Oligonucleotides are finding an extremely large number of applications in molecular biology and diagnostics, and might become a very selective class of drugs for the treatment of a vast palette of diseases. Oligonucleotides are polyanions that exert their specific activity following hybridization to a complementary sequence borne by another polyanionic nucleic acid. As drug candidates, they must also be capable of crossing the anionic cell membrane. Simple electrostatic considerations imply that hybridization energy and cell binding could benefit from the addition of cationic groups to the oligonucleotide structure. To reach this goal, many synthetic approaches for introducing ammonium or guanidinium residues into oligonucleotides have been explored: phosphate backbone replacement, ribose or nucleic base modification and end conjugation of a polycation. However, hybridization specificity, nucleic acid-processing enzyme activity and metabolite toxicity concerns all point to the diblock approach, in which the polycation is appended to an otherwise natural oligonucleotide, as the best solution. Unfortunately, stepwise automated synthesis of oligonucleotide–cationic peptide conjugates is not yet routine. Unfortunately, conjugation chemistry between preformed large blocks is not straightforward, especially in water, where “super” zwitterions raise intractable solubility, purification and characterization problems. Moreover, molecular biology and diagnostics applications require fast and straightforward synthesis of any given base sequence linked to any ACHTUNGTRENNUNGorganic cation length. With these considerations as guidelines, we have focused our efforts towards developing a high-yield automated synthesis of oligonucleotide–oligocation conjugates that makes exclusively use of classical phosphoramidite chemistry. In addition to the four A-, T-, Gand C-protected trityl phosphoramidite vials, a linear polyamine N-protected trityl phosphoramidite vial was plugged into the oligonucleotide synthetizer machine to allow online, computer-driven, synthesis (Scheme 1). Decamer oligonucleotides bearing a global charge of up to +9 were synthesized in good yield and characterized by ESMS and PAGE. Duplex melting temperature measurements showed that stability increases up to DTm=45 8C for the largest oligocation moiety. In order to avoid potential toxicity of metabolites, the cationic synthon was based on the largest naturally occurring polyamine, namely spermine. Spermine is present at millimolar concentration in cells, and end-alkylation seems harmless. Amine protection followed by a,w-bishydroxylalkylation led to a diol compatible with oligonucleotide synthesis (Scheme 2). Classical dimethoxytrityl (DMT)/phosphoramidite elongation chemistry was implemented together with base-labile trifluoroacetyl (TFA) protecting groups. The spermine synthon 1 was obtained in eight steps with an overall yield of 12%. A series of decamer oligonucleotides of identical sequences N10= CACCGTAGCG appended with increasing numbers of spermine residues S was synthesized on the micromolar scale on an Expedite 8909 oligonucleotide synthetizer. Individual coupling yields were determined by spectrophotometric determination of the total amount of DMT released during each step. For the nucleoside phosphoramidites, typical coupling yields over 97% were obtained under standard conditions, whereas coupling of 1 had to be optimized to finally reach 90– 96% routinely. The synthesis was stopped with the last DMT group remaining on the oligomer to facilitate separation from truncated sequences. Oligomer cleavage from the solid support and full deprotection were performed under standard aqueous ammonia conditions. Oligomers were purified by reversed-phase HPLC or by using a Poly-Pak II cartridge. DMT was removed by using either standard acetic acid conditions or 2% aqueous TFA to give 80–250 nmol decaoligonucleotide– spermine conjugates N10Sn with up to six spermines [a] B. Pons, Dr. M. Kotera, Dr. G. Zuber, Dr. J.-P. Behr Laboratoire de Chimie G!n!tique associ! au C.N.R.S. Universit! Louis Pasteur de Strasbourg, Facult! de Pharmacie B. P. 24, 67401 Illkirch (France) Fax: (+33)390-244-306 E-mail : kotera@bioorga.u-strasbg.fr behr@bioorga.u-strasbg.fr Supporting information for this article is available on the WWW under http://www.chembiochem.org or from the author. Scheme 1. Synthesis of oligonucleotide–oligospermine conjugates NmSn.