DOI: 10.1002/cbic.201200611 Functionalization of the 3’-Ends of DNA and RNA Strands with N-ethyl-N-coupled Nucleosides: A Promising Approach To Avoid 3’-Exonuclease-Catalyzed Hydrolysis of Therapeutic Oligonucleotides Montserrat Terrazas,* [a] Adele Alagia, [a] Ignacio Faustino, [b, c] Modesto Orozco, [b, c] and Ramon Eritja* [a] Introduction Over the past three decades, the inhibition of gene expression by synthetic oligonucleotides has been a widely explored field. [1] Relevant established classes of oligonucleotide agents include antisense oligonucleotides, short interfering RNAs (siRNAs), [2] aptamers, [3] and DNA/RNAzymes. [4] Unfortunately, the application of oligonucleotides as therapeutic agents in vivo faces some key problems. [5] One of the most important ones is that ordinary DNA and RNA are highly vulnerable to serum nucleases; this leads to short half-lives of the oligonu- cleotides in serum. Much research effort has been focused on overcoming these limitations. [2, 5, 6] In particular, major efforts have been made to increase the biostability of oligonucleotide-based agents with- out compromising their biological activity. [5] This research has crystallized in the synthesis of a wide variety of modified oligo- nucleotides that contain chemical modifications involving the sugar ring [7] and/or the phosphate backbone. [7a, g, 8] Among them, the phosphorothioate modification [9] provides high levels of nuclease protection and has been widely and success- fully employed. [5] In contrast to the large effort made to create modified backbones with improved biostability, there has been little exploration of the potential impact of nucleobase modifi- cations. [10] An alternative solution is the incorporation of modified nu- cleotides in the 3’-dinucleotide overhangs of siRNAs. [11] Incor- poration of two peptide nucleic acid (PNA) units has been found to improve the biostability of these oligonucleotides without impairing the siRNA activity. [11a] Positive results have also been obtained for carbohydrate conjugates, [11b] for siRNAs bearing C-5 polyamine-substituted nucleosides, [11c] and for siRNAs containing terminal bis(hydroxymethyl)benzene [11d] and biaryl units. [11e] Many of the studied modifications increased the biostability of oligonucleotides, but in some cases modified oligonucleotides were found to have negative effects on activi- ty. [12] Thus, the search for efficient oligonucleotide chemistries remains a focus of continued study. We have created and analyzed a new class of modification aimed at increasing the stability of oligonucleotides against 3’- exonucleases (the predominant nuclease activity present in serum [13] ) without affecting biological action. In particular, ra- tional design showed the possibility of blocking the hydrolytic activity of 3’-exonucleases by creating a new nucleotide scaf- fold characterized by its lack of a phosphodiester bond linking the two 3’-terminal nucleotide building blocks. Our approach is based on the replacement of the two 3’-terminal nucleotides The development of nucleic acid derivatives to generate novel medical treatments has become increasingly popular, but the high vulnerability of oligonucleotides to nucleases limits their practical use. We explored the possibility of increasing the sta- bility against 3’-exonucleases by replacing the two 3’-terminal nucleotides by N-ethyl-N-coupled nucleosides. Molecular dy- namics simulations of 3’-N-ethyl-N-modified DNA:Klenow frag- ment complexes suggested that this kind of alteration has negative effects on the correct positioning of the adjacent scis- sile phosphodiester bond at the active site of the enzyme, and accordingly was expected to protect the oligonucleotide from degradation. We verified that these modifications conferred complete resistance to 3’-exonucleases. Furthermore, cellular RNAi experiments with 3’-N-ethyl-N-modified siRNAs showed that these modifications were compatible with the RNAi machinery. Overall, our experimental and theoretical studies strongly suggest that these modified oligonucleotides could be valuable for therapeutic applications. [a] Dr. M. Terrazas, A. Alagia, Prof. Dr. R. Eritja Institute for Research in Biomedicine (IRB Barcelona) and Institute for Advanced Chemistry of Catalonia (IQAC) Spanish Research Council (CSIC) Cluster Building, Baldiri i Reixac 10, 08028 Barcelona (Spain) E-mail : montserrat.terrazas@irbbarcelona.org recgma@cid.csic.es [b] I. Faustino, Prof. Dr. M. Orozco Joint IRB–BSC Program on Computational Biology Institute for Research in Biomedicine (IRB Barcelona) Baldiri i Reixac 10, 08028 Barcelona (Spain) [c] I. Faustino, Prof. Dr. M. Orozco Department of Biochemistry, University of Barcelona Diagonal 647, 08028 Barcelona (Spain) Supporting information for this article is available on the WWW under http://dx.doi.org/10.1002/cbic.201200611. # 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim ChemBioChem 2013, 14, 510 – 520 510 CHEMBIOCHEM FULL PAPERS