Current Drug Metabolism, 2012, 13, 000-000
1
Intracellular Delivery: Exploiting Viral Membranotropic Peptides Galdiero Stefania2,3,4, Vitiello Mariateresa1, Falanga Annarita2, Cantisani Marco2, Incoronato Novella1 and Galdiero Massimiliano1,3,* 1 Department of Experimental Medicine - II University of Naples; Via De Crecchio 7; 80138; Napoli; Italy, 2Department of Biological Sciences; Division of Biostructures and 3Centro Interuniversitario di Ricerca sui Peptidi Bioattivi - University of Naples “Federico II”; Via Mezzocannone 16; 80134; Napoli; Italy, 4Istituto di Biostrutture e Bioimmagini – CNR; Via Mezzocannone 16; 80134; Napoli; Italy
Abstract: Recent advances in the understanding of cellular and molecular mechanisms of the pathogenesis of several diseases offer the possibility to address novel molecular targets for an improved diagnosis and therapy. In fact; in order to fulfill their function; macromolecular drugs; reporter molecules; and imaging agents often require to be delivered into specific intracellular compartments; usually the cytoplasm or the nucleus. From a medical perspective; biological membranes represent a critical hindrance due to their barrier-like behaviour not easily circumvented by many pharmacologically-active molecules. Therefore; identifying strategies for membrane translocation is essential. Several technologies have been designed to improve cellular uptake of therapeutic molecules; including cell-penetrating peptides (CPPs). These peptides; which are able to efficiently translocate macromolecules through the plasma membrane; have attracted a lot of attention; and new translocating peptides are continuously described. In this review; we will focus on the viral derived peptides; and in particular those derived by viral entry proteins that may be useful as delivery vehicles due to their intrinsic properties of inducing membrane perturbation.
Keywords: Cell-penetrating peptides, endocytosis; intracellular delivery, membrane fusion, viral fusion proteins. INTRODUCTION The plasma membrane surrounding mammalian cells is extremely effective as a selectively permeable barrier. The presence of this finely tuned barrier allows cells to be able to survive and function properly. However; the lipophilic nature of biological membranes restricts the direct intracellular delivery of most compounds. In fact; many biologically active molecules (therapeutic proteins or DNA) need to be delivered intracellularly to exert their therapeutic action inside the cytoplasm; the nucleus; or other specific organelles. Chemically modified compounds have been showed to improve permeability; but these modifications that allow a better translocation through the membrane barrier often result in a partially sacrificed activity. At the present; several viral delivery strategies have been developed and have given much hope especially for gene therapies; but their clinical application has often been drawn back by several problems such as side-effects and toxicity [1]. Therefore; many researches have switched to the development of safer non-viral strategies [2-4] and have proposed different methods: lipid; polycationic nanoparticles and peptide-based formulations; but only few of these technologies have been efficient in vivo and have reached the clinic. So; although many pharmaceutical compounds show a promising potential in vitro; they cannot be used in vivo due to bioavailability problems. Alternatively; synthetic transporters can be utilized to promote cellular uptake. Linking or complexing therapeutic molecules to peptides that can translocate through the cellular membranes could enhance their internal delivery; andconsequently; a higher amount of active compound would reach the site of action. Cell penetrating peptides (CPP) are among the most promising non-viral strategies. Although defining CPPs is rather difficult in view of the fact that new CPPs are continuously discovered; *Address correspondence to this author at the Department of Experimental Medicine-II University of Naples; Via De Crecchio 7; 80138; Naples; Italy; Tel: +39 0815667646; Fax: +39 0815667578; E-mail:
[email protected] 1389-2002/12 $58.00+.00
rendering any classification constantly evolving and therefore never updated; they can generally be considered as short peptides of less than 30 amino acids either derived from proteins or from chimeric sequences. They are usually amphipathic and possesses a net positive charge [5-7]. CPPs can be subdivided into two main classes; the first requiring chemical linkage with the cargo and the second involving the formation of stable; noncovalent complexes [8-11]. CPPs from both classes have been described to efficiently deliver a large number of cargos (plasmid DNA; oligonucleotide; siRNA; PNA; protein; peptide; liposome; nanoparticle; etc.) into a wide variety of cell types and in vivo models [7]. The following introduction will briefly summarize how the CPPs field originated and how these peptides managed to attract so much interest as delivery agents. The idea of molecular transporters came about 20 years ago based on the observation that some proteins; mainly transcription factors; could move within cells and from one cell to another. The first protein able to translocate across cell membranes and to reach intracellular access was described in 1988. This was the human immunodeficiency virus (HIV) transactivator of transcription protein; called Tat [12].This key discovery was soon after followed by the identification of Antennapedia (Antp); a transcription factor of Drosophila melanogaster that was shown to be internalized by neuronal cells [13]. These works laid the foundation for the discovery that also selected parts of proteins could be translocated inside cells. The first of these domains; which was named Protein Transduction Domain (PTD); was a 16-mer peptide derived from the third helix of the homeodomain of Antennapedia named Penetratin. The first use of a PTD as a vector was described when penetratin was employed for the delivery of a small exogenous peptide in 1994 [14]. In 1997; Vives et al. reported the identification of the minimal sequence of Tat required for cellular uptake [15] and the first application of PTDs in vivo was reported by Schwarze et al. for the delivery of a biologically active beta-galactosidase protein into the mouse [16]. The notion of cell penetrating peptide (CPP) was introduced in 1998; with the design of the first synthetic peptide carrier; the Transportan; which is a 27 amino acid-long peptide containing 12 functional amino acids from the amino terminus of the neuropeptide © 2012 Bentham Science Publishers
2 Current Drug Metabolism, 2012, Vol. 13, No. ??
galanin and mastoparan (a wasp venom peptide) in the carboxyl terminus [17]. But; we should consider that nowadays the definitions PTD and CPP are overlapping; therefore from now on we will simply refer to CPPs for the rest of the review. Since the early discoveries; the list of available CPPs has impressively grown and their number continues to increase [7]. Peptides have been derived from VP22 protein of Herpes Simplex Virus [18]; from the PreS2-domain of hepatitis B virus surface antigen [19]; from calcitonin [20]; from antimicrobial or toxin peptides [21]; as well as from arginine-rich peptides [22]. Nowadays; CPPs constitute one of the most hopeful tool for the non-invasive intracellular transport of cargos and have been successfully applied for the delivery of therapeutic molecules varying from small chemical molecules; nucleic acids; proteins; peptides; liposomes to particles. This review will highlight the use of CPPs for biological applications; with particular attention paid to peptides derived from viral proteins and will also briefly discuss factors affecting mechanism of peptide uptake. Since CPPs have recently been comprehensively reviewed in the literature [8,23-29]; the purpose of this review will be to emphasize mainly viral-derived peptides; how we can exploit current knowledge of viral entry; and the impact of viral membranotropic sequences in the field of peptide mediated delivery. DESCRIPTION OF KNOWN CPPS Novel classes and applications of CPPs are being constantly discovered and have been reported through the study of proteins of the most various organisms. They differ significantly in their sequence; hydrophobicity; and polarity. Mainly; CPPs can be divided into three classes: a) protein derived peptides represented by short stretches of a protein domain that are the primary responsible for their translocation ability (examples: TAT; penetratin or pANTP; transportan; HSV-1 VP22); b) amphipathic peptides resulting from the sequential assembly of hydrophobic and hydrophilic domains (examples: MAP; MPG; Pep-1); c) synthetic or cationic peptides such as poliarginines (Arg9). Penetratin The Antennapedia motif is derived from a family of Drosophila homeoproteins; a class of trans-activating factors involved in the developmental process [31].These proteins recognise and bind DNA through a 60 amino acid carboxy-terminal region arranged in three -helical sequences [14]; called the homeodomain. The third helix of the homeodomain; comprising just 16 amino acids; also possesses this unusual property; enabling small molecules to be taken up into live cells [32]. Today; this peptide is commonly referred to as penetratin. The two residues of Trp present in penetratin are considered a necessary requirement for traslocation; their replacement with hydrophobic amino acids strongly decreases the internalization [14]. The high arginine and lysine content of penetratin gives it a positive charge at physiological pH. Penetratin shows high propensity to adopt an -helical conformation in membrane mimetic conditions; but in presence of phospholipids vesicles it adopts a -sheet structure with a residual -helical content at low concentrations [33;34;35]. Transportan Trasportan is a synthetic analogue of the peptide galanin synthesized by Poogie et al. and described in their study in 1998 [17]. It is a 27 amino acids long peptide and contains 12 functional amino acids taken from the neuropeptide galanin followed by a lysine connection and the remaining constituted by mastoparan; a wasp venom peptide toxin. Transportan can; in contrast to the galanin and mastoparan peptides alone; translocate through biological membranes and carry large hydrophilic molecules into the cell
Stefania et al.
without destroying the membrane [36]. After entering the cell; the peptide can be detected in the nuclear membrane and subsequently in the nucleus. Circular dichroism studies of transportan revealed a random coil secondary structure in water which could reach 60% helical conformation in SDS mimetic conditions [37]. Further studies by Nuclear Magnetic Resonance (NMR) in lipidic bicelles showed a well-defined -helix in the C-terminal mastoparan part of the peptide and a weaker tendency to form an -helix in the Nterminal domain [38]. Model Amphipathic Peptides (MAP) Model amphipathic peptides (MAP) are derived from the helical amphipathic model peptide KLALKLALKALKAALKLA; designed by Steiner et al. in 1991[39]. Lysine is the main positive charge contributor and is distributed in one face of the -helix; while the opposite face is rich in alanine residues. This peptide has been shown to internalise by multiple; nonspecific; energy-dependent and -independent processes into several types of cells [40]. Arg9 Arg9 is the prototype of the so-called polyarginine–based peptides; that simply derive from the observation that positive charged residues are crucial for translocation through biological membranes. Several studies using different cationic peptides showed the higher efficiency of polyarginines compared to the use of other residues such as histidines or lysine.; As a result; homoarginine sequences have been investigated as drug delivery vehicles [50]. DESCRIPTION OF OTHER MAJOR VIRAL-DERIVED CPPS The CPPs family puts together a vast number of peptides with the ability to translocate across membranes; regardless of whether they are natural; synthetic or chimaeric peptides. The peptides described above remain the most widely studied CPPs; and have been used to translocate a wide range of macromolecules into living cells. In virtue of the limitless diversity of peptide sequences and natural source molecules; it seems that peptide-guided intracellular translocation is a field in clear expansion; and the peptides derived from viral proteins seem very interesting as intracellular transporters also in consideration that the first protein that showed the ability to cross membranes was the HIV-1 Tat protein. Some examples of viral derived peptide investigated and included in the CPPs family are briefly described below. Tat 48-60 Peptide Tat is a transcription activating factor of 101 amino acids derived from HIV-1; necessary for the replication of the virus; which has the ability to cross the plasma membrane of neighbouring cells. In particular a short basic sequence from amino acid 48 to 60 is required for the binding of Tat to its activator RNA; which has been shown to adopt an extended structure [15;30]. Several peptides carrying mutations within the cationic domain of the original peptide were useful in clarifying that the translocating activity of the Tat protein is controlled by this cationic cluster of amino acids and that deletion of arginine led to non-translocating peptide. MPG MPG is an amphipathic peptide whose primary sequence is composed of a motif mainly constituted of hydrophobic amino acids derived from the fusion sequence of the HIV-1 protein gp41 (GALFLGFLGAAGSTMGA) associated to an hydrophilic domain with positively charged residues derived from the Nuclear Localization Sequence (NLS) of Simian virus 40 (SV40) large T antigen (PKKKRKV). These hydrophilic and hydrophobic segments are separated by a three amino-acid spacer (WSQ)[41;42]. It represents an interesting synthetic peptide that utilise the known properties of
Viral Peptides for Intracellular Delivery
the glycine-rich region of viral gp41 essential for membrane fusion activity and structural stabilization in addition to the NLS of the SV40 large T antigen able to improve the nuclear addressing of the peptide [43;44;45;46;47] Pep-1 Pep-1 is at least partially derived from MPG; in fact it conserves a identical C-terminal hydrophilic domain (the NLS of SV40 large T antigen) and spacer motif; while the hydrophobic region of Pep-1 corresponds to a W-rich segment (KETWWETWWTE) derived from the HIV-1 reverse transcriptase [48; 49]. Pep-1 has been shown to be able to efficiently deliver a variety of peptides and proteins into several cell lines in a fully biologically active form. Herpes Simplex Virus Type 1 (HSV-1) Protein VP22 VP22; the product of the UL49 gene of herpes simplex virus [51]; is a major component of the tegument; a structure residing between the capsid and envelope of herpes virus particles and has been reported to traffic between cells [52]. The C-terminal 40 amino acids region of the 301 residues VP22 protein is principally responsible for the transduction property of the whole protein and can mediate the delivery of DNA and RNA oligonucleotides into cells [53]. Sciortino et al. [54] have shown that VP22 can deliver a complete mRNA that is translated within the recipient cell. Hepatitis-B Virus (HBV) PreS2 Domain The property of cell permeability was observed in the PreS2domain (PLSSIFSRIGDP; subtype ayw) of the hepatitis-B virus surface antigens [19]. PreS2 has been shown to present an amphipatic -helix between amino acids 41 and 52. This domain is involved in PreS2 dimerization and the mutagenesis of amino acid residues within the amphipatic -helix resulted in a loss of the helical structure and led to -sheet conformation [55].These mutations also led to a reduced oligomerisation tendency and no significant cell permeability. Flock House Virus (FHV) Coat Protein -Peptide Flock House virus (FHV); an insect nodavirus; contains a 4kDa peptide called “gamma” (); which shares many of the characteristics of other nonenveloped virus lytic peptides [56]. The Nterminal region of the -peptide can alter membrane structure [57] and increase bilayer permeability [58]. Biophysical studies have shown that the -peptide possess membranotropic characteristics and strongly binds to fluid phase lipid bilayers increasing transmembrane permeability. An interesting thermodynamic description of the peptide binding to lipid vesicles showed that it displays a similar activity and thermodynamic profile to known membrane active peptides such as melittin and alamethicin [59]; suggesting the possibility of peptide-triggered disruption of the endosomal membrane as a prelude to viral uncoating in the host cytoplasm. A recent detailed analysis of the internalization methods of the FHV peptide revealed that this is mediated by relatively high cellsurface adsorption leading to enhanced macropinocytic uptake and cytosolic distribution. FHV peptide was found to internalise with a higher efficiency compared to Tat peptide [60]. gH625 gH625 is a membrane-perturbing domain derived from the glycoprotein gH of Herpes simplex virus type 1 [61;62]. The peptide gH625 is a mainly hydrophobic peptide which interacts with biological membranes; contributing to the merging of the viral envelope and the cellular membrane and has proved to be able to transport a cargo across the cell membrane. Avian Infectious Bursal Disease Virus Rath Peptide A novel CPP (named “Rath”) from VP5 protein domain of infectious bursal disease virus of poultry which is a member of genus
Current Drug Metabolism, 2012, Vol. 13, No. ??
3
Avibirnavirus of the family Birnaviridae has been recently described [63]. It has a dominant –structure and self-assembling properties. Rath peptide is also able to interact with protein and nucleic acids cargo molecules forming regular size particles. Its internalization mechanism has not yet been investigated. Among viral-derived peptides; the so-called viral fusion peptides (FP) are of paramount importance for their intrinsic properties of being able to interact with membranes. CPPs based on viral FPs are mainly represented by INF (derived from the N-terminal fusion peptide of influenza virus) and the previously described amphipathic peptide MPG which is a chimeric peptide containing the FP of the HIV-1 glycoprotein gp41. Several stretches of viral proteins have; indeed; been found to possess intrinsic ability to interact with membranes during the fundamental step of the viral life cycle represented by the viral penetration into the susceptible cell [64]. VIRAL ENTRY Viruses; like all obligate intracellular pathogens; must overcome numerous barriers in the host cell to deliver their nucleic acid into the cytosol or nucleus. This is the key to initiating their infectious cycle; and involves a number of distinct and subsequent steps like binding to one or more cellular receptors and entry; capsid destabilization and genome uncoating; ending in the release of viral nucleic acids at the proper site of replication. The most critical barrier is represented by the complex membranous system surrounding and residing within the host cell. Viruses can be classified as enveloped viruses when a membrane bilayer surrounds the capsid; whereas viruses with capsid proteins alone present on their surface are called nonenveloped viruses. The mechanisms of genome release are widely different between enveloped and non-enveloped viruses. In fact; enveloped viruses fuse with the plasma or endosome membranes; thereby exposing the genome or capsid to the cytosol; whereas nonenveloped viruses partially disrupt membranes to release viral nucleic acids. [65;66;67]. But; in any case; the transfer of the genome through the barrier of a cellular membrane into the cytosol is imperative for viral infections to occur. The molecular mechanisms of viral entry are beginning to be understood in increasingly fine details and will be discussed below. NON-ENVELOPED VIRUSES Non-enveloped viruses can seize the cytoplasm of the host cell by direct penetration through the plasma membrane or by a variety of endocytotic mechanisms that collectively lead to the penetration of internal membranes. In such way they do not differ much from what enveloped viruses do while invading cells; that is taking control of common cellular pathways. The main difference with enveloped viruses is posed by the absence of a membrane surrounding the proteinaceous capsid; therefore fusion events cannot happen; but capsid-dependant mechanisms for penetrating the cell membrane or; alternatively; for lysing the endosome are likely to be required. The process of non-enveloped viruses entry is generally started by an interaction with a cellular stimulus (e.g.; receptors; low pH; proteases) at the penetration site which drives a conformational change that in some instances triggers the release of viral components with membrane lytic activity. This is followed by the binding of the hydrophobic component or lytic factor to the cell membrane; which mars the lipid bilayer to allow transport of the viral particle across the limiting membrane. Therefore; membrane penetration seems to be mediated by short; membrane altering; amphiphatic or hydrophobic sequences contained in proteins which have undergone through a conforma-
4 Current Drug Metabolism, 2012, Vol. 13, No. ??
tion transition; allowing such sequences to interact with membranes [56]. Despite several salient differences in the mechanism of entry among non-enveloped viruses; similar events during entry collectively group these membrane lytic peptides: induced modifications of capsid proteins resulting in peptide exposure followed by outward projection of peptides able to interact with host membranes and disrupt them; resulting in the delivery of the viral genome inside the host cell. Examples of these membrane altering sequences of several nonenveloped virus capsid proteins are: i) peptides generated by autocatalytic cleavage of a precursor such as in the case of the (gamma) peptide of nodaviruses [68]; the VP4 and the N-terminal region of VP1 of picornaviruses [69]; and the μ1N of reoviruses [70]; ii) peptides generated by trypsin cleavage such as in the case of VP4 of rotavirus [71]; iii) peptides generated by cellular proteolysis such as the protein VI of adenoviruses [72]; iv) peptides generated by viral proteolysis such as pep46 and additional peptides of birnavirus [73; 74]. Enveloped Viruses All enveloped viruses share common steps of virus entry using two main routes to enter the cell; either by the endocytic or nonendocytic routes. Clathrin coated vesicles; but non-clathrin-coated pits; macropinocytosis or caveolae are all pathways exploited by viruses preferring the endocytic route; while the non-endocytic path involves direct crossing of the plasma membrane at neutral pH. Regardless of the chosen route; the basic mode of entry by enveloped viruses is through membrane fusion; an essential and ubiquitous mechanism in most cellular events. Virally-induced fusion is mediated by viral membrane proteins which undergo remarkable conformational modifications as a consequence of a trigger that is represented either by low endosomal pH or receptor binding. These conformational changes lead to the exposure of hydrophobic peptides; loops or patches (the so called fusion peptides); which then interact with and destabilize one or both of the opposing membranes. A large diversity of conformations has been revealed by the detailed atomic structure of complete ectodomains or core regions of many viral fusion proteins. However the overall view presented is that the structural transition from a pre- to a post-fusion conformation leads to a stable hairpin conformation resulting in the posi-
Stefania et al.
tioning of the two membrane anchors; the transmembrane and the fusion peptide domains; at the same end of the of a trimeric elongated rod-like structure. Three different classes of viral fusion proteins have been identified to date based on their common post-fusion structural motifs [65,75-81]. These are: (i) class I fusion proteins; characterized by trimers of hairpins containing a central -helical coiled-coil structure; (ii) class II fusion proteins; characterised by trimers of hairpins composed of structures; (iii) class III fusion proteins; that show features of both classes. Similar to class I fusion proteins; the post-fusion trimer of class III fusion proteins displays a central helical trimeric core; however; each fusion domain is composed of two fusion loops located at the tip of an enlongated -sheet revealing a striking convergence with class II fusion proteins. Monomers of the post-fusion ectodomains of the three classes are shown in Fig. 1. Viral Fusion Proteins Class I viral fusion proteins tend to form spiky projections on the surface of the virion and are commonly generated as fusioninactive precursors. As a consequence of a proteolytic cleavage; the fusion peptide is exposed at the N-terminus of the protein. The prototype fusion protein of this class (identified in orthomyxoviruses; paramyxoviruses; retroviruses; filoviruses and coronaviruses) [82-86]; forms homotrimers; which; once cleaved; exhibits a surface subunit and a transmembrane subunit anchored to the viral membrane and containing the fusion peptide. This fusion protein extends to form a rod-like structure as a reaction to an activating trigger. The hydrophobic -helical fusion peptide becomes accessible and penetrates into the membrane of the apposing cell. A further conformational change brings together two heptad repeats (HR; the N-terminal HR domain has been generally named either HR1 or NHR; while the C-terminal one has been named HR2 or CHR) located downstream of the fusion peptide; to form a coiled-coil structure resulting in a stable hairpin conformation. Fusion competent proteins of this class share the common 6-helix bundle; which is the functional unit that causes the folding back of the fusion protein upon itself and leads to the close contacts between the viral and cellular membranes in order to allow lipid mixing [87]. Class II fusion proteins present a three-dimensional structure radically different from those belonging to class I and their main
Fig. (1). Post-fusion conformations of class I; II; and III fusion proteins. Post-fusion ectodomains of the three classes form trimers; and only monomers are shown here. The crystal structures of the class I fusion proteins; Influenza virus HA2; a class II fusion protein; Dengue Virus E; and a class III fusion protein; HSV-1 gB; are shown. The monomeric ectodomains are in dark blue; with fusion peptides in red. C-terminal domains; which connect to the virus membrane; are in green. C-terminal linkers; transmembrane domains; and fusion peptides not visible in the structure are represented by dashed green lines; green triangles; and red triangles; respectively.
Viral Peptides for Intracellular Delivery
representatives are members of the Flaviviridae and Togaviridae families [88; 89; 90]. Class II fusion proteins contain predominantly -strand secondary structures; do not form coiled-coils and essentially their fusion peptide is located in an internal position. These glycoproteins are often associated with another protein to form heterodimers lying almost flat on the virion surface. The polypeptide chain of the class II proteins are organised in three globular domains; essentially constituted by -sheets; with the C terminus and the fusion peptide located at the two end of a rod-like molecule. Domain I; which contains the N terminus; is a -barrel with an “upand-down” topology. Two of the connections between adjacent strands in this barrel are long and elaborated; and comprise the “finger-like” domain II with the fusion loop at the tip of the molecule. Domain III; which lies at the opposite end of domain I; has an immunoglobulin-superfamily fold and is connected to the C terminus of domain I by an ~12 aminoacid polypeptide. Activation occurs following the cleavage of the accessory protein leading to an irreversible rearrangement of the fusion protein into a more stable homotrimer which protrudes from the viral envelope allowing the penetration of the internal fusion peptide into the cell membrane. The crystal structure of the class II homotrimers showed a foldback arrangement strikingly reminiscent of that of class I fusion protein; despite the different architecture of the proteins. This movement brings the two membranes close enough to start merging [79]. Class III proteins contain a central trimeric coiled-coil; a hallmark of class I proteins in their post-fusion states; three of their domains are predominantly made of -sheets and their fusion peptide is internal; typical of class II proteins. The structures of the ectodomains of the fusogenic proteins G of vesicular stomatitis virus (VSV) and of the gB protein of HSV-1 were described in 2006 revealing a remarkable structural similarity between the two proteins [91; 92]. Two more glycoproteins have been added to the class III of fusion proteins; namely gp64 from baculovirus [93] and gB from Epstein–Barr virus (EBV) [94]. These proteins share a common structural organization of their five domains. Class III fusion proteins all contain a fusion region (domain I) with an internal fusion peptide organised in two fusion loops flanked by -sheet of a domain with a pleckstrin-like fold (domain II). The domain II is in turn embedded within the largely -helical domain III; which is composed of trimers that give rise to an elongated; rod-like shape molecule. The -helical domain is inserted in domain IV made entirely of -sheets and continues with domain V which is composed of an extended segment connecting the domain IV with the membrane-proximal regions that precede the transmembrane domain [81]. However; the overall activity of the three classes of fusion proteins; irrespective of their structural and biochemical differences; seems to induce membrane fusion in a similar manner. In fact; after fusion activation all fusion proteins will finish up forming a similar hairpin structure. The principal element of the fusion machinery is always represented by a fusion peptide able to insert into cell membranes and to drive regimented membrane destabilization. Despite differences in the mode of activation; the structural motifs used and the differences in initial oligomerization states of viral fusion protein (native trimeric conformation in the case of class I and III fusion proteins vs. homo- or hetero-dimeric conformations in the case of class II fusion proteins); a common refolding mechanism is highly suggestive for the conservation of several steps during viral glycoprotein mediated fusion. Following activation by receptor binding or acidification of the endosomal compartment; the fusion peptide is projected toward the foremost side of the glycoprotein where it inserts into the target membrane. The structural changes in the fusion protein end up in the juxtaposition of the target membrane; held by the peptide or fusion loops; and the viral membrane; held by the trans-membrane region of the fusion protein. Further awesome refolding steps result in the merging of the two lipid lay-
Current Drug Metabolism, 2012, Vol. 13, No. ??
5
ers and the consequent release of the viral nucleocapsid inside the host cells. VIRAL MEMBRANOTROPIC PEPTIDES AND INTRACELLULAR DELIVERY The success in the application of nanomedicines and gene therapy is largely dependent on the development of vectors able to selectively and efficiently deliver the gene or therapeutic agents to the target cells with minimal toxicity. Delivery across cellular membranes involves several mechanisms such as direct transfer through cell surface membrane by lipid membrane fusion or transient permeabilization of the cell membrane; or after endocytosis; transfer across vesicular membranes by lipid disruption; pore formation or fusion. Several of these membrane riorganization steps are involved in the entry of viruses or microorganisms; and are also triggered by protein toxins and defense peptides [95; 96]; related processes are important in biological events like the intracellular vesicle budding and fusion [97; 98] or fusion of cells; sperm-egg fusion or the immune response. Nowadays; non-viral vectors with minimal toxicity and immunogenicity have been developed to partially mimic the mechanism of receptor-mediated viral entry even though early attempts of drug deliverly following a mechanism of receptor-mediated endocytosis have mainly proven to be inefficient due to inability to escape the endosomal pathway [99; 100]. In fact; molecules entering the cells via the endocytic pathway become entrapped in endosomes and eventually finish up in the lysosome; where active enzymatic degradation processes take place. This results in a limited delivery of therapeutic agents to the intracellular target. Many viruses have evolved quite efficient systems for endosomal release [101; 102]; therefore; it is of paramount importance to discover new methodologies to reproduce their behaviour. Since; as described above; viruses may enter cells either through a endosomal pathway or via direct fusion on the plasma membrane; through the activity of membranotropic peptides; it will be interesting to analyse the different structural characteristics of viral entry peptides that may be able to circumvent the endosomal entrapment either favouring the escape from the endosome or by translocating a cargo through the plasma membrane directly into the cytosol and to compare them with known cell-penetrating peptides. Despite the fundamental differences between enveloped and non-enveloped viruses and between different envelope viral fusion proteins or capsid proteins; the hallmark of the mechanism of viral penetration is the presence and membranotropic activity of small peptides which are profoundly different from each other; but share similar entry related events and intrinsic properties of membrane disruption and/or membrane translocation which may be exploited for the delivery of a cargo inside the cell; facilitating transmembrane transport [64]. Membrane fusion and disruption are related processes; although leakage and fusion capacities of peptides do not always correlate and the characteristics and the activities of membranotropic peptides may depend on the particular process and environment involved as well as on their secondary structures. Since not all membranotropic peptides are able to cross the membrane bilayer; it is essential to identify structural characteristics of peptides know to enter the cell membrane to highlight any feature that is involved in the penetration. Thus; an important feature to be considered is the structural requirements for cellular uptake and the ability of membranotropic peptides to interact with the cell surface and lipid moieties of the cell membrane. In this paragraph we will compare the characteristics of membranotropic peptides that have been used for drug delivery (Table 1). In general CPPs are cationic peptides of less than 30 aminoacids and are internalised via energy-dependent or independent mechanisms. Positively charged amino acids; hydrophobicity and amphipathicity are common features shared among them.
6 Current Drug Metabolism, 2012, Vol. 13, No. ??
Table 1.
Stefania et al.
Examples of Commonly Used CPPs and Some of their Properties Related to Interaction with Membranes
Name
Origin
Sequence
AA
Net Charge
Structure Water/lipid
Mean Hydrophobicity
Protein Transduction domains penetratin
Antennapedia Drosophila melanogaster
RQIKIWFQNRRMKWKK
16
+6
Random/- sheet
-2.33
Tat 48-60
Hiv-1 Transactivator (Tat)
GRKKRRQRRRPPQ
13
+8
Random/ Random
-7.27
Pep1
NLS from Simian Virus 40 large antigenti and reverse transcriptase of HIV
KETWWETWWTEWSQPKKKRKV
21
+2
Random/ - sheet or -helix
-2.55
Transportan
Galanin and mastoparan
GWTLNSAGYLLGKINLKALAALA KKIL
27
+4
Random/- helix
0.84
Arg9
Oligoarginine
RRRRRRRRR
9
+9
Random/ Random
-10.0
MAP
Model amphipatic peptides
KLALKLALKALKAALKLA
18
+5
Random/- helix
0.65
Chimeric CPPs
Syntheic CPPs
Viral membranotropic peptides gH625
Glycoprotein gH of HSV typeI
HGLASTLTRWAHYNALIRAF
20
+2
Random/- helix
0.32
MPG
A hydrofobic domain from the fusion sequence of HIV gp41 and NLS of SV40 Tantigen
GALFLGFLGAAGSTMGAWSQPKK KRKV
27
+5
Random/- sheet
-0.63
INF
Influenza HA2
GLFEAIEGFIENGWEGMIDGWYGC
24
-5
Random/- helix
0.66
CADY
PPTG1 peptide
GLWRALWRLLRSLWRLLWRA
20
+5
Random/- helix
2.39
CPPs have a net positive charge at physiological pH; due to their high content of the basic amino acids lysine and arginine. Positive charges are fundamental in the initial steps of internalization that involve the interaction with cell-surface proteoglycans; but also play an important role in the interaction with the nearby phosphate heads of the membrane lipids. Both lysines and arginines interact with negatively charged sulphate and phosphate groups. Arginine residues contribute more than lysine residues to internalization; although both contribute to the net charge of the peptide; the guanidinium group of arginine is believed to be the crucial structural component for the internalization for its special ability to form bidentate hydrogen bonds with phosphate or sulphate groups. CPPs show overall hydrophobicity; an essential characteristic for the interaction with the lipidic part of the cell-membrane. Although; if too hydrophobic the peptide can stick to the membrane rather than internalize. The combination of a charged and a hydrophobic domain is pivotal for the interaction with natural membranes; the charged moieties are involved in the initial interaction with the membrane while the hydrophobic domain is crucial for the insertion. With the exception of polyarginine all other CPPs are amphipathic; consisting of two domains: a hydrophilic (polar) domain and a hydrophobic (non-polar) domain [64]. The peptides can be amphi-
pathic in their primary structure or secondary structure. Primary amphipathic peptides correspond to the sequential assembly of a domain of hydrophobic residues with a domain of hydrophilic residues divided by a spacer domain; while secondary amphipathic peptides are generated by the conformational state which allows positioning of hydrophobic and hydrophilic residues on opposite sides of the molecule. Peptides that have been used for intracellular delivery [6,9,10305] are generally characterised by the formation of -helices; although -sheet structures have also been detected. The length of the helix is fundamental for discriminating the mechanism of action. Peptides longer than 20 amino acids can form -helices of sufficient length to individually span a lipid membrane. The transmembrane orientation may simply destabilise the membrane at sufficiently high peptide concentrations or may oligomerise into either ‘barrelstave’ or toroidal pores. Shorter peptides (
