ABSTRACT

This PhD dissertation is based on work performed at the Department of Tumour Endocrinology, Institute of Cancer Biology, Danish Cancer Society, and at the Department of Medicine and Pharmacology, Columbia Presbyterian Hospital, New York, from March 2000 to June 2003.

This work was funded by the Danish Research Agency as a part of the SUE-programme. The project involved collaboration between three partners: Exiqon A/S, Institute of Chemistry, University of Copenhagen and University of Southern Denmark and the Department of Tumour Endocrinology, Danish Cancer Society.

The aim of this basic research project has been to investigate the utility of locked nucleic acid (LNA) as antisense molecule and the potential of LNA to inhibit the growth of cancer cells.

The antisense (AS) technique holds the potential to specifically down-regulate the expression of any selected gene. Thus, in principle it should be possible to design an antisense drug to target any mRNA encoded by the human genome and potentially have specific treatment for any disease caused by overexpression of a single (or a few) gene(s).

LNA is a novel class of nucleic acid analogues, which exhibits unprecedented thermal stability towards complementary LNA, DNA and RNA. Furthermore, LNA obeys Watson-Crick base pairing rules, is stable in serum and can be taken up by mammalian cells. In vivo, LNA displays low toxicity and an LNA antisense oligonucleotide (AON) has been demonstrated to inhibit tumour growth in a murine xenograft model. All together, these characteristics of LNA make it a very promising new player in the antisense field of research.

Firstly, we optimized the uptake in the human breast cancer cell line MCF-7 and studied the intracellular localization of oligonucleotides. The results have been evaluated by fluorescence microscopy showing that LNA is effectively taken up in MCF-7 cells with a predominant nuclear localization. The uptake of oligonucleotide has been quantified by flowcytometry and shows that approximately 60-70% of the MCF-7 cells take up oligonucleotide.

Secondly, we evaluated the potential of various LNA AONs to down-regulate the expression of five proteins in cancer cells, namely the Estrogen Receptor α (ER α ), p21, Survivin, Bcl-xL and Bcl-2. These studies strongly support that LNA oligonucleotides are more potent than the currently used phosphorothioate AONs. The exact design of LNA-containing AONs appear to be crucial, since fully modified LNA and LNA/DNA mix-mer oligonucleotides were ineffective as AONs in our experiments. However, a consistent result from this study is that LNA gap-mer AONs are highly efficient antisense molecules.

Thirdly, to test if LNA AONs are efficient as an agent to inhibit growth of cancer cells we chose a model system with estrogen dependent growth of MCF-7 cells. The growth promoting effects of estrogen is based on the binding to ER α . Thus, AON mediated down-regulation of ER α prevents estrogen from exerting its growth stimulatory effects. Compared to an iso-sequential phosphorothioate AON, an LNA containing oligonucleotide is approximately ten times as potent in suppression of estrogen dependent growth of MCF-7 cells.

This study is a basic research project, in which the ability of a new nucleic acid analogue has been examined for its antisense properties in cancer cells. It is the first systematic study to describe how LNA AONs should be designed in order to achieve the most efficient down-regulation of endogenous proteins encoded by the targeted mRNA. We hope that these results have increased the understanding of how to apply LNA as antisense molecules and that this knowledge can be utilized by other researchers. Ultimately, we hope that the data presented in this dissertation may eventually form the basis upon which LNA antisense oligonucleotides can be developed as cancer therapeutic drugs.