Target Choice and Oligonucleotide Design

The choice of the target mRNA selected for inhibition by antisense ODNs is dictated by the biology of a particular disease process and by the ability to predict the effects that may be achieved by inhibiting the expression of a particular cancer gene. For example, the bcr/abl ( BCR-ABL1) transcripts of chronic myelogenous leukemia (CML) cells serve as an ideal target because of the role of the BCR/ABL oncoprotein in hematopoietic cell transformation and in the maintenance of the leukemic phenotype. Since bcr/abl genes are only found in leukemic cells, targeting their mRNA transcripts might also provide the advantage of a specific effect against tumor cells. Targeting  BCL-2 mRNA in lymphomas with the t(14;18) translocation is appropriate not only for the disease-causing effect of  BCL-2 expression but also for the importance of interfering with antiapoptotic pathways in drug response. Thus,  BCL-2 antisense ODNs, in addition to their direct effects on target cells, may also sensitize these cells to chemotherapeutic agents that promote  apoptosis. In most published studies, the sequence of the ODN targeting mRNA transcripts of a disease-causing gene is selected empirically with a preference for the mRNA transcription initiation sequence or the nucleotides surrounding the translation initiation codon. However, there are now novel approaches of oligonucleotide design based on the use of the DNA chip ( microarray (cDNA) technology) technology and hybridization with labeled RNA to dissect accessible sites in the mRNA tertiary structure.

Early investigations of ODN-targeting of growth-regulatory mRNA transcripts employed natural DNA; the realization that natural ODNs are rapidly cleaved by endo-and exonucleases led to the development of nuclease-resistant ODNs by modification of the internucleotide linkages. The most common modification is the replacement of the nonbridging oxygen atoms in the phosphate group with a sulfur group. This type of modification generates the so-called phosphorothioate ODNs extensively used in preclinical studies and in phase I clinical trials. The phosphorothioate modification results in several desirable properties such as nuclease resistance, water solubility, and activation of RNase H. Nevertheless, it presents also certain disadvantages, including impaired uptake caused by the polyanionic nature of phosphorothioate ODNs, and nonsequence-dependent effects attributed to charge interactions between phosphorothioate ODNs and proteins in the extracellular environment, on the cell surface, and intracellularly. A number of strategies have been utilized to minimize the undesirable effects of the phosphorothioate ODNs while preserving their useful properties. Since these modified phosphorothioate ODNs have not been tested sufficiently in vitro and in in vivo models, we will focus on first-generation phosphorothioate ODNs with regard to delivery, subcellular trafficking, pharmacodynamics, and applications in mouse models and in humans.

Delivery, Subcellular Trafficking, and Pharmacodynamics of ODNs

Native and phosphorothioate ODNs are polyanionic molecules that cross cell membranes inefficiently. There is evidence that ODN uptake is time- and concentration-dependent. Below a concentration of 1 mmol/l, uptake of phosphorothioate ODNs is predominantly via a receptor-like mechanism, while fluid-phase endocytosis appears to predominate at higher concentrations. Several receptor-like proteins potentially involved in ODN uptake have been identified, but evidence that they are responsible for ODN uptake is still lacking. In culture, ODN uptake may be enhanced by a number of procedures directly or indirectly modifying the permeation properties of the ODNs. The most common methods are  electroporation and streptolysin treatment which result in physical disruption or enhanced permeabilization of cell membranes. Such procedures are impractical for in vivo studies, which, at present, rely on the administration of naked DNA. Inside the cells, ODNs accumulate in vacuoles, presumably endosomes and lysosomes, and slowly redistribute to the cytoplasm and nucleus where they may interact with their target mRNA molecules. Accordingly, strategies that promote the release of ODNs from endosomal structures may enhance the ODN’s antisense effects. Pharmacokinetics and metabolism of antisense ODNs have been investigated in a variety of animal systems and also in few human trials. In most reports, the analyses were carried out after intravenous or intraperitoneal administration. Approximately 30% of the injected dose is excreted in the urine within 24 h and intact material is detected in most tissues up to 48 h, and up to 7 days in liver and kidney, the organs where most ODNs accumulate. Plasma clearance is biphasic with an initial half-life of 15–25 min and a second half-life of 20–40 h. Potential toxic effects of ODN administration have been reported in rodents and in primates. Mice receiving high doses of phosphorothioate ODNs show decreased platelet counts probably related to the polyanionic charge of the ODNs. Cardiovascular toxicity, rapid peripheral vasodilatation, and death have been reported in monkeys. These effects were noted after rapid bolus administration of large doses, while slow infusion of similar doses appeared to be well tolerated.

Clinical Applications of Antisense ODNs in Hematological Malignancies

Early clinical experiences with antisense ODNs have been reported by groups targeting oncogene or apoptosis regulators. These studies were based on encouraging antitumor effects of systemically delivered ODNs in mice injected with leukemia or lymphoma cells of human origin. For example, the disease process induced by Philadelphia1 leukemia cells was suppressed by the systemic delivery of antisense ODNs targeting  BCR-ABL or c- myb transcripts. In particular, the antileukemia effects of the bcr/abl antisense ODNs was markedly enhanced by the combination with low doses of  cyclophosphamide. In the context of  chronic myelogenous leukemia (CML), oligodeoxynucleotides targeting  BCR-ABL or c-myb mRNA have been used as marrow purging agents in the chronic as well as accelerated phase of the disease. Eight patients with CML in advanced phase were subjected to autologous bone marrow transplantation after bone marrow purging with bcr/abl antisense ODNs.

Infusion of the ODN-treated cells was followed by prompt engraftment and hematologic reconstitution in all patients. Evaluation of antileukemia effects by standard cytogenetic analysis and fluorescence in situ hybridization showed a complete karyotypic response in two cases and a minimal or no response in the other six. Survival of transplanted patients exceeded three years in some cases, but it is not clear that the protocol had therapeutic efficacy.

However, lack of toxicity, prompt hematopoietic reconstitution, and karyotypic response in some cases are all encouraging observations for designing additional clinical trials. In a different study, eight CML patients were subjected to bone marrow transplantation using autologous hematopoietic progenitor (CD34⁺) cells that were pretreated with antisense ODNs targeting the mRNA transcripts of the c-myb gene, a key regulator of normal and leukemic hematopoiesis. After transplantation, seven of eight patients engrafted. Of these, four patients showed 80–90% normal metaphases 3 months post autologous bone marrow transplant, suggesting that the antisense ODNs treatment eliminated the majority of Philadelphia1 CML cells. These patients showed hematologic improvement during the period (6–24 months) following the bone marrow transplant.

Eighteen patients with refractory acute myelogenous leukemia were also treated by continuous infusion of c-myb antisense ODNs at dose levels ranging from 0.3 to 2.0 mg/kg/day for 7 days. There was no treatment-related toxicity, but only one patient showed a therapeutic response.

Studies in a mouse model of lymphoma with the t(14;18) associated with BCL-2 overexpression have demonstrated dose-dependent disease eradication in most mice treated with antisense ODNs targeting a segment of the BCL-2 open reading frame.

On the basis of these preclinical data, BCL-2 antisense ODNs were given via a continuous subcutaneous infusion for 2 weeks to lymphoma patients with high BCL-2 expression and resistant to conventional therapies. Therapeutic responses assessed by computed tomography scanning were demonstrated in six out of nine patients. The specificity of the antisense effects was validated by showing a decrease in BCL-2 levels in lymph node aspirates taken at different times after initiating the antisense ODN therapy.

Prospects for Antisense DNA Therapy

Continuous advances in understanding the genetic basis of tumorigenesis are leading to the identification of an ever-increasing number of gene targets for antisense ODNs-based therapies. Most disease-causing genes identified by molecular genetics belong to the class of cell cycle and apoptosis regulators. Accordingly, antisense ODNs may be used individually or in combination against these targets; moreover, antisense ODNs might be combined with conventional chemotherapeutic drugs to enhance apoptosis susceptibility of tumor cells. It might therefore be conceivable that various therapeutic strategies involve ODNs that target tumor-causing genes. Considering this, the success of antisense ODNs-based antitumor therapies is likely to depend on the development of antisense ODNs as effective therapeutic agents. Delivery of sufficient amounts of ODNs to tumor cells remains an important problem. Administration procedures that may guide ODNs to tumor cells are of great interest; in a in vitro study, neuroblastoma cells were targeted on the basis of the expression of the neuroectodermal-specific GD2 disialoganglioside by antibody-coupled neutral liposomes encapsulated with c-myb antisense ODNs. Although it is unknown if such an approach may function in vivo, this is an example of a potentially useful strategy. The delivery of sufficient amounts of ODNs to tumor cells does not guarantee that they will find the mRNA targets once inside the cells. Thus, methods promoting the intracellular trafficking of ODNs, to enhance the access to as many as possible mRNA target molecules, will be invaluable for efficacious ODNs-based therapies. The development of novel classes of ODNs with fewer nonspecific interactions to nontarget molecules will also improve the efficacy of antisense ODN therapies.

If the goal of making effective ODN drugs is to be achieved, these and other problems need to be addressed.

While the principles underlying ODNs-based therapies remain highly attractive, the field of DNA therapeutics is now at a crossroad where rigorous validation in clinical trials is necessary.