Rapid synthesis of oligodeoxyribonucleotides VII. Solid phase synthesis of oligodeoxyribonucleo-tides by a eontinuous flow phosphotriester method on a kieselguhr-polyamide support

volume 10 Number 20 1982 Nucleic Acids Research Rapid synthesis of oligodeoxyribonucleotides VII. Solid phase synthesis of oligodeoxyribonucleotides by a eontinuous flow phosphotriester method on a kieselguhr-polyamide support Michael J.Gait , Hans W.D.Matthes, Mohinder Singh, Brian S.Sproat and Richard C.Titmas Laboratory of Molecular Biology, Medical Research Council Centre, Hills Road, Cambridge, CB2 2QH.UK Received 29 June 1982 A new kieselguhr-polydimethylacrylamide support has been used in a continuous flow, column system for solid phase synthesis of oligodeoxyribonucleotides by a phosphotriester procedure. Using only protected mononucleotides a 14-mer, 20-mer and 27-mer were assembled in high repetitive yield using a simple manually operated,bench top apparatus. INTRODUCTION In previous papers in this series we have developed the use of swellable polydimethylacrylamide gel copolymers as solid supports for oligodeoxyribonucleotide synthesis. In the most recent paper [1] we showed that 17-mers could be efficiently prepared using phosphotriester chemistry and dimer building blocks. Several other phosphotriester routes have also been described recently involving a diversity of solid phases, viz: polyacryloylmorpholide [2], polystyrene [3] and cellulose [4]. Such non-rigid materials (including polydimethylacrylamide gels) require gentle handling with solvents and reagents being added batchwise followed by gentle agitation and filtration. More efficient washing can be achieved by packing the solid phase into a column and passing solvents through continuously. This has been used recently with silica gel in conjunction with phosphite [5] and phosphoroamidite [6] chemistry. Although attempts are presently being made to use polystyrene [7], silica gel [8] and glass beads [9], especially on a small scale in columns with phosphotriester chemistry, conventional polyamide gels are too swollen to be handled in this way [10]. Recently the preparation and use in peptide synthesis of a new kieselguhr-polydimethylacrylamide composite was described, specifically designed for use in continuous flow systems [11]. In a preliminary communication we showed [12] that this material was particularly suitable for oligodeoxyribonucleotide synthesis. The methods are now © IR L Press Limited, Oxford, England. 0305-1048/82/1020-6243S 2.00/0 6243 ABSTRACT Nucleic Acids Research presented in more detail and in particular we describe some significant improvements to apparatus, monomer purification and oligonucleotide isolation conditions. The efficient synthesis of a 14-mer, 20-mer and 27-mer is described in each case using only monomer building blocks. DISCUSSION AND RESULTS The solid phase The composite support, which is commercially available, consists of particles of fabricated, inert kieselguhr having extremely large pores of o several thousand A diameter, in which the polydimethylacrylamide gel has been The macroporous support packed in a glass column exerts negligible back-pressure during solvent flow. Thus far only one loading of functional group has been investigated (0.108 mmoleg . ) These are sarcosine methyl ester groups and attachment of the first deoxynucleoside residue is achieved by methods similar to those previously described [13] (Figure 1). Essentially two rounds of peptide synthesis are carried out to obtain a double glycine spacer, followed by formation of a 3'-0-succinamido linkage to the first deoxynucleoside residue by an analagous coupling step. All reactions can be followed by the ninhydrin test, and final resin quantitated for glycine and dimethoxytrityl content. Reproducible loadings of deoxynucleoside of 80-90 ymole g of final support are achieved. |3 CH-OCO.CHjN-® 2) Fmoc Gly anhydride then piperidine 3) Fmoc Gly anhydride then piperidine Thy | succinate | Gly | Gly Figure 1 6244 | ethylene | sarcosine diamine prepared in situ. Nucleic Acids Research Oligonucleotide assembly For the assembly reactions we have constructed a simple apparatus from low cost, commercially available components, consisting principally of solvent bottles, a six-way rotary solvent selector valve, small glass column and assorted Teflon tubes and connectors. Originally solvent delivery was by a pump [12] but we have now replaced this by a more convenient gas pressure delivery (Figure 2). Once assembled the only manipulations required are operation of the single six-way valve for solvent selection and injection of activated nucleotides via a septum port. The basic phosphotriester route as applied to solid phase synthesis Only two chemical The first is removal of 5'-O- dimethoxytrityl groups using 10% trichloroacetic acid in chloroform [14]. Deprotection is reproducibly complete within three minutes and very little unwanted depurination is observed. During deprotection a bright orange colour (the dimethoxytrityl cation) is released and serves as a useful qualitative guide to the efficiency of assembly. The second reaction is coupling of appropriately N-protected 5'-Odimethoxytrityl-2'-deoxynucleoside-3'-0-2-chlorophenyl phosphates to the support. In the past we have used dinucleotides rather than mononucleotide units in the coupling reactions in order to reduce the number of synthesis cycles [13]. However, the use of monomers has the advantage that only four (rather than 16) coupling units are necessary for any synthesis. In further examination of the use of mononucleotides we noticed large variations in Solvent bottles Argon Nitrogen Waste bottle Figure 2 6245 (Figure 3) has been described several times before [1,13]. reactions are necessary per synthesis cycle. Nucleic Acids Research (MeC r V ^ f OC-CH CH C-NH-© 6 2 2 6 oCI3CCOOH/CHCI3 ' . A . ,B2 + O con*ng agent f O Vo-p-CTEt,NH* CMSNT) a 6 (Me O) Tr O 6 6 (B-T. teA.bzC.ot ibG ) Figure 3 coupling yields when different batches of the same deoxynucleotide derivative were used. We concluded that small amounts of reactive impurities were being carried through in preparation of the monomers by standard methods. Such impurities become magnified by virtue of the use of large excesses of nucleotide in solid phase reactions and presumably react preferentially to cap off the oligonucleotide chains at their 5'-hydroxyl groups [15]. After investigating alternative purification techniques we have found that consistently high purity can be obtained by chromatographing the monomers as their triethylammonium salts on columns of Kieselgel 60H by the short column method. Monomers purified in this way give significantly improved results in solid phase synthesis. Hence for 60 mg of composite support (5 pinole) a solution of monomer (40 pmole) in anhydrous pyridine (0.35 ml) is pre-activated by addition to mesitylenesulphonyl-3-nitro-1,2,4-triazole (MSNT) (200 pinole) , injected on to the column and left in contact with the support for 45 min. A large excess of coupling agen (...truncated)