Evaluation of Cell-Penetrating Peptides (CPPs) as Vehicles for Intracellular Delivery of Antisense Peptide Nucleic Acid (PNA) Nadia Bendifallah, ² Frank Winther Rasmussen, § Vladimir Zachar, # Peter Ebbesen, # Peter E. Nielsen, ²,§ and Uffe Koppelhus* ,²,# Center for Biomolecular Recognition, Department of Medical Biochemistry and Genetics, Biochemistry Laboratory B, The Panum Institute, University of Copenhagen, Blegdamsvej 3c, DK-2200 Copenhagen N, Denmark; Pantheco A/S, Bøge Alle ´ 3, DK-2970 Hørsholm, Denmark; and Laboratory for Stem Cell Research, Institute for Health Technologies, Aalborg University, Fredrik Bajers Vej 3B, DK-9220 Aalborg Ø, Denmark. Received September 21, 2005; Revised Manuscript Received March 22, 2006 Cell-penetrating peptides (CPPs) are characterized by their ability to be internalized in mammalian cells. To investigate the relative potency of CPPs as carriers of medicinally relevant cargo, a positive read-out assay based on the ability of a peptide nucleic acid (PNA) oligomer to promote correct expression of a recombinant luciferase gene was employed. Seven different CPPs were included in the study: Transportan, oligo-arginine (R 7-9 ), pTat, Penetratin, KFF, SynB3, and NLS. The CPP-PNA conjugates were synthesized by different conjugation chemistries: continuous synthesis, maleimide coupling, and ester or disulfide linkage. Under serum-free conditions PNA-SS-Transportan-amide (ortho)-PNA was found to be the most potent conjugate, resulting in maximum luciferase signal at a concentration of 1-2 µM. (D-Arg) 9 -PNA showed optimal efficacy at 5 µM but gave rise to only one-third of the luciferase signal obtained with the Transportan conjugate. The pTat- and KFF-PNA conjugates showed significantly lower efficacy. The penetratin-, SynB3-. and NLS-PNA conjugates showed only minimal or no activity. Serum was found to have a drastic negative impact on CPP-driven cellular uptake. PNA-SS-Transportan-acid (ortho) and (D-Arg) 9 -PNA were least sensitive to the presence of serum. Both the chemical nature and, in the case of Transportan, the position of the peptide PNA coupling were found to have a major impact on the transport capacity of the peptides. However, no simple relationship between linker type and antisense activity of the conjugates could be deduced from the data. INTRODUCTION Over the past decade a group of cell-penetrating peptides (CPPs) have been identified. CPPs are characterized by their ability to become internalized by mammalian cells and, in some cases, to cotransport macromolecules to the cellular interior. Proteins, peptides, oligonucleotides, and other macromolecules with a therapeutic potential related to intracellular processes are prospective cargos for CPPs. The most intensively studied CPPs include protein-derived peptides such as pTat, pAntp (Penetra- tin), and VP22 and synthetic or chimeric peptides such as oligo- arginine and Transportan, respectively (1-7). The mechanism underlying the translocation of CPPs across the lipid bilayer of mammalian cells is yet not fully understood. Initially, a “passive” passage over the membrane based on a temperature- and energy-independent “physicochemical” mech- anism was suggested (8-10). However, later studies strongly suggested that (at least some of) the peptides designated CPPs are internalized in cells by classic endocytotic pathways (11- 13), and most recently evidence for multiple pathways has emerged (14, 15). Thus, further investigations on the mechanism of CPPs are warranted. Another largely unresolved aspect of CPPs is their (relative) potency as transport vehicles for therapeutically relevant macromolecules. Due to the large diversity in application, cell type, and methodology, it is not possible to make a relevant comparison of the CPPs based on meta-analysis of the many data reported. Moreover, only a few studies (comprising only a few CPPs) have been concerned with a direct comparison of CPP performance (2, 5, 11, 16-19). Therefore, to establish data on the relative efficacy of some of the most prominent CPPs, we employed a transgenic reporter cell line, pLuc 705 HeLa, in which a PNA targeted to an aberrant splice site in luciferase mRNA is known to induce correct splicing and thus produce a positive read-out signal in the form of increased luciferase activity (20, 21). We use this assay to evaluate the potency of seven different CPPs as vehicles for intranuclear delivery of PNA. We chose to analyze pTat, pAntp (Penetratin), and Transportan, as these peptides are some of the most widely used CPPs that have also been employed in earlier studies with PNA (11, 22, 23). We also included oligo-arginine, KFF, SynB3, and NLS. Oligo-arginine peptides have been shown to be potent cellular delivery vehicles for morpholino oligomers (2). Like PNA oligomers, morpholino oligomers have uncharged backbones. KFF has been shown to induce cellular uptake of PNA in prokaryotic cells (24, 25), but data from eukaryotic cells are lacking. SynB3 has been claimed to enhance passage of the blood-brain barrier of conjugated molecules such as doxorubicin (26) and morphine-6-glucuronide (27) but has not previously been tested as a cellular vehicle for PNA. NLS was included in the study as earlier papers have claimed this peptide to promote PNA translocation over the plasma cell membrane (28). NLS is not usually perceived as having capacity to transgress membranes other than the nucleolemma. In addition to the PNA-peptide conjugates, we tested a PNA-biotin * Corresponding author [e-mail uffek@hst.aau.dk; fax (+45) 9635 7816]. ² University of Copenhagen. § Pantheco A/S. Present address: Novozymes A/S, Krogshoejvej 36, DK-2880 Bagsvaerd, Denmark. # Aalborg University. 750 Bioconjugate Chem. 2006, 17, 750-758 10.1021/bc050283q CCC: $33.50 © 2006 American Chemical Society Published on Web 04/15/2006