Increased stability and specificity through combined hybridization of peptide nucleic acid (PNA) and locked nucleic acid (LNA) to supercoiled plasmids for PNA-anchored "Bioplex" formation.

Low cellular uptake and poor nuclear transfer hamper the use of non-viral vectors in gene therapy. Addition of functional entities to plasmids using the Bioplex technology has the potential to improve the efficiency of transfer considerably. We have investigated the possibility of stabilizing sequence-specific binding of peptide nucleic acid (PNA) anchored functional peptides to plasmid DNA by hybridizing PNA and locked nucleic acid (LNA) oligomers as "openers" to partially overlapping sites on the opposite DNA strand. The PNA "opener" stabilized the binding of "linear" PNA anchors to mixed-base supercoiled DNA in saline. For higher stability under physiological conditions, bisPNA anchors were used. To reduce nonspecific interactions when hybridizing highly cationic constructs and to accommodate the need for increased amounts of bisPNA when the molecules are uncharged, or negatively charged, we used both PNA and LNA oligomers as "openers" to increase binding kinetics. To our knowledge, this is the first time that LNA has been used together with PNA to facilitate strand invasion. This procedure allows hybridization at reduced PNA-to-plasmid ratios, allowing greater than 80% hybridization even at ratios as low as 2:1. Using significantly lower amounts of PNA-peptides combined with shorter incubation times reduces unspecific binding and facilitates purification.

Protease-induced release of functional peptides from bioplexes.

Linking peptide functions directly to nucleic acids can be used to improve transfection. We have previously demonstrated this by sequence-specific hybridization of a bifunctional peptide nucleic acid (PNA) consisting of a nucleic acid binding moiety conjugated to a peptide. The resulting biological complex of PNA/DNA is called a Bioplex. The bifunctional PNA is continuously synthesized with one or more functional entities. For certain applications, it might be preferable to eliminate a functional entity after it has served its purpose. We have addressed this issue by adding a specific protease cleavage site to the construct. In this first approach, cathepsin L was used to cleave a linker sequence including a cathepsin L site: afrsaaq, thereby releasing the tri-peptide Arg-Gly-Asp (RGD) from the PNA anchor. In vitro and in vivo experiments showed an efficient cleavage of the peptide. Moreover, bifunctional PNA constructs were shown to retain activity of the second entity following removal of the first function. Since cathepsin L is ubiquitously expressed in eukaryotic cells and becomes active as the endosomal pH drops, inclusion of cathepsin sites makes it possible to remove functional entities in late endosomes/early lysosomes.