Polynucleotides, in addition to their natural role as carriers of genetic information, offer a wide array of unique chemistries that can be exploited for use as therapeutic and diagnostic reagents,1 and as highly selective materials for biorecognition2 where specific binding to target molecules is desired. Unfortunately, native polynucleotides are susceptible to ubiquitous endoand exonucleases present in many biological systems. In addition, the natural phosphodiester linkage is particularly labile to harsh conditions often found in abiotic environments including extremes in pH and temperature. Conversely, replacing the phosphodiester linkage with unnatural bonds can result in a wide array of functionalized polymers. This strategy has been exploited to some extent for the chemical derivatization of thymine and related bases and nucleosides with vinyl groups to give functionalized vinyl polymers for use as selective absorbents.3,4 Polynucleotide analogues have also been synthesized using a variety of other polymer backbones including poly(trimethyleneimine),5 poly(vinyl alcohol)s,6 poly(ethylenimine),7 and poly(vinylamine).8 Nucleic acid analogues possessing high biostability and affinity based on polypeptide backbones have also been developed.9 In all cases, polymers with a broad range of sizes (103 to 105 Da) and controlled hydrophobicity/hydrophilicity, chain flexibility, and stability were produced for specific applications including aptamers that selectively bind organic ligands10 and polymeric drugs for chemotherapy.6,11 Nearly all of the aforementioned synthetic schemes require multiple chemical steps that provide little selectivity in the attachment of the nucleoside or nucleoside derivative to the polymeric backbone. Selectivity, if necessary, is achieved by complex site-specific chemistries involving blocking and deblocking steps. Such selectivity may be critical in the preparation of polymers with side chains occupying a specific orientation around the polymer backbone, thereby resulting in unique properties. In the present work, we have developed a unique, simple, and effective technique to prepare novel polynucelosides with unnatural polymeric backbones and with high inherent selectivity. Specifically, we have used a two-enzyme coupled reaction system to generate nucleoside-based polyphenols. The resulting polymers may provide highly selective and stable materials for therapeutic, diagnostic, and materials applications. The synthetic approach is shown in Scheme 1. Thymidine was used as a model nucleoside and a trifluoroethyl ester derivative of p-hydroxyphenylacetic acid (1)12 was used as the phenolic derivative. Regioselective acylation of thymidine at the 5′-hydroxyl was achieved in nearly anhydrous CH3CN using the lipase from Candida antarctica. This was followed by polymerization of the phenolic nucleoside derivative catalyzed by the peroxidase from soybean hulls (SBP). The lipase-catalyzed formation of the thymidine 5′p-hydroxyphenylacetate (2) was facile and over 95% of a 50 mM solution of 1 in CH3CN was consumed in 18 h to give almost entirely monoester 2.13 Purification of the monoester resulted in two distinct products in a ratio of ca. 5:1 with overall 70% isolated yield.13 1H and 13C NMR confirmed that the major product was the 5′ester (1.10 g, 59% isolated yield based on conversion of 1). The minor product was presumably the 3′-ester (0.21 g, 11% isolated yield); thus, the enzymatic transformation in CH3CN was highly regioselective. Similar specificity was observed for the Pseudomonas cepacia lipase-catalyzed acylation of thymidine with vinyl butyrate in tetrahydrofuran.14 The resulting thymidine monoesters were soluble in CH3CN and dimethyl sulfoxide (DMSO). The reaction was essentially catalyzed in a solid phase; the concentration of thymidine dissolved in CH3CN did not exceed 3 mM. Hence, C. antarctica lipase is effective in catalyzing the efficient regioselective acylation of a poorly organic solventsoluble nucleoside in CH3CN. SBP-catalyzed polymerization of 2 was carried out in the presence of H2O2 in a solution of aqueous buffer (50 mM Na phosphate, pH 7.0) containing 60% (v/v) CH3CN following a procedure developed previously for polyhydroquinone synthesis.15 Conversion of 2 reached 65% (as determined by reversed-phase HPLC) and the reaction was terminated by centrifuging the suspension and separating the reaction components into soluble and insoluble fractions. The soluble fraction was precipitated by adding to neat CH3CN. Extensive washing with CH3CN followed by water gave an isolated yield of 65 mg (28% of reacted 2) of poly(thymidine 5′-p-hydroxyphenylacetate) (3). 1H NMR in DMSO-d6 showed a clear presence of the thymidine (essentially identical with that in 2); however, the phenolic peaks were significantly broadened, indicating polyphenol synthesis (data not shown). Molecular weight determination16 provided an Mn ) 21 700 and an Mw/Mn ) 1.2. DSC17 of the polymer indicated the presence of a Tg at 60 °C and a broad melting point * To whom correspondence should be addressed. † Department of Chemical and Biochemical Engineering. ‡ Present address: Chemical Technology Division, Oak Ridge National Laboratory, Oak Ridge, TN 37831-6226. § Division of Medicinal and Natural Products Chemistry. Scheme 1 941 Macromolecules 1998, 31, 941-943
Read full abstract