The Different Cellular Entry Routes for Drug Delivery Using Cell Penetrating Peptides

ABSTRACT

The cell plasma membrane acts as a semi‐permeable barrier essential for cellular protection and function, posing a challenge for therapeutic molecule delivery. Conventional techniques for crossing this barrier, including biophysical and biochemical methods, often exhibit limitations such as cytotoxicity and the risk of genomic integration when viral vectors are involved. In contrast, cell‐penetrating peptides (CPPs) offer a promising non‐invasive means to deliver a broad range of molecular cargoes, including proteins, nucleic acids and small molecules, into cells. CPPs, typically 5 to 30 amino acids long and rich in basic or non‐polar residues, interact favourably with different cell membranes. These peptides have evolved since the discovery of the HIV‐1 TAT peptide in the 1980s, expanding into various CPP families with diverse therapeutic applications. CPPs can form covalent or non‐covalent complexes with their cargo, influencing their stability and efficacy. Based on their sequence properties and interactions, CPPs can be amphipathic or non‐amphipathic, with distinct mechanisms of membrane penetration, such as direct penetration and endocytosis. While their uptake mechanisms are complex and not fully elucidated, ongoing optimization aims to enhance CPP specificity and efficacy. CPPs have demonstrated potential in drug delivery, gene therapy, cancer treatment and vaccine development, addressing key safety and efficiency concerns associated with viral vectors. This review explores the classification, mechanisms of action and therapeutic potential. It focuses on the intracellular vesicular trafficking of CPPs, highlighting their role as transformative tools in advancing cellular therapies and medical treatments.

Cell‐penetrating peptides (CPPs) represent a promising, non‐invasive strategy for delivering a wide variety of molecular cargoes including proteins, nucleic acids, and small molecules into cells. Due to their versatility, CPPs have shown significant potential in areas such as drug delivery, gene therapy, cancer treatment, and vaccine development, offering safer and more efficient alternatives to viral vectors. Comprising 5 to 30 amino acids, and typically enriched in basic or non‐polar residues, CPPs interact efficiently with cellular membranes and employ diverse entry mechanisms, including direct translocation and various forms of endocytosis. In this review, we examine their classification, mechanisms of cellular entry, therapeutic applications, and particularly emphasize their intracellular vesicular trafficking, underscoring their value as innovative tools for advancing cell‐based therapies and modern medicine.

1. Introduction

The cell plasma membrane is a semi‐permeable barrier that acts as a protective layer of cells and is crucial for their survival and function. This barrier enables the selective transport of small molecules through specific channels and transporters, while preventing the entry of macromolecules (Madani et al. 2011). The development of new therapeutic molecules to overcome the limitations of conventional therapies has been limited by this barrier. Accordingly, several cell techniques have been developed to cross this membrane, including biophysical methods (microinjection, electroporation, magneto‐injection), biochemical methods (using amphipathic detergents) and viral vectors (Reissmann 2014). Despite their efficacy, these methods are often not suitable for therapeutic use, may have cytotoxic side effects and, in the case for some viral vectors, have a high likelihood of integrating viral genes into the host genome (Silva et al. 2019). To overcome the various limitations of these approaches, cell penetrating peptides (CPPs) that can be covalently conjugated to therapeutic molecules (later noted cargoes) or form non‐covalent complexes with their cargoes have been seen as promising tools for several decades. However, challenges such as limited cellular uptake and low target specificity have driven researchers to continuously develop new CPPs and deepen their understanding of their cellular entry mechanisms.

CPPs are short peptides, typically consisting of 5 to 30 amino acids, which possess the unique ability to cross cell membranes. These peptides, also known as protein transduction domains, are characterized by their capacity to facilitate the intracellular delivery of various molecular cargoes ranging from small molecules to large proteins and nucleic acids. CPPs are often rich in basic residues like arginine and lysine, which contribute to their overall positive charge at physiological pH, aiding their interaction with the negatively charged cell membranes (Ruseska and Zimmer 2020; Trofimenko, Grasso, Heulot, et al. 2021). They were first discovered in the late 1980s with the TAT peptide encoded by HIV‐1 and later with the Drosophila penetration peptide Antennapedia and have since evolved into a myriad of peptides with diverse applications in medical research (Frankel and Pabo 1988; Derossi et al. 1998; Silva et al. 2019).

CPPs can be classified according to their interaction with therapeutic agents and their sequence properties (Table 1). They also can be modified to form covalent or non‐covalent complexes with their cargoes. Covalent CPP‐cargo conjugates are suitable for oligonucleotides but can interfere with the biological activity of charged molecules such as siRNA/miRNA due to steric hindrances (Deshayes et al. 2005; Tai and Gao 2017). Non‐covalent CPPs, often amphipathic peptides, form stable supramolecular complexes through electrostatic interactions, suitable for the transfer of different biologically active molecules (Deshayes et al. 2005).

Table 1.

Classification of cell‐penetrating peptides based on their chemical properties. B = β‐alanine.

Class	Name	Sequence	Mechanism	Reference
Amphipathic	MPG	GALFLGFLGAAGSTMGAWSQPKKKRKV	Water‐pore formation, inverted micelle	Simeoni et al. (2003)
	αR5W4	RRWWRRWWR	Endocytosis	Walrant et al. (2020); Okafor et al. (2024)
	(WH)5	WHWHWHWHWH	Direct penetration	Shirazi et al. (2018)
Cationic	TAT(47‐57)	YGRKKRRQRRR	Macropinocytosis, clathrin‐mediated and caveolae endocytosis.	Schwarze et al. (1999)
	Penetratin	RQIKIWFQNRRMKWKK	Inverted micelles, macropinocytosis, clathrin‐mediated and caveolae endocytosis.	Derossi et al. (1998)
	hPP10	KIPLPRFKLKCIFCKKRRKR	Endocytosis.	Wang, Ma, Yang, et al. (2016)
	dNP2	CKIKKVKKKGRKKIKKVKKKGRK	Endocytosis	Lim et al. (2015; Xiang et al. (2018)
Anionic	p28	LSTAADMQGVVTDGMASGLDKDYLK PDD	Direct penetration and caveolae endocytosis	Taylor et al. (2009; Yamada et al. (2015)
	Roseltide rT7	CVSSGIVDACSECCEPDKCIIMLPTWPPRYVCSV	Direct penetration and endocytosis	Kam et al. (2019)
	SAP(E)	VELPPPVELPPPVELPPP	Endocytosis	Martín et al. (2011)
Chimeric	Pep‐1	KETWWETWWTEW‐SQP‐KKKRKV	Direct penetration	Morris et al. (2001)
	Transportan	GWTLNSAGYLLG‐K‐INLKALAALAKKIL	Endocytosis, inverted micelle and carpet model	Pooga et al. (1998)
	LP‐C18	KRKRRRR‐ (CH3(CH2)16–COOH)2	N/A	Do et al. (2017)
Hydrophobic	TP10	XAGYLLGKINLKALAALAKKIL	Endocytosis, inverted micelle and carpet model	Tsuchiya et al. (2023)
	SG3	RLSGMNEVLSFRWL	Endocytosis	Gao et al. (2011)
	C105Y	CSIPPEVKFNKPFVYLI	Direct penetration	Rhee and Davis (2006)

When classifying CPPs, their sequence and binding properties are also considered. Primary amphipathic CPPs, such as Transportan, contain hydrophobic and hydrophilic residues and can cross membranes directly (Pooga et al. 1998; Reissmann 2014), whereas secondary amphipathic CPPs, such as Penetratin, form structured alpha‐helices or beta‐sheets when interacting with phospholipid membranes (Derossi et al. 1998; Reissmann 2014; Zhang et al. 2014). Non‐amphiphilic CPPs such as TAT (HIV‐1 transactivator of transcription) and R9 (9‐residues poly‐arginine sequence) are short with a high proportion of cationic amino acids and bind to lipid membranes with a high proportion of anionic lipids (Frankel and Pabo 1988; Rothbard et al. 2000).

In summary, CPPs represent a promising solution to overcome one of the most significant challenges in cellular therapy and drug delivery into cells, especially for molecules that cannot readily cross the cell membrane, including most peptides, proteins, siRNA and plasmid DNA. Their ability to penetrate cell membranes non‐invasively and efficiently, together with ongoing research to optimize their properties and interactions, make CPPs a valuable tool for the advancement of medical treatments. CPP uptake has been observed in a variety of cell types and in combination with different loading molecules. However, the exact mechanism by which CPPs cross cell membranes involves several processes, not all of which are yet fully understood. Some CPPs have been used to deliver chemotherapeutic agents directly to cancer cells, potentially reducing the adverse side effects associated with traditional chemotherapy by minimizing effects on healthy cells (Guo et al. 2020; Li, Li, Xie, et al. 2024, Li, Zhang, Liu, et al. 2025; Zorko et al. 2022). CPPs can also facilitate the delivery of genetic material for the treatment of genetic disorders, offering a potential method to correct or mitigate diseases at a genetic level. Others have been used to enhance the delivery of antigens and adjuvants into cells, improving the efficacy of vaccines, and some are emerging as powerful tools in gene therapy, offering a promising alternative to traditional viral vectors (Oba et al. 2019; Holl et al. 2021). The non‐viral nature of CPPs addresses many of the safety concerns associated with viral vectors, such as immunogenicity and oncogenicity, while providing an efficient means to deliver genetic material into cells. CPP can be classified into various categories. As well they penetrate cells in different ways, which can be classified into two main groups: direct penetration and endocytosis. The CPP classifications, as well as their mechanisms of cell membrane crossing and their fate once entered, will be the focus of the present review.

2. Classification of Cell‐Penetrating Peptides

Cell‐penetrating peptides (CPPs) are classified into several types based on their structural characteristics and the nature of their interactions with cell membranes (Table 1). These classifications help in understanding their mechanisms of entry and potential applications. Here's an overview of the primary types of CPPs but a more detailed classification can be found elsewhere (Deshayes et al. 2005; Reissmann 2014; Ruseska and Zimmer 2020).

2.1. Cationic CPPs

Cationic CPPs are rich in positively charged amino acids such as lysine and arginine. The positive charge enhances their interaction with the negatively charged components of the cell membrane, such as phospholipids and glycosaminoglycans. Peptides derived from the Trans‐Activator of Transcription (TAT) protein of HIV‐1 (Frankel and Pabo 1988) and the synthetic oligoarginine sequences such as R8 and R9 (Rothbard et al. 2000; Duchardt et al. 2007) are classic examples of cationic CPPs (Table 1). These peptides typically rely on their charge density to facilitate cell entry, where the positive charges promote binding to cell membranes, followed by various proposed mechanisms of entry, including direct translocation or endocytosis. TAT and R9 have been extensively used in research to deliver a wide range of cargoes, including proteins, nucleic acids and nanoparticles, into cells due to their simplicity and high efficiency (Gao et al. 2009; Jeong et al. 2016; Xiang et al. 2018; Zhang et al. 2014).

2.2. Amphipathic CPPs

Amphipathic CPPs contain both hydrophobic and hydrophilic domains, enabling them to interact with both the aqueous environment and the lipid bilayers of cell membranes. They can form alpha‐helices or beta‐sheets that facilitate their insertion into lipid bilayers. They often form structures that interact efficiently with membrane lipids, aiding in their uptake. Their structure allows these peptides to disrupt the membrane locally or form pores transiently, facilitating their own uptake or the uptake of cargo. For example, Penetratin is a 16‐amino acids peptide derived from the third helix of the Antennapedia homeodomain protein in Drosophila that combines both hydrophobic and positively charged residues, giving it amphipathic properties (Derossi et al. 1998) (Table 1). It primarily enters cells via endocytosis but can also translocate directly depending on the environment and cargo. They have been shown to effectively transport a range of different bioactive cargoes across cellular membranes (Patel et al. 2019; Tyler et al. 2023).

These CPPs adopt specific secondary structures such as alpha‐helices, beta‐sheets or random coils depending on the environment. Their efficiency can be linked to the stability and dynamics of these structures. The formation of secondary structures can facilitate interactions with membranes and promote endocytosis or direct translocation depending on the structural context.

2.3. Hydrophobic CPPs

Hydrophobic CPPs primarily contain hydrophobic or non‐polar amino acids, which facilitate their interaction with the lipid components of membranes. Some segments of larger protein domains that act as CPPs are predominantly hydrophobic, allowing them to associate with cellular membranes deeply enough to initiate translocation. These CPPs may insert into membranes similarly to amphipathic peptides but rely more on their hydrophobic interactions to cross the membrane barrier. This is the case for C105Y, which was shown to be very effective in entering the cell through direct membrane translocation, in conditions devoid of endocytosis (Rhee and Davis 2006) (Table 1).

2.4. Chimeric CPPs

Chimeric CPPs are designed by combining sequences or functional domains from different CPPs or other functional proteins to enhance their properties such as specificity, efficiency or cargo‐carrying capacity. This is the case, for instance, with fusion peptides that combine, for example, TAT with another sequence to improve specificity or cellular targeting. The mechanisms of cellular entry then depend on the combination of features from the parent peptides, potentially allowing tailored interactions with cells and improved intracellular delivery. An example is the Transportan peptide originally designed by linking the N‐terminal sequence of the neuropeptide Galanin to the wasp venom peptide Mastoparan, therefore exhibiting both hydrophobic and hydrophilic characteristics, enhancing their ability to interact with cellular membranes (Pooga et al. 1998). Pep‐1 is another synthetic peptide designed to improve intracellular delivery of proteins. It includes a sequence that facilitates the formation of a stable complex with the cargo containing peptide sequences from both SV40 T‐antigen and TAT (Morris et al. 2001; Deshayes, Heitz, Morris, et al. 2004) (Table 1). Typically, Pep‐1 enters cells by forming non‐covalent complexes with cargo, allowing for efficient cellular uptake without the need for covalent bonding and has been used to deliver large proteins and even full‐length plasmids into cells (Deshayes, Gerbal‐Chaloin, Morris, et al. 2004; Deshayes, Heitz, Morris, et al. 2004; Li et al. 2017).

These CPPs illustrate the diversity of structures and origins found within this class of peptides. Each type of CPP has been tailored for specific types of cellular uptake and delivery applications, reflecting the dynamic and adaptable nature of CPP technology in biomedicine and has unique advantages and limitations, influencing their suitability for specific applications in drug delivery, gene therapy or as research tools in cell biology. Understanding the fundamental properties of these peptides allows researchers to design more effective CPPs tailored to specific therapeutic needs or experimental conditions.

3. General Mechanisms of Cell Entry

CPPs utilize multiple pathways to enter cells, and the mechanism can vary depending on the peptide sequence, the cargo attached and the cell type. The two primary mechanisms of cellular uptake are direct plasma membrane translocation and endocytosis, which are respectively energy‐independent and energy‐dependent processes (Gehan et al. 2020; Morris et al. 2008; Xu et al. 2019). The precise mechanism of entry for a given CPP may depend heavily on experimental conditions, such as peptide concentration and temperature, as well as the physical and chemical nature of the cargo being delivered. An in‐depth understanding of these mechanisms continues to be a critical area of research, aiming to enhance the efficiency and specificity of CPP‐mediated delivery systems. This versatility and efficacy make CPPs a continually evolving field with significant implications for drug delivery, gene therapy and molecular biology research (Figure 1).

Figure 1.

Illustration of direct cell penetration and endocytosis‐mediated cell penetration mechanisms of CPPs.

3.1. Direct Translocation Mechanisms

Direct translocation refers to the ability of CPPs to cross cell membranes without the involvement of cellular energy. This mechanism is especially intriguing as it allows CPPs to penetrate cells even under conditions where metabolic energy is depleted, such as at low temperatures or in metabolically inactive cells (Bode et al. 2012; Gao et al. 2019; Lee et al. 2019; Takeuchi and Futaki 2016; Walrant et al. 2013). Understanding the specific mechanisms behind direct translocation is critical for improving the design and efficiency of CPPs for therapeutic applications. Several mechanisms of direct plasma membrane translocation have been proposed (Figure 1).

3.1.1. Transitory Membrane Disruption

Some CPPs can induce temporary disruptions in the lipid bilayer of cell membranes. This disruption can be facilitated by the insertion of the peptide into the membrane, followed by a rapid rearrangement of the lipid molecules, creating transient pores or gaps through which the peptide can pass. This process is believed to be facilitated by the high concentration of positively charged residues in many CPPs, which can interact electrostatically with the negatively charged components of the cell membrane, such as phospholipid headgroups (Gehan et al. 2020; Ruseska and Zimmer 2020). However, it should be noted that at higher concentrations, some CPPs prove to be toxic to cells, due to their property to disrupt plasma membrane.

3.1.2. Inverted Micelle Formation

Another proposed model involves the formation of inverted micelles. In this scenario, the CPP interacts with the cell membrane initially through electrostatic interactions and induces the lipid molecules to form a spherical structure (an inverted micelle) within the membrane itself, which in consequences changes curvature (Alves et al. 2010; Swiecicki et al. 2014). Thanks to the hydrophobic environment inside the micelle, the CPP and its hydrophilic cargo are encapsulated inside the core of this micelle, which then crosses the bilayer, releasing the peptide and cargo directly into the cytoplasm.

3.1.3. Carpet Model

Inspired by the action of antimicrobial peptides, a model termed “carpet model” suggests that some CPPs, at sufficiently high concentrations, can coat the surface of the cell membrane, like a carpet (Pouny et al. 1992). This extensive coverage disrupts the membrane's structural integrity, leading to the formation of micelles or direct translocation across the disrupted bilayer. This mechanism is less about forming structured pores and more about creating a general destabilization of the membrane, which facilitates the passive diffusion of peptides into the cell and presumably might be less prone to induce toxicity through damages of the plasma membrane.

3.1.4. Membrane‐Thinning

An alternative to the “carpet” model is the “membrane‐thinning” effect in which the interaction of the negatively charged lipids in the outer leaflet of the membrane and the cationic groups of the CPP causes a lateral rearrangement of the negative charges and a thinning of the membrane. The self‐assembly of the CPPs at the membrane surface provokes a reduction of the local surface tension and allows for intercalation of the CPP within the membrane. After the internalization of the peptide, the membrane reseals (Hedegaard et al. 2020; Ludtke et al. 1995).

3.1.5. Membrane Fusion

Some CPPs can directly facilitate membrane fusion, a process by which the CPP and its cargo merge directly with the cell membrane. This process might involve the formation of transient structures that allow the CPPs to integrate into the lipid bilayer. Membrane fusion typically requires specific sequences or structures in CPPs that can disrupt the phospholipid bilayer sufficiently to allow merging without causing permanent damage. This mechanism can be particularly useful for delivering cargoes directly to the cytoplasm, avoiding intracellular membrane compartment entrapment (Weinberger et al. 2017).

3.1.6. Water Pore Formation

More recently, direct translocation of cationic CPPs was shown to occur via water pores that can form by several mechanisms depending on the sequence of the CPP (Ruseska and Zimmer 2020). On one side, the “barrel‐stave” model is typical of amphipathic α‐helical peptides that form bundles forming a channel with hydrophilic surface at the centre (Copolovici et al. 2014). On the other side, the “toroidal” model has been attributed to positively charged CPPs that accumulate at the plasma membrane surface through ionic interactions causing bending of the membrane and transient pore formation (Bechara and Sagan 2013).

Some CPPs can also induce membrane megapolarization by bringing positive charges on the outer surface of the plasma membrane, thereby decreasing the transmembrane potential. The megapolarization can then trigger the formation of water pores used by CPPs to enter cells. This type of membrane translocation can be inhibited through plasma membrane depolarization or inactivation of specific potassium channels (e.g., KCNN4 in HeLa cells), without affecting endocytosis of CPPs or transferrin (Torriani et al. 2019; Trofimenko, Grasso, Heulot, et al. 2021, Trofimenko, Homma, Fukuda, et al. 2021).

3.2. Endocytosis Pathways

The major itinerary by which CPPs seem to enter cells involves one or several of the many endocytic routes. Unlike direct translocation, endocytic pathways are energy‐dependent, meaning that they require ATP consumption and are influenced by cellular metabolic activity. CPPs can exploit several endocytic mechanisms, each of which involves the engulfment of substances outside the cell, enclosed within a vesicle formed from the plasma membrane (Figure 1). Here's an overview of the main types of endocytic pathways utilized by CPPs:

3.2.1. Clathrin‐Mediated Endocytosis (CME)

This is the most well‐characterized form of endocytosis (Conner and Schmid 2003). It involves the formation of clathrin‐coated pits on hot‐spots of the plasma membrane enriched with phosphatidylinositol‐4,5‐biphosphate (PI(4,5)P2), which interacts with adaptor proteins such as AP2, changing its conformation and thus triggering clathrin assembly (Mettlen et al. 2018). Subsequently, clathrin‐coat components bind to the cytosolic regions of the transmembrane cargo molecules, driving the clustering of all cargo molecules to one region of the membrane, where the clathrin‐coated vesicle will be formed (Kaksonen and Roux 2018). A combination of endocytic actors contributes to the curvature of the plasma membrane including the clathrin coat, the actin filaments and different scission proteins (Haucke and Kozlov 2018). Assembly of the GTPase dynamin into tight oligomers allows constriction of the membrane neck. After GTP hydrolysis, dynamin oligomers further constrict the membrane causing spontaneous transitions to a hemi‐fission and then to a fission state leading to the budding of clathrin‐coated vesicles (Kaksonen and Roux 2018).

Some CPPs bind to the cell surface, where they accumulate and induce the assembly of clathrin and other adaptor proteins, leading to vesicle formation. It has equally been shown that the interaction of CPPs with proteins on the plasma membrane might be important in inducing CME (Kawaguchi et al. 2016). This is particularly significant for the uptake of various CPPs because it can handle a broad range of cargo sizes, from small molecules to larger complexes. For instance, a large body of work has shown that TAT enters cells mostly through CME by interacting with proteoglycans (Richard et al. 2005; Christianson and Belting 2014).

3.2.2. Caveolae‐Mediated Endocytosis

Lipid rafts are microdomains within the cell membrane, rich in cholesterol and sphingolipids, which can serve as platforms for cellular signalling and trafficking. CPPs that interact with lipid rafts or specific receptors may trigger caveolae formation and subsequent internalization into caveosomes. Among the CPPs that use caveolae‐mediated endocytosis are the proline rich CPPs and Transportan (Pujals and Giralt 2008; Säälik et al. 2009).

Caveolae are small (50–100 nm), flask‐shaped invaginations in the plasma membrane that are rich in cholesterol and sphingolipids. Members of the caveolin family, together with cavins and dynamin, but also with the activity of the cytoskeleton, kinases and phosphatases, are important actors of caveolae‐mediated endocytosis (Johannes et al. 2015). Given that caveolae‐mediated endocytosis also bypasses lysosomal degradation, it is particularly useful for delivering sensitive biological cargoes like proteins or RNA.

Moreover, the partitioning of certain macromolecules into lipid rafts eases their internalization via a different endocytic pathway, which is clathrin‐ and caveolae‐independent. However, caveolae are considered a lipid raft subtype and are sometimes classified as identical structures. It is now believed that an additional type of clathrin‐independent endocytosis occurs even in cells devoid of caveolae (Pelkmans 2005; Sandvig et al. 2008). Although less abundant, some CPPs preferentially associate with these lipid rafts, facilitating their internalization through pathways that may involve caveolae or other specialized mechanisms. Such a process has been reported for Azurin and its fragments p18 and p28 and for Transportan (Taylor et al. 2009). Lipid raft‐mediated endocytosis can provide a more targeted approach to internalization, potentially influencing the intracellular routing and efficiency of cargo delivery.

3.2.3. Macropinocytosis

Macropinocytosis does not consist of the ‘envelopment’ of a ligand‐coated particle but involves the non‐specific uptake of extracellular fluid and its contents through large vesicles or macropinosomes. This process is induced by actin‐driven membrane ruffling leading to membrane protrusion that (Futaki et al. 2007). collapse into and fuse with the plasma membrane to generate large endocytic vesicles called macropinosomes (Conner and Schmid 2003). As with the other endocytic routes, micropinocytosis is highly regulated by kinases (such as Src) and Rho‐, Ras‐ and Rab‐GTPases, which trigger the actin‐driven formation of membrane protrusions (Swanson and Watts 1995).

Some CPPs can induce or enhance membrane ruffling leading to the formation of macropinosomes. This pathway is less selective towards the intended cargo and can internalize larger volumes of extracellular material. Macropinocytosis is advantageous for the internalization of large or multiple cargoes simultaneously and accordingly is often exploited by various types of CPPs involved in high‐volume cellular delivery (Tomono et al. 2023; Morimoto et al. 2024). It is also the preferred pathway for non‐CPP dependent cell transduction such as the process termed “iTOP” for induced transduction by osmocytosis and propanebetaine (D'Astolfo et al. 2015). This method induces macropinocytosis allowing the transduction of several cell types that are not suitable for efficient transfection (ex: primary cells) and can also boost CPP activity (D'Astolfo et al. 2015; Wang, Zhang, Zeng, et al. 2016).

3.2.4. Adsorptive‐Mediated Endocytosis

This form of endocytosis is characterized by the non‐specific adsorption of CPPs to the cell surface followed by internalization, which differentiate it from macropinocytosis. It is driven by electrostatic interactions between positively charged CPPs and negatively charged surface components on the cell. Unlike clathrin‐ or caveolae‐mediated pathways, adsorptive‐mediated endocytosis does not involve specific receptor‐ligand interactions but relies on the charge and other physicochemical properties of the CPPs. Unlike the translocation mechanisms, this type of endocytosis involves the internalization of portions of the plasma membrane and is energy‐dependent. This method is less selective than other types of endocytosis but can be highly effective, especially for highly cationic CPPs (Zhang et al. 2013; Lin et al. 2016).

4. Mechanistic considerations

In understanding the mechanisms used by CPPs to penetrate cell membranes, there are several factors that can affect the ability of CPPs to cross the cellular plasma membrane, including their concentration, cell status and external factors.

4.1. CPP and Cargoes Concentration

Higher concentrations of CPPs tend to increase the efficiency of direct translocation, possibly by overcoming the energy barrier associated with membrane disruption or micelle formation, whereas at low concentrations, CPPs may enter cells primarily through energy‐dependent endocytic pathways (Duchardt et al. 2007). Secondly, the structure of CPPs and their interaction with cellular membranes is crucial for their function and effectiveness as delivery vehicles. The amino‐acid residues composition and the sequence directly determine its physical and chemical properties, such as the charge, hydrophobicity and secondary structure of the CPP as well as its solubility, stability and affinity for lipid membranes, all of which influence how it interacts with and penetrates cell membranes. Peptides rich in cationic amino‐acid residues may preferentially enter cells via endocytic pathways, which are facilitated by electrostatic interactions with the negatively charged components of the cell membrane, such as phospholipids proteoglycans and glycosaminoglycans, whereas hydrophobic amino acids can interact with the hydrophobic core of lipid bilayers, aiding in the insertion and penetration of the peptide into the membrane. The ability of CPPs to form secondary structures such as alpha‐helices or beta‐sheets can significantly affect their interaction with cellular membranes and disrupt lipid bilayers (Eiríksdóttir et al. 2010; Ruzza et al. 2010; Walrant et al. 2020). Flexibility and length of a CPP allow a peptide to adapt its conformation to fit the membrane surface, potentially increasing its efficiency in crossing the membrane as longer peptides might have more potential binding sites for membrane components, but they also might be more prone to structural changes that affect their function (Kalafatovic and Giralt 2017).

The cargo itself, including its size, charge and nature can also significantly influence between choosing direct penetration or endocytosis. In fact, larger cargoes often use endocytosis, as direct translocation mechanisms may not accommodate large molecules or complexes. Therefore, although there is limited information on the relative efficiency of the different mode of CPP internalization, one needs to tailor the choice of CPP to the nature and size of the cargo.

4.2. The Status of the Recipient Cell and Its Membrane Composition

Another important factor that affects the internalisation of CPPs, that is sometimes set aside, is the specific lipid or protein composition and physical state of the cells, that can also influence the efficiency of cell penetration (Fretz et al. 2007; Takeuchi and Futaki 2016; Islam et al. 2018). Importantly, different cell types may express varying levels of receptors, or possess different membrane compositions and exhibit diverse endocytic activities, all of which could affect CPP uptake (Zahid and Robbins 2015; Sitinjak et al. 2023). Cardiac targeting peptide, CTP (NH2‐APWHLSSQYSRT‐COOH), is one of the rare CPPs that have been shown to target a specific cell type (Zahid et al. 2010). CTP targets, quite specifically cardiomyocytes, and this has been shown to be based on potassium and sodium levels controlled by voltage‐gated sodium or potassium channels (Zahid et al. 2010; Zahid et al. 2018).

A potentially good target for designing cell‐specific targeting CPPs, are neuronal cells. Neurons, have unique membrane properties and vesicular traffic patterns, which might influence how CPPs are internalized compared to other cell types like fibroblasts or cancer cells (Prescott and Koninck 2002; Valsecchi et al. 2007). There is however very limited evidence that neurons display more efficient CPP internalization than fibroblasts or epithelial cells due to their rich endocytic pathways, high membrane plasticity and anionic lipid composition. Furthermore, the efficiency may be comparable to or slightly lower than that in highly endocytic immune cells, which can internalize CPPs rapidly but may degrade them more quickly. For functional delivery, neurons may have an advantage because their CPP uptake may be enhanced by neuronal activity and receptor‐mediated endocytosis, which can bypass degradation pathways. Therefore, for many therapeutic applications, neurons represent a highly promising target for CPP‐based drug delivery. Thus far, targeting neurons has been done through the integration of nanomaterials with CPPs or the addition of a peptide sequence to specifically target neurons (Tuffereau et al. 1998; Arora et al. 2021; Esteruelas et al. 2025).

4.3. Additional External Physical Parameters

External factors such as temperature, pH, and the presence of serum proteins can also affect the entry pathways of CPPs. For example, low temperatures typically inhibit endocytic processes but may not affect direct translocation. Also, the presence of serum plays a part, as certain CPPs may bind to proteins, which can either hinder or facilitate their uptake depending on the nature of the interaction and the resulting changes in CPP properties (Diaz and Pellois 2023). Furthermore, the metabolic activity of the cell can influence energy‐dependent processes like endocytosis. Cells in different phases of growth or under stress conditions may exhibit altered uptake mechanisms. Starved or metabolically compromised cells might have reduced endocytic activity, favouring ATP‐independent CPP entry mechanisms (Diaz and Pellois 2023).

4.4. Enhancement of CPP's Membrane Penetration

Some modifications can be added to existing CPPs to increase activity. Some CPPs are designed to form dimers or higher‐order oligomers, which can change their membrane interaction properties. Multimeric forms tend to have increased surface area for interacting with membranes or might form structures like pores or channels (Nair et al. 2019; Sauter et al. 2020). Other modifications such as cyclization, ramification or glycosylation can alter the charge or hydrophobicity (Jeong et al. 2016; Do et al. 2017; Schulze et al. 2024; Diedrichsen et al. 2025), thereby modifying their interactions with membranes. For instance, phosphorylation can introduce negative charges to a CPP, potentially increasing its affinity for positively charged membrane components (Gallego et al. 2019; Zhang et al. 2023; Zhao et al. 2024). Also, tweaking the peptide sequence for optimal interaction with a specific cell type's membrane, or adjusting the dosage to balance efficiency with potential cytotoxicity, can significantly enhance the outcome of CPP‐based treatments (Kerkis et al. 2006).

Some treatments, such as DMSO, chloroquine, hypertonic treatments and other chemical treatments, have been shown to enhance the penetration of CPP peptides (D'Astolfo et al. 2015; Ma et al. 2015; Wang, Zhang, Zeng, et al. 2016). These treatments usually don't perpetuate their effect through alteration of cell membrane integrity, but rather by modulating endocytosis rate (Wang et al. 2010; Ma et al. 2011; D'Astolfo et al. 2015). By understanding and manipulating these factors, researchers and clinicians can tailor the delivery system to maximize the therapeutic potential of CPPs (Ruseska and Zimmer 2020; Qian et al. 2024).

4.5. Methodological Approaches to Study CPP Cell Internalization

Therefore, experimental studies focusing on membrane dynamics and the interaction of CPPs with these membranes are crucial for elucidating the mechanisms by which these peptides traverse cellular barriers and deliver their cargo into cells. These studies often employ a variety of techniques to visualize and quantify the interactions between CPPs and cell membranes, shedding light on the biophysical processes involved. They include (i) fluorescence spectroscopy as fluorescently labelled CPPs allow researchers to monitor the peptide's localization, concentration and environmental changes, (ii) circular dichroism spectroscopy allowing to follow the changes in structure of a CPP upon interaction with membranes, (iii) atomic force and cryo‐electron microscopy that can visualize the interaction of CPPs with membranes at the nanoscale, or high‐resolution imaging of CPPs with membranes and vesicles, capturing the formation of complexes and translocation events, (iv) surface plasmon resonance to study the real‐time binding kinetics and affinity of CPPs to lipid bilayers or cell membrane components, (v) liposome leakage assays to determine the membrane‐disruptive abilities of CPPs, elucidating mechanisms of action such as pore formation or membrane fusion, (vi) cell‐based assays including confocal microscopy and flow cytometry, used to study the uptake, localization, and intracellular trafficking of fluorescently labelled CPPs in living cells and (vii) biophysical simulations to provide insights into the interactions at the atomic level between CPPs and lipid bilayers can also give additional mechanistic information (Pooga et al. 1998; Richard et al. 2005; Bond and Khalid 2010; Walrant et al. 2020; Okafor et al. 2024).

5. The CPPs Along the Different Endosomal Compartments

The previously described experimental approaches have significantly advanced our understanding of how CPPs interact with cell membranes, providing a foundation for improving CPP design for therapeutic applications. However, an aspect that has not been fully elucidated is the fate of CPPs once endocytosed. Indeed, this is a significant aspect that one must keep in mind to be able to deliver intact cargoes to the desired destination and possibly escape the degradative pathway.

Endosomes are dynamic structures that undergo coordinated fusion and fission events, allowing intense intracellular signalling (Gautreau et al. 2014). Early endosomes mature into multivesicular bodies (MVBs), late endosomes, and finally, lysosomes, where degradation of the endocytosed material occurs (Huotari and Helenius 2011; Scott et al. 2014). The endocytosed material can also be recycled back to the plasma membrane or trafficked toward other cellular compartments such as the Golgi apparatus for instance, from which material can then be transported back to the plasma membrane (Huotari and Helenius 2011; Naslavsky and Caplan 2018). Each stage of endosomal trafficking is minutely regulated by the sequential recruitment of a set of endosomal proteins and lipids such as the Rab GTPases and phosphoinositides. For example, Rab5, activated by its guanine‐nucleotide exchange factor (GEF) Rabex‐5, controls the local generation of phosphatidylinositol‐3‐phosphate (PI(3)P) by recruiting the Vps34 PI3 kinase to early endosomes (Buckles et al. 2020; Tremel et al. 2021). This in turn, leads to recruitment of early endosome antigen 1 (EEA1) via its capacity to bind PI(3)P through its FYVE domain. EEA1 can also directly interact with the active GTP‐bound form of Rab5. The ability of EEA1 to bind simultaneously Rab5 and PI(3)P on separate vesicles makes it a tethering protein that contributes to endosomal fusion (Murray et al. 2016). The subsequent and concomitant inactivation of Rab5 by its specific GTPase‐activating protein (GAP) induces the activation/recruitment of Rab7 to these endosomes, thus becoming more mature endosomes that will end up as lysosomes. Hence, the Rab5/Rab7‐controlled endocytic pathway is currently seen as the only molecularly characterized route taken by endocytosed material that ends up in lysosomes (Langemeyer et al. 2018).

5.1. CPP in Different Compartments of the Endo‐Lysosomal System

Recent studies using modern cell biology approaches on KCNN4 knockout HeLa cells, to drastically reduce direct membrane translocation, have recently shown that several cationic CPPs (TAT, R9, Penetratin, MAP, TAT‐Ras‐GAP_317‐326 and Transportan) enter cells by CME and micropinocytosis. These CPPs are first located in EEA1‐positive early endosomes and later Lamp1‐positive vesicles that are non‐acidic and probably non‐degradative (Trofimenko, Homma, Fukuda, et al. 2021). Surprisingly, very little colocalization of these CPPs with either Rab5 or Rab7 was observed, even when cells were kept at 20°C to slow down the kinetics of endosomal transition (Trofimenko, Homma, Fukuda, et al. 2021). A siRNA screen for the various Rabs indicated that this pathway requires, in fact, the small GTPase Rab14, unlike the classical endocytosis pathway that is regulated by Rab5 and Rab7 (Trofimenko, Homma, Fukuda, et al. 2021). It is of note that while this study showed that Rab14 is mandatory for the maturation of CPP‐containing endosomes, it also reported that Rab14 is not a specific marker for the CPP endocytic pathway, as Rab14 was also found in CME endosomes containing (ex: transferrin‐positive vesicles) as on CPP‐containing endosomes, and to a lesser extent on dextran‐positive structures (Trofimenko, Homma, Fukuda, et al. 2021).

However, similar experiments involving pharmacological treatments, siRNA silencing and colocalization experiments with GFP‐Rab fusion proteins, indicated that the poly‐arginine/tryptophane CPP, αR5W4, is internalized by an ATP‐dependent endocytosis pathway involving both Rab5 and Rab14 endosomes routes (Okafor et al. 2024). This points to the fact that different entry pathways may be employed by a similar CPP in different cell lines. Additionally, this αR5W4 derived CPP was also found in Rab4 and Rab11 positive endosomes suggesting that a significant portion may be recycled to the plasma membrane (Okafor et al. 2024). This may be of particular interest in terms of potential therapeutic use, as this CPP may thus be capable of multiple cellular entries from the outer cellular space where it could capture molecules for delivery to the cytosol before degradation by the lysosome (Figure 2).

Figure 2.

Entry routes of αR5W4 derived Cu‐shuttle in HeLa cells. The fluorescence of AKH‐αR5W4‐NBD is quenched in the medium upon binding to Cu. The Cu‐shuttle is endocytosed via Rab5 and Rab14 endosomes in which Cu rapidly dissociates from the shuttle as seen by an increase in fluorescence. Then, the shuttle is either recycled through Rab 4 and Rab11, recycling endosomes or degraded through the endolysosomes in a pathway involving Rab7‐positive vesicles. Adapted from (Okafor et al. 2024).

5.2. CPP Escape of the Endo‐Lysosomal System

Once the CPPs and their associated cargoes have successfully crossed the cell membrane, their intracellular fate becomes a critical factor in determining the overall efficacy of this delivery method. Understanding the subsequent behaviour of CPPs and their cargoes inside cells is also essential for optimizing their therapeutic potential and minimizing unwanted effects. Some CPPs are capable of endosomal escape through several mechanisms such as influx of protons or water into the endosomes, leading to osmotic swelling and eventual rupture of the endosomal membrane, or a direct interaction with the lipid components of the endosomal membrane, destabilizing the membrane sufficiently to release the endocytosed cargo into the cytoplasm (Soni et al. 2024). Cargo binding to CPPs within the endosomal compartments may also be sensitive to the gradual pH decrease along the endocytic pathway, leading to potential delivery of the cargo or dissociation of CPP‐cargo complex. Of interest, we have recently shown that a Cu‐shuttle containing the αR5W4 CPP released Cu from the endosomes within minutes after cell entry thanks to the increasing acidity of endosomes (Okafor et al. 2024).

Once inside the cytoplasm, CPPs and their cargoes can interact with various cellular components and pathways. The interaction can be passive, driven by diffusion or active, involving specific binding to intracellular proteins and participation in cellular trafficking systems. The ultimate location and interactions within the cell can greatly influence the functionality of the delivered cargo, affecting aspects like gene expression (in the case of nucleic acids) or enzyme activity (for therapeutic proteins). Therefore, enhancing the delivery of CPPs into the cytosol is a critical aspect of improving their effectiveness as delivery vehicles for therapeutic agents. Various strategies can be employed to maximize cytosolic delivery while minimizing entrapment in endosomes or degradation by cellular enzymes including, (i) incorporating compounds like polyethylenimine into CPP formulations that can induce a proton sponge effect, leading to endosomal swelling and rupture, facilitating the release of the CPP and its cargo into the cytosol. Equally co‐delivering CPPs with agents that inhibit endosomal acidification (like chloroquine) that can raise the pH of endosomes, reduces enzyme activity and increases the chance of CPP escape into the cytosol (Ma et al. 2011; Hu et al. 2014; Akbarzadeh et al. 2019), (ii) adding fusogenic lipids or peptides that destabilize the endosomal membrane thereby promoting the escape of CPPs into the cytosol (Ye et al. 2012; Kim et al. 2019), (iii) chemical modification such as hydrocarbon stapling can stabilize alpha‐helical structures in CPPs, enhancing their ability to interact with and penetrate cellular membranes (Walensky and Bird 2014), (iv) encapsulating CPP‐cargo complexes in nanocarriers like liposomes or biodegradable polymeric or gold nanoparticles that can protect CPPs from degradation and promote endosomal escape or enhance their cellular uptake and facilitate their intracellular release under appropriate conditions. These carriers can be engineered to release their contents in response to specific intracellular triggers (Walrant et al. 2013; Silva et al. 2019; Gessner and Neundorf 2020; Yang et al. 2021; Reich et al. 2024) and (v) use of light‐ or thermo‐sensitive compounds that induce rupture of the endosomal membrane, releasing the CPP and its cargo into the cytosol (Bartlett et al. 2013; Behzadipour and Hemmati 2024). These strategies collectively aim to overcome the natural barriers within the cell that limit the efficacy of CPPs. By combining these methods or improving upon them with innovative technologies, the potential of CPPs as powerful tools for targeted therapy and molecular diagnostics can be improved.

6. CPP in Clinical Trial

Several CPPs have or are currently undergoing clinical trial to deliver treatments to various organs. p28, a CPP derived from the cupredoxin azurin with a preference for cancerous cells, has therefore been used in a few clinical trials to deliver anti‐cancer agents to tumours (Taylor et al. 2009; Yamada et al. 2009). This particular CPP is capable of stabilizing/increasing p53 in cancerous cells, leading to cell cycle arrest. In phase I clinical trials, p28 proved safe to both children and adults at the highest dose levels and was able to increase expression of p53 in tumours (Warso et al. 2013; Lulla et al. 2016). Another clinical CPP candidate is the drug PGN‐EDO51 (NCT06079736), which is the fusion of an exon‐skipping oligonucleotide with a CPP peptide developed for the treatment of Duchenne muscular dystrophy (DMD). In nonclinical studies and a phase 1 clinical ongoing trial in healthy volunteers, PGN‐EDO51 was shown to effectively induce skipping of the target exon and dystrophin production (Mellion et al. 2023, Mellion et al. 2024).

Nomlabofusp (NCT06681766) is a drug currently in a phase 1 clinical trial to treat Fridreich's ataxia. The molecule is a combination of the complete human frataxin and the TAT CPP. Nomlabofusp has been shown to increase levels of mature human frataxin in patients (Baile et al. 2025). Another CPP drug in an active clinical trial is TB511, which is intended for use in advanced solid tumours. TB511 is a fusion between the amphiphilic CPP, Melittin that targets type 2 Tumor‐associated macrophages and pro‐apoptotic peptide (NCT06400160).

Over the years, more and more CPPs have been included in clinical trials and appear advantageous as they are generally well tolerated and don't induce an immune response. However, given the fact that they are generally injected intravenously in patients, a few participants in clinical trials have reported transient infusion‐related reaction (Warso et al. 2013; Lulla et al. 2016). Going forward, modifications to the peptides or encapsulation, that reduce susceptibility of CPP drugs to degradation, will be needed to permit oral drug delivery (Nicze et al. 2024).

7. Conclusion and Perspectives of Improvement

Advancements in the design and synthesis of CPPs, as well as a deeper understanding of their interaction with cellular membranes and intracellular pathways, continue to improve their utility in drug delivery. The development of stimuli‐responsive CPPs, targeted delivery systems, and hybrid approaches combining CPPs with other delivery technologies are promising areas that may address current limitations and expand the potential of CPPs in clinical applications. Despite extensive studies, the exact mechanisms by which different CPPs enter cells remain not fully understood. The ambiguity around whether a particular CPP uses direct translocation, endocytosis, or a combination of pathways under different conditions complicates the design of more efficient CPPs and will require further studies. In addition, there is a significant need to develop targeting strategies that could allow CPPs to selectively deliver cargoes to specific cell types or tissues, minimizing side effects and enhancing therapeutic outcomes. Addressing all these challenges requires a multidisciplinary approach, combining insights from cell biology, chemistry, pharmacology and materials science. Advances in CPP design, such as incorporating targeting motifs, stabilizing modifications, or smart, responsive systems, along with more in‐depth studies into their mechanisms of action, will be essential for overcoming current limitations and enhancing the clinical applicability of CPP‐based therapies.

Author Contributions

All authors revised the manuscript.

Conflicts of Interest

The authors declare no conflicts of interest.

Data Availability Statement

Data sharing is not applicable to this article as no new data were created or analysed in this study.