Trusting your mail to a reliable carrier is something we all tend to take for granted, and usually items end up at the right destination. The same cannot be said for delivery of peptide and protein therapeutics across relatively impermeable cell membranes. A technical report in the December issue of Nature Biotechnology describes a new peptide, Pep-1, that is designed to deliver peptide and protein drugs reliably into cells.

Several naturally occurring small protein domains have been shown to cross biological membranes efficiently, without interacting with transporters or receptors. Attaching these domains to peptides or proteins promotes the delivery of these large molecules into cells. An HIV-1 protein, tat, is able to deliver biologically active proteins. Similarly, a protein from the herpes simplex virus, VP22, and one α-helix of the Antennapedia homeodomain also promote the delivery of peptides or proteins into cells. However, major drawbacks with these carriers are that they all require cross-linking to the target peptide or protein. The tat-fusion-protein system sometimes even requires target denaturation, a requirement that would preclude this system from delivering native protein therapeutics.

Morris et al. present a new protein delivery strategy based on a short amphipathic peptide that contains both hydrophobic and hydrophilic regions. The 21-residue peptide, Pep-1, can efficiently deliver various biologically active peptides and proteins into a number of cell lines, without requiring covalent coupling or denaturation of the protein cargo. When Pep-1 is mixed with peptides or proteins in solution, rapid association occurs through non-covalent hydrophobic interactions, and stable complexes are formed. Although Pep-1 alone localizes to the nucleus, this does not affect the subcellular localization of the protein being delivered. Pep-1 was shown to be able to deliver peptides to the nucleus or the cytoplasm, depending on the signal motif expressed by the transported peptide. Importantly, Pep-1 showed very low toxicity, and also did not seem to have any effect on the biological activity of the protein, as shown using the enzyme β-galactosidase.

It remains to be seen whether Pep-1 will be as successful in vivo as it has proved to be in vitro, and what limits there will be on the size of the molecules that can be delivered. As genomic and proteomic information filter down, biopharmaceuticals will become more common, and Pep-1 could be important in the delivery of these proteins. Furthermore, Pep-1 and other protein delivery technologies constitute useful tools for basic research.