-S bond fission as seen in when using fresh PADS.
Available under License Creative Commons Attribution Non-commercial No Derivatives.
Download (4MB) | Preview
Downloads
Downloads per month over past year