Johannes Oehlke

Cellular uptake of an α-helical amphipathic model peptide with the potential to deliver polar compounds into the cell interior non-endocytically

Evidence that multiple, probably non-endocytic mechanisms are involved in the uptake into mammalian cells of the alpha-helical amphipathic model peptide FLUOS-KLALKLALKALKAALKLA-NH2 (I) is presented. Extensive cellular uptake of N-terminally GC-elongated derivatives of I, conjugated by disufide bridges to differently charged peptides, indicated that I-like model peptides might serve as vectors for intracellular delivery of polar bioactive compounds. The mode of the cellular internalization of I comprising energy-, temperature-, pH- and ion-dependent as well as -independent processes suggests analogy to that displayed by small unstructured peptides reported previously (Oehlke et al., Biochim. Biophys. Acta 1330 (1997) 50-60). The uptake behavior of I also showed analogy to that of several protein-derived helical peptide sequences, recently found to be capable of efficiently carrying tagged oligonucleotides and peptides directly into the cytosol of mammalian cells (Derossi et al., J. Biol. Chem. 269 (1994) 10444-10450; Lin et al., J. Biol. Chem. 270 (1995) 14255-14258; Fawell et al., Proc. Natl. Acad. Sci. USA 91 (1994) 664-668; Chaloin et al., Biochemistry 36 (1997) 11179-11187; Vives et al., J. Biol. Chem., 272 (1997) 16010-16017).

Structural requirements for cellular uptake of α-helical amphipathic peptides

The structure of the cell‐permeable α‐helical amphipathic model peptide FLUOS‐KLALKLALKALKAALKLA‐NH2 (I) was modified stepwise with respect to its helix parameters hydrophobicity, hydrophobic moment and hydrophilic face as well as molecular size and charge. Cellular uptake and membrane destabilizing activity of the resulting peptides were studied using aortic endothelial cells and HPLC combined with CLSM. With the exceptions that a reduction of molecule size below 16 amino acid residues and the introduction of a negative net charge abolished uptake, none of the investigated structural parameters proved to be essential for the passage of these peptides across the plasma membrane. Membrane toxicity also showed no correlation to any of the parameters investigated and could be detected only at concentrations higher than 2 μm. These results implicate helical amphipathicity as the only essential structural requirement for the entry of such peptides into the cell interior, in accord with earlier studies. The pivotal role of helical amphipathicity was confirmed by uptake results obtained with two further pairs of amphipathic/non‐amphipathic 18‐mer peptides with different primary structure, net charge and helix parameters from I. The amphipathic counterparts were internalized into the cells to a comparable extent as I, whereas no cellular uptake could be detected for the non‐amphipathic analogues. The mode of uptake remains unclear and involves both temperature‐sensitive and ‐insensitive processes, indicating non‐endocytic contributions. Copyright

Extensive cellular uptake into endothelial cells of an amphipathic β-sheet forming peptide

Extensive internalization into endothelial cells has been found for a water soluble amphipathic 26‐mer β‐sheet peptide (FLUOS‐DPKGDPKGVTVTVTVTVTGKGDPKPD‐NH2; VT5). With the d‐Val13,d‐Thr14 di‐d‐amino acid analog of VT5 (DD‐VT5), exhibiting an identical primary structure but no propensity to adopt a β‐sheet conformation, only about 5% of the cellular uptake of VT5 was found. The mechanism of entry of VT5 into the cells remained unclear, but proved to be energy, temperature and pH dependent and, therefore, clearly distinct from that reported for helical amphipathic peptides. No detectable cytotoxicity, high solubility in water and the found extensive entry into endothelial cells make VT5 appear a good lead for developing new types of vectors for delivering oligonucleotides and peptides into intact cells.

Evidence for extensive and non-specific translocation of oligopeptides across plasma membranes of mammalian cells

After exposure of bovine aortic endothelial cells to various small peptides (tetra- to undeca-mer), extensive transport of the peptides across the plasma membrane was observed in the concentration range 10(-7) to 10(-2) M. The observed transport events, which contradict the generally anticipated poor permeability of peptides across plasma membranes, exhibited high complexity and showed no saturability up to a concentration of 10(-2) M. Evidence was found for the involvement of mdrp-like transporters as well as of energy-independent facilitated diffusion events. The peptide levels within the cells approximated those of the incubation solution within 30 min, indicating high capacity and velocity for the involved transport processes. Correspondingly, preloaded cells exported about 80% of the internalized peptide within 5 min at 37 degrees C. Analogous results were found after peptide exposure to several other mammalian cell types, indicating a more general importance of the transport phenomena described here. Our findings contradict the prevailing opinion that the often observed lack of activity of externally administered peptides against their targets within intact cells is accounted for primarily by poor cellular uptake and point to export processes counteracting the uptake to be more important in this context.

Cellular uptake and biological activity of peptide nucleic acids conjugated with peptides with and without cell-penetrating ability.

A 12‐mer peptide nucleic acid (PNA) directed against the nociceptin/orphanin FQ receptor mRNA was disulfide bridged with various peptides without and with cell‐penetrating features. The cellular uptake and the antisense activity of these conjugates were assessed in parallel. Quantitation of the internalized PNA was performed by using an approach based on capillary electrophoresis with laser‐induced fluorescence detection (CE‐LIF). This approach enabled a selective assessment of the PNA moiety liberated from the conjugate in the reducing intracellular environment, thus avoiding bias of the results by surface adsorption. The biological activity of the conjugates was studied by an assay based on the downregulation of the nociceptin/orphanin FQ receptor in neonatal rat cardiomyocytes (CM). Comparable cellular uptake was found for all conjugates and for the naked PNA, irrespective of the cell‐penetrating properties of the peptide components. All conjugates exhibited a comparable biological activity in the 100 nM range. The naked PNA also exhibited extensive antisense activity, which, however, proved about five times lower than that of the conjugates. The found results suggest cellular uptake and the bioactivity of PNA‐peptide conjugates to be not primarily related to the cell‐penetrating ability of their peptide components. Likewise from these results it can be inferred that the superior bioactivity of the PNA‐peptide conjugates in comparison with that of naked PNA rely on as yet unknown factors rather than on higher membrane permeability. Several hints point to the resistance against cellular export and the aggregation propensity combined with the endocytosis rate to be candidates for such factors. Copyright

Gonadotropin-releasing hormone (GnRH) analogs: relationship between their structure, proteolytic inactivation and pharmacokinetics in rats

There are two types of superactive agonists of gonadotropin-releasing hormone (GnRHa-I: (D-amino acid)6-GnRH and GnRHa-II: (D-amino acid)6-(desGly)10-GnRH- ethylamide) the high hormonal activity of which is understood to be due to their higher receptor affinity and their higher proteolytic stability as compared with the native GnRH sequence. Using the soluble fractions of various rat tissues in studies on the inactivation of GnRH peptides, we confirmed the higher proteolytic resistance of GnRHa-II, but not of D-Phe6-GnRH (GnRHa-I) and of another analog, D-Trp3-D-Phe6-GnRH, as compared with GnRH. The exact behaviour of the peptides during degradation was found to be dependent on the peptide concentrations used, showing the importance of using conditions as near to the physiological ones a possible. Towards the membrane fractions, however, the order of degradability was found to be GnRH much greater than D-Phe6-GnRH much greater than D-Trp3-D-Phe6-GnRH. The pharmacokinetic consequences of the different proteolytic degradabilities of the GnRH peptides, observed in rats, were a moderate increase in the biological half-life of D-Phe6-GnRH by 2.5-fold, as compared with GnRH, and a small increase in half-life of D-Trp3-D-Phe6-GnRH by 1.4-fold when compared with D-Phe6-GnRH. Whereas no intact GnRH was recovered in rat urine, small amounts of D-Phe6-GnRH (about 1% of dose) and high amounts of D-Trp3-D-Phe6-GnRH (25.5%) were excreted into urine. Combining the biochemical and pharmacokinetic data, it is concluded that proteolytic stability of GnRH analogs in pharmacological terms means stability towards membrane enzymes (pharmacologically-related stability) and that designing analogs with further increased proteolytic stability will be of only limited consequences with respect to their biological half-lives, the glomerular filtration rate of the kidney becoming the determining factor in the peptide clearance.

Induction of caspase-8 in human cells by the extracellular administration of peptides containing a C-terminal SLV sequence

Extracellular administration of a membrane permeable model peptide containing the tripeptide sequence, SLV, at the C-terminus to human endothelial and kidney cells resulted in an induction of caspase-8 (FLICE), the apical enzyme of the apoptosis caseade. The unmodified or N-terminally SLV-tagged peptide had no effect, thereby eliminating an unspecific induction of apoptosis as the cause of the caspase activity observed. Drastic alterations of primary structure and structure forming properties of the carrier peptide did not significantly influence the caspase-8 inducing activity of the C-terminal SLV-tag, supporting previous findings that translocation into the cell interior is a more general ability of peptides.

Growth Factor- and Adhesion Protein-Like Components of Fetal Calf Serum Can Significantly Enhance the Intracellular Delivery of Peptide Nucleic Acids

Evidence is presented that components of fetal calf serum (FCS) can significantly enhance the splicing correction activity of peptide nucleic acids (PNA) in HeLa pLuc 705 cells. The effect proved more pronounced for PNAs bearing fluorescence tags and relies on the ability of specific components of FCS to mediate a mainly nonendocytotic intracellular delivery of PNA. Attempts to isolate and characterize the active serum components using PNA-loaded beads and nano-LC-ESI mass spectrometry revealed the growth-factor related inter-alpha-trypsin inhibitor and the adhesion protein fibronectin to be substantially responsible for the delivery activity of FCS.

Evidence for Extensive Non-Endocytotic Translocation of Peptide Nucleic Acids Across Mammalian Plasma Membranes

The ability of peptide nucleic acids (PNA) to enter and to cross filter-grown MDCK, HEK and CHO cells was studied by means of a protocol based on capillary electrophoresis combined with laser-induced fluorescence detection. The used approach avoided possible errors encountered in protocols based on confocal laserscanning microscopy and FACS analysis. In contradiction to the commonly anticipated unability of PNA to cross biomembranes, extensive translocation of unmodified PNA into and across the investigated cell types was found. The transport mode comprised a variety of energy dependent and -independent as well as temperature sensitive mechanisms being probably destined to natural substrates and hijacked by PNA. The presented results suggest active as well as passive export mechanisms rather than poor penetration into cells to be responsible for the only weak biological activity of unmodified PNA.