Review
Keynote
Cell-penetrating peptides: classes, origin, and current landscape

https://doi.org/10.1016/j.drudis.2012.03.002Get rights and content

With more than ten new FDA approvals since 2001, peptides are emerging as an important therapeutic alternative to small molecules. However, unlike small molecules, peptides on the market today are limited to extracellular targets. By contrast, cell-penetrating peptides (CPPs) can target intracellular proteins and also carry other cargoes (e.g. other peptides, small molecules or proteins) into the cell, thus offering great potential as future therapeutics. In this review I present a classification scheme for CPPs based on their physical–chemical properties and origin, and I provide a general framework for understanding and discovering new CPPs.

Highlights

► More than 100 CPPs are presented, divided into cationic, amphipathic, and hydrophobic. ► Origin-based classification: natural proteins/peptides; designed; random libraries ► Natural sources of CPPs: heparin-, DNA/RNA binding proteins; viral proteins; antimicrobial/signal peptides.

Section snippets

Physical–chemical properties-based classification

Even though CPPs have a great sequence variety, it is possible to identify three major classes: cationic, amphipathic and hydrophobic. Figure 1 presents a broad overview of the current CPPs landscape. The data presented contain more than 100 diverse CPPs (corresponding to the CPPs shown in Table 1, Table 2, Table 3, Table 4, Table 5, Table 6) collected from publications and patents, excluding mutants around the same peptides. Most of the CPPs in this set have a net positive charge (83%);

Cationic CPPs

The first CPP discovered was cationic and was derived from the HIV-1 protein Tat (RKKRRQRRR) [34]. Studies on arginine-based peptides (from R3 to R12) have shown that the minimal sequence for cellular uptake is octaarginine (R8), and that increasing the number of arginines increases the level of uptake [20]. Polylysine, in comparison, has a much poorer uptake profile [20].

Studies suggest that at least eight positive charges are needed for efficient uptake of several other cationic CPPs [35].

Primary amphipathic

Several primary amphipathic CPPs are chimeric peptides obtained by covalently attaching a hydrophobic domain for efficient targeting to cell membranes to a NLS. For example, MPG (GLAFLGFLGAAGSTMGAWSQPKKKRKV) and Pep-1 (KETWWETWWTEWSQPKKRKV) are both based on the SV40 NLS PKKRKV. The hydrophobic domain of MPG was derived from the fusion sequence of the HIV glycoprotein 41 (GALFLGFLGAAGSTMGA), while that of Pep-1 corresponds to a tryptophan-rich cluster (KETWWETWWTEW), which has high affinity for

Hydrophobic CPPs

In this article I consider hydrophobic peptides that either: contain only apolar residues; have a low net charge (less than 20% of the sequence); or have a hydrophobic motif or chemical group that is crucial for uptake regardless of the rest of the sequence. Compared with cationic and amphipathic peptides, only a few hydrophobic CPPs have been discovered, however this might be a mere reflection of an historical bias towards cationic and amphipathic CPPs, which were discovered first. Indeed, it

Prenylated peptides

The addition of either a farnesyl (C15) or geranylgeranyl (C20) isoprenoid moiety, known as prenylation, has been reported to give peptides inherent cell-penetrating ability through an ATP-independent, non-endocytic pathway [72]. Although more work is needed to fully understand the role of prenylation on cellular uptake, current studies suggest that uptake is independent of the specific sequence, as is the case for stapled peptides.

Pepducins

Pepducins are a class of N-terminally lipidated peptides that can cross the cell membrane and bind to the cytosolic region of a variety of transmembrane proteins (GPCRs, MMPs, among others) [73]. The N-terminal lipidation consists of either palmitoyl or other fatty acids. Two pepducins, x1/2pal-i3 and x1/2LCA-i1, have entered preclinical development. Contrary to CPPs, pepducins remain anchored to the cell membrane, and are not released in the cytosol, thus focusing their application to

The role of peptide 3D structure on cellular uptake

The role of peptide secondary structure in cell-penetration remains elusive. First, secondary structure depends on the medium [74]. Peptides can adopt completely different conformations depending on whether they are in water, near the membrane interface, inside the membrane, or bound to a protein. Second, the importance of the secondary structure depends on the mode of uptake and on the peptide class (cationic, amphipathic or hydrophobic). α-helical and β-strand CPPs can be sensitive to

Origin-based classification

The discovery of the first CPP, the cationic peptide Tat, was followed by the identification of a cationic and partially amphipathic CPP, Penetratin, from the homeodomain of Antennapedia. Mutation studies demonstrated that positive charges and amphipathicity are important features for cell penetration, and these characteristics were carefully considered to design new synthetic CPPs. Concurrently, researchers also looked for these features within domains of natural proteins. Presently, new CPPs

CPPs derived from natural proteins or peptides

Natural protein motifs are a rich source of CPPs, but establishing a link between a CPP and its role in the protein from which it is derived presents considerable challenges. Proteins might be able to enter the cell through a variety of motifs which, if isolated, would not be cell-penetrating. For example, certain plasma membrane proteins are ubiquinated at the level of their cytosolic tail to control their subsequent endocytosis. Many transmembrane proteins contain specific and diverse

Designed CPPs and CPPs derived from large peptide libraries

While the majority of designed peptides have a characteristic amphipathic structure (Table 5), peptides discovered from high-throughput screening on large peptide libraries are much more diverse, with many having a larger number of hydrophobic, rather than cationic amino acids, and no clear preference for amphipathic structures.

CPPs from large combinatorial peptide libraries

Randomized peptide libraries can be obtained through DNA-encoded peptide libraries and thus enable the generation of billions of peptides. Several methods are available to link peptides to their encoding DNA, such as phage display, plasmid display, micro-organism surface display and ribosome display [88]. Alternatively, peptide arrays can be used for high-throughput synthesis of tens of thousands peptides. Several CPPs have been discovered from such libraries, such as SG3 [66] (plasmid

Phylomers

Phylomers are libraries of natural peptides encoded by natural genes of diverse bacterial genome [93]. Compared with random peptide sequences obtained by phage display, phylomer libraries contain millions of peptides with subdomains that already evolved to maintain some structural stability. New CPPs obtained from this approach have also been reported, some cationic, others amphipathic with a net negative charge, and others mainly hydrophobic. Examples include peptides BEN_1079 and BEN_0805

CPPs prediction from natural proteins and random libraries

Only a few prediction models for CPPs have been proposed, including that by Hansen et al. [94] based on Sandberg z descriptors for amino acids [95], and that by Sanders et al. based on biochemical properties of peptides (such as amino acid composition, peptide length and net charge) [96]. Peptides discovered through these methods, which are reported in Table 5 have a varied amino acid composition.

Concluding remarks

With a great sequence variety and large differences in terms of physical chemical properties, CPPs can be linear, cyclical, cationic, anionic, hydrophobic, hydrophilic, amphipathic, non-amphipathic, random coiled, α-helical, or β-sheets. However, CPPs differ from most other peptides with respect to specific features that reflect various mechanisms used to enter the cell.

Highly cationic CPPs (i.e. CPPs with at least eight positive charges) can interact with GAGs and enter the cell through

Francesca Milletti leads the Cheminformatics and Statistics group at Roche (Nutley, NJ, USA). She joined Roche in 2010 after a postdoctoral fellowship at Novartis, Basel. Before that, she was a visiting student at the University of California, San Francisco (2008). She received her PhD and undergraduate degree in chemistry from the University of Perugia, Italy, where she developed computational tools for pKa prediction and tautomer enumeration. Her research focuses on novel computational

References (128)

  • A. Prochiantz

    Getting hydrophilic compounds into cells: lessons from homeopeptides

    Curr. Opin. Neurobiol.

    (1996)
  • A.D. Ragin

    Cellular import mediated by nuclear localization signal peptide sequences

    Chem. Biol.

    (2002)
  • K. Oglecka

    Relevance of the N-terminal NLS-like sequence of the prion protein for membrane perturbation effects

    Biochim. Biophys. Acta: Biomembr.

    (2008)
  • J. Oehlke

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

    Biochim. Biophys. Acta: Biomembr.

    (1998)
  • W. Li

    GALA: a designed synthetic pH-responsive amphipathic peptide with applications in drug and gene delivery

    Adv. Drug Deliv. Rev.

    (2004)
  • M. Charton et al.

    The structural dependence of amino acid hydrophobicity parameters

    J. Theor. Biol.

    (1982)
  • K.M. Biswas

    Evaluation of methods for measuring amino acid hydrophobicities and interactions

    J. Chromatogr., A

    (2003)
  • J. Kyte et al.

    A simple method for displaying the hydropathic character of a protein

    J. Mol. Biol.

    (1982)
  • U. Soomets

    Deletion analogues of transportan

    Biochim. Biophys. Acta: Biomembr.

    (2000)
  • J. Oehlke

    Extensive cellular uptake into endothelial cells of an amphipathic [beta]-sheet forming peptide

    FEBS Lett.

    (1997)
  • T. Yoshida

    Bax-inhibiting peptide derived from mouse and rat Ku70

    Biochem. Biophys. Res. Commun.

    (2004)
  • M. Rhee et al.

    Mechanism of uptake of C105Y, a novel cell-penetrating peptide

    J. Biol. Chem.

    (2006)
  • C. Gao

    A cell-penetrating peptide from a novel pVII–pIX phage-displayed random peptide library

    Bioorg. Med. Chem.

    (2002)
  • F. Nakayama

    Fibroblast growth factor-12 (FGF12) translocation into intestinal epithelial cells is dependent on a novel cell-penetrating peptide domain

    J. Biol. Chem.

    (2011)
  • H. Zhang

    A cell-penetrating helical peptide as a potential HIV-1 inhibitor

    J. Mol. Biol.

    (2008)
  • J.D. Ochocki

    Evaluation of a cell penetrating prenylated peptide lacking an intrinsic fluorophore via in situ click reaction

    Bioorg. Med. Chem. Lett.

    (2011)
  • D. Derossi

    Cell internalization of the third helix of the antennapedia homeodomain is receptor-independent

    J. Biol. Chem.

    (1996)
  • S. Lu

    A cell-penetrating peptide derived from mammalian cell uptake protein of Mycobacterium tuberculosis

    Anal. Biochem.

    (2006)
  • J.P.M. Langedijk

    translocation activity of c-terminal domain of pestivirus erns and ribotoxin L3 loop

    J. Biol. Chem.

    (2002)
  • J.P.M. Langedijk

    Application, efficiency and cargo-dependence of transport peptides

    Int. Congr. Ser.

    (2005)
  • S.S. Sidhu et al.

    DNA-encoded peptide libraries and drug discovery

  • S. Nishimura

    Combinatorial targeting of the macropinocytotic pathway in leukemia and lymphoma cells

    J. Biol. Chem.

    (2008)
  • C. Wu

    A peptide-based carrier for intracellular delivery of proteins into malignant glial cells in vitro

    J. Control. Release

    (2008)
  • M. Hansen

    Predicting cell-penetrating peptides

    Adv. Drug Deliv. Rev.

    (2008)
  • I. Nakase

    Cell-surface accumulation of flock house virus-derived peptide leads to efficient internalization via macropinocytosis

    Mol. Ther.

    (2009)
  • J.M. Reichert

    Development Trends for Peptide Therapeutics

    (2008)
  • A.L. Hopkins et al.

    The druggable genome

    Nat. Rev. Drug Discov.

    (2002)
  • A. Mullard

    Protein-protein interactions get into the groove

    Nat. Rev. Drug Discovery

    (2012)
  • F. Heitz

    Twenty years of cell-penetrating peptides: from molecular mechanisms to therapeutics

    Br. J. Pharmacol.

    (2009)
  • B.N. Taylor

    Noncationic peptides obtained from azurin preferentially enter cancer cells

    Cancer Res.

    (2009)
  • H.J. Johansson

    Characterization of a novel cytotoxic cell-penetrating peptide derived from p14ARF protein

    Mol. Ther.

    (2007)
  • F. Madani

    Mechanisms of cellular uptake of cell-penetrating peptides

    J. Biophys.

    (2011)
  • S. Mayor et al.

    Pathways of clathrin-independent endocytosis

    Nat. Rev. Mol. Cell. Biol.

    (2007)
  • P. Lundin

    Distinct uptake routes of cell-penetrating peptide conjugates

    Bioconjugate Chem.

    (2008)
  • R.M. Steinman

    Endocytosis and the recycling of plasma membrane

    J. Cell. Biol.

    (1983)
  • K. Jacobson

    Lipid rafts: at a crossroad between cell biology and physics

    Nat. Cell. Biol.

    (2007)
  • A. Shevchenko et al.

    Lipidomics: coming to grips with lipid diversity

    Nat. Rev. Mol. Cell. Biol.

    (2010)
  • K. Simons et al.

    Lipid rafts and signal transduction

    Nat. Rev. Mol. Cell. Biol.

    (2000)
  • H.T. McMahon et al.

    Membrane curvature and mechanisms of dynamic cell membrane remodelling

    Nature

    (2005)
  • G. van Meer

    Membrane lipids: where they are and how they behave

    Nat. Rev. Mol. Cell. Biol.

    (2008)
  • Cited by (661)

    • Helix-specific properties and applications in synthetic polypeptides

      2023, Current Opinion in Solid State and Materials Science
    • mRNA: A promising platform for cancer immunotherapy

      2023, Advanced Drug Delivery Reviews
    View all citing articles on Scopus

    Francesca Milletti leads the Cheminformatics and Statistics group at Roche (Nutley, NJ, USA). She joined Roche in 2010 after a postdoctoral fellowship at Novartis, Basel. Before that, she was a visiting student at the University of California, San Francisco (2008). She received her PhD and undergraduate degree in chemistry from the University of Perugia, Italy, where she developed computational tools for pKa prediction and tautomer enumeration. Her research focuses on novel computational methods for drug discovery with a special interest on cell-penetrating peptides.

    View full text