Cell-penetrating peptides improve pharmacokinetics and pharmacodynamics of anticancer drugs

ABSTRACT

This review concentrates on the research concerning conjugates of anticancer drugs with versatile cell-penetrating peptides (CPPs). For a better insight into the relationship between the components of the constructs, it starts with the characteristic of the peptides and considers its following aspects: mechanisms of cellular internalization, interaction with cancer-modified membranes, selectivity against tumor tissue. Also, CPPs with anticancer activity have been distinguished and summarized with their mechanisms of action. With respect to the conjugates, the preclinical studies (in vitro, in vivo) indicated that they possess several merits in comparison to the parent drugs. They concerned not only better cellular internalization but also other improvements in pharmacokinetics (e.g. access to the brain tissue) and pharmacodynamics (e.g. overcoming drug resistance). The anticancer activity of the conjugates was usually superior to that of the unconjugated drug. Certain anticancer CPPs and conjugates entered clinical trials.

GraphicleAbstract

KEYWORDS:

• anticancer CPPS
• anticancer drugs
• cell-penetrating peptides
• clinical trials
• conjugates
• CPP
• internalization
• preclinical studies

1. Introduction

Despite the great progress in pharmacotherapy, many diseases, as for example cancers, are still a great challenge for the clinicians. Their pharmacological treatment is often unsuccessful and burdened with severe complications and a high mortality rate. The main reason for this fact is the unsatisfactory efficacy of the drugs, which results from their pharmacokinetic and/or pharmacodynamic features, and the most important of which refer to low cellular bioavailability, poor penetration to the target, toxicity and development of drug resistance.

The ability to permeabilize in the membrane and gain access to the cytosolic compartment is largely dependent on the physicochemical qualities of the therapeutic entities including the most relevant such as lipophilicity (determined by drug hydrophilicity/hydrophobicity), polarity and molecular size. Among them, a fairly but not excessively hydrophobic character is most conducive to the drug transport; very large, polar hydrophilic molecules are being poorly internalized similarly as extremely hydrophobic (the former ones will fail to enter the membrane, the latter will enter readily but may have difficulty in leaving it).

Briefly, getting a drug to its target can be a considerable challenge because the compound has to cross fewer or more barriers along its way. They comprise endothelial (the blood–brain barrier [BBB]) and epithelial (gastrointestinal mucous, skin) sheets, which are represented by tightly interconnected cells resulting in highly impermeable structures that compartmentalize different physiological microenvironments.^1–4 Thus, transepithelial and transendothelial transports of drugs may become limited or even arrested

On the other hand, cellular internalization of the drug may be completely changed under pathological conditions. Such a phenomenon is characteristic of anticancer therapy. Intratumoral drug penetration and direct contact with the malignant cells are hampered by deregulation and disorganization of the tumor stroma (it becomes stiff and dense due to its reduced turnover and excessive production) and the abnormal blood flow with preference for the peripheral tumor region.^1,2

At the cellular level, most anticancer drugs act on membrane surface receptors or have intracellular targets. In the latter case, a drug must cross the cellular membrane that constitutes an additional barrier for it. A further complication for drug traversing the cellular membranes is their remodeling induced by cancer itself, which brings about complex changes in the membrane structure and composition. They lead to a drop in anticancer drug efficacy manifested by simultaneous cancer cell proliferation, avoidance of apoptosis and development of cellular resistance.^5–8

Generally, anticancer therapy bears a high incidence of toxicity due to a low therapeutic index of the drugs and the simultaneous necessity for the usage of high doses in order to produce the clinical effect. An explanation for these unfavorable parameters of the anticancer therapy is lack of drug selectivity (they damage not only cancerous cells but also healthy rapidly growing ones) and poor bioaccessibility to the tumor cells. Another shortcoming of this therapy is multidrug resistance (MDR), which constitutes one of the dominant reasons for its failure. Versatile mechanisms are involved in this process including decreased drug uptake, increased drug efflux, reprogramming of the target structure, enhanced enzymatic inactivation, etc.

The presented limitations of cancer treatment have been imperative to the development of diverse strategies of targeted drug delivery to tumor tissue by exploiting passive and active technologies. The passive technologies may utilize such phenomena as charge selectivity (preference for negatively charged tumor membranes over those of neutral healthy ones)^9–11 and/or enhanced permeability and retention (EPR) effect that arises from the unique anatomical and pathophysiological characteristics of solid tumors (e.g. leaky vasculature, impaired lymphatic drainage).^12,13 The active ones involve the usage of special ligands (e.g. transferrin, folate, epidermal growth factor, antibodies) that target molecules or receptors within the tumor cells. It should be emphasized that the EPR effect constitutes an indispensable step for starting the active targeted process.¹³

An attractive application of the EPR effect in tumor chemotherapy comprises therapeutics in the form of liposomes, polymers or micelles.^14–16 This effect has also been utilized by experimental technologies based on cell-penetrating peptides coupled with diverse types of cargoes (low molecular weight anticancer drugs). The CPP may be a part of a simple construct or a component of a macromolecular delivery system.

CPPs constitute a class of oligopeptides rich in basic amino acids characterized by exceptional translocation properties across cell membranes including those which form vital biological barriers without significant interference with their integrity. These peptides originate from a wide variety of sources (as for example humans, mice, viruses, synthesis) and usually carry a positive net charge. It is generally accepted that both endocytosis and direct translocation contribute to the internalization process of the CPPs. Besides being inert vectors for translocating and transporting cargoes into the intracellular milieu, some of them also show biological activity [17–22; 23–26].

This review will focus on CPPs conjugated covalently or noncovalently with anticancer drugs approved by the FDA. These constructs will be presented in terms of the anticancer efficacy and clinical potential in different types of tumors.

2. Classification of CPPs

CPPs are a class of diverse short sequence peptides (usually <30 amino acids) of natural (protein-derived or chimeric) or synthetic origin.^27,28 There are several classifications of the CPPs, but none of them is ideal to the very end. They adopt versatile criteria, and the most often used is that which considers the physical–chemical properties of the amino acid components of the peptide. According to this classification they can be categorized into three main classes, i.e. hydrophilic (e.g. Tat), hydrophobic (e.g. Pep-7) and amphipathic (e.g. TP10) peptides. The latter ones contain both hydrophilic and hydrophobic domains and include primary (e.g. pVEC) and secondary (e.g. azurine-derived p28 peptide, proline-rich CPPs) compounds with peculiar α-helical or β-sheet structures.^29,30 With respect to the net charge of the CPPs, most of them (83%) are cationic. Anionic CPPs do not form a class of their own³¹ nevertheless, SAP(E) and p28 may serve as examples of these peptides.^32–35

3. Internalization mechanisms of CPPs

The exact mechanisms by which CPPs cross the cell membrane still remain to be carefully clarified. One of the reasons for this fact is that the cellular membrane itself constitutes a highly complex and heterogeneous structure composed of specific domains as glycoproteins, polysaccharides and lipids. Moreover, the bending and fluidity properties of the membrane change together with local modifications in its potential.

According to the available data, CPPs are internalized by two possible main pathways, i.e. endocytosis (energy dependent) and/or direct translocation (energy independent). Endocytosis is of vital significance for the cationic CPPs and an exclusive mechanism for the anionic peptides.^29,33,35 The first phase which precedes endocytosis is usually an electrostatic interaction with negatively charged cell surface components such as proteoglycan glycosaminoglycan (GAG) platform (the major constituent of the extracellular matrix). This process is followed by actin remodeling and GTPase activation resulting in changes in membrane fluidity.^7,36

It is not surprising that the anionic CPPs do not share the electrostatic mechanism with cationic peptides, but instead they use another process as the first step of cellular entry, i.e. peptide aggregation on the membrane surface.^{33–35,37–39}

Next, the peptides undergo endocytosis, which comprises several diverse pathways classified as macropinocytosis, clathrin/caveolae-mediated endocytosis and clathrin/caveolae-independent endocytosis.^29,40–42 Some CPPs have been shown to use more than one of these internalization routes, even simultaneously.⁴² During endocytosis, a CPP remains entrapped in endosomes and therefore, after internalization a peptide must escape to the cytosol to avoid enzymatic degradation from lysosomes.^29,41 Entrapment in the endosomal compartments appears to be the principal limiting factor in the efficient intracellular delivery of functional macromolecules, e.g. drugs. Hence, versatile approaches have been undertaken to facilitate the escape of CPPs from the endosomes.²⁹

On the other hand, direct translocation involves peptides entering straight through the cytosol after recruiting negatively charged phospholipid headgroups and/or exploiting membrane defects. Additionally, this process requires permanent or temporary membrane destabilization for internalization to occur. As a process which requires no energy, direct translocation is regarded as a single-step episode including mechanisms involving the formation of pores (“toroidal” or “barrel-stave”),^43–45 inverted micelles (entrapment of the peptide by membrane curvatures or invaginations and its release into the cytosol by their inversion)^46,47 or a “carpet” with the positively charged segments of the peptide covering the negatively charged membrane (local reorganization and disruption of the membrane due to a change in the CPP secondary structure as the basic residues turn toward the membrane surface and the hydrophobic residues interact with the hydrophobic membrane core).^{29,30,40,41,48–53}

Whether a CPP utilizes endocytosis, direct translocation, or both processes for internalization depends on many factors, as for example the chemical structure of the peptide,⁵⁴ temperature,⁵⁵ the targeted cell type,⁴⁰ the condition of the lipid membrane.⁵¹ Thus, it is rather impossible to ascribe a specific mechanism of internalization to each individual CPP. Taking into account one of the best known CPP, i.e. Tat, it employs, depending on the concentration, both direct translocation (forming transient pores)⁴⁴ and endocytosis (caveolae-mediated pathway or macropinocytosis).⁵⁶ On the other hand, the few known anionic peptides utilize only endocytosis (caveolae-dependent pathway) as a mechanism of cellular uptake.^33,57,58 In addition, the cargo (its type, size and charge) as well as the CPP-cargo formulation design may be decisive for the mechanism involved in membrane permeation.⁵⁹ It is likely that multiple mechanisms are active simultaneously, and the down-regulation of one pathway might lead to an upregulation of the other.⁴⁰

Irrespective of the route of internalization, the majority of the CPPs localizes in the cytoplasm or the nucleus.^28,59,60 They may also provide subcellular organelle delivery as happens in the case of mitochondria-specific CPPs, i.e. mitochondria-penetrating peptides or vesicle-based CPPs.^61,62 It is worth stressing that the nucleus and the mitochondria are relevant targets for many cargoes including anticancer drugs.

4. The role of CPPs in cancer treatment

4.1. CPPs as vectors for anticancer drug delivery

4.1.1. Interaction of CPPs with cancer-modified cell membranes

The interaction of a CPP with cancerous cells should be considered in light of the current concept of cancer-induced modifications within the cell membrane itself. They concern both its heteropolysaccharide and lipid components, ensuring the uncontrolled growth, progression and invasiveness of the tumor cells.

The heteropolysaccharides such as GAGs are constituents of proteoglycans and include several categories: chondroitin sulfate, dermatan sulfate, heparan and heparin sulfates, hyaluronan. As a rule, they are overexpressed in cancerous membranes. Moreover, GAGs undergo complex disorganization which concerns their structural pattern and composition. They are implicated in various phases of carcinogenesis as for example the activation of crucial cell signaling cascade, epithelial/mesenchymal transition, the adaptive protection of malignant cells to natural killers.^6–8 Moreover, the higher abundance of GAGs in tumor cell membranes results in their increased anionic profile relative to those of non-transformed ones.⁷

Also, the lipid component contributes to the anionic character of the cancerous membrane due to the exposition of the negatively charged phosphatidylserine (instead of electrically neutral zwitterionic lipids) in its outer leaflet which correlates with a more acidic pH of their external media as well. Furthermore, there are changes in cholesterol metabolism leading to either a lower or higher content relative to the type of cancer, stage and sensitivity. All these modifications result in denser packing of the lipids and a decreased fluidity of the membrane, turning the latter more rigid and less permeable for the drugs.^57,63–65

Generally, the CPPs possessing mostly cationic nature due to arginine and lysine residues favor the negatively charged tumor membranes over those of healthy cells, which are electrically neutral. Of great importance is the electrostatic interaction of the positively charged arginine guanidinium head groups with the negatively charged GAGs, sialic acid or phospholipid head groups displayed in abundance on the surface of the tumor cells. This type of interaction ensures that the CPP indicates a certain level of selectivity to tumor cells referred to as passive charged selectivity. Although it is not complete, the CPPs themselves are not reported to produce toxicity below the concentrations of hundreds of μmoles.^9–11

4.1.2. Improvement of anticancer selectivity of CPPs

An innovation with respect to the classic peptides is the group of tumor-homing CPPs, which are originally obtained by mRNA display technology. The most characteristic feature of them is that they are able not only to distinguish between cancer and normal cells but also target specific tumor cell types. One of them which has been described in a more detailed way is CPP44. A critical role in its internalization plays endogenous M160 protein, which is overexpressed in myelogenous leukemias and hepatic tumors.⁶⁶

Another method of increasing the tumor cell selectivity of CPPs is utilization of a homing device such as an antibody or a targeting ligand (e.g. RGD peptides, folic acid, transferrin, hyaluronic acid) that will enable to attach to tumor cells through antigens or receptors on their surface, usually overexpressed in certain tumor types. For example, folate receptors are highly expressed in ovarian, cervical, breast, kidney, colorectal, lung and brain tumors.^28,67–74 Versatile CPPs (e.g. octaarginine – R8, Tat) have been functionalized by a homing device demonstrating the potential usefulness for enhancing the delivery of a cargo to its target.⁷⁵

Also, there are novel active strategies, which adopt various methods of temporal inactivation of the CPPs by neutralizing their charge or placing a steric hindrance that restrains the nonspecific penetration of the peptides in normal tissue. The CPPs regain their activity within the tumor microenvironment thanks to specific triggers derived from intrinsic stimuli of the target site (acidic pH, MMP-2/9) or their extrinsic counterparts (topical application of heat, light, magnetism). Within this paradigm of controlled CPP delivery, the distinctive examples comprise activable CPPs, ATTEMPT (Antibody Targeted Triggered Electrically Modified Prodrug-Type strategy), Pop-up, cleavable protecting polymer strategies. A detailed description of all of them has been delivered by others.^{28,67,73,75–79}

4.2. CPPs with additional anticancer activity

The privileged access of the CPPs to the tumor cells does not exhaust entirely their significance in the prospect of anticancer therapy. As has been indicated, some members of both cationic and anionic CPPs possess anticancer activity per se, which may be primarily accomplished by a lytic effect on the cancerous cell membrane or non-lytic actions. The diverse mechanisms responsible for this activity are summarized in Table 1. Taking into account the primary mode of action, the CPPs in question may be divided into four categories and each of them with the most recognizable representatives is outlined below.

Table 1. Representatives of CPPs with anticancer activity, mechanisms of action investigated on versatile experimental models

CPP	Anticancer activity	Mechanism of anticancer action	Experimental model		References
			In vitro (cell lines)	In vivo (animal species)
TP10	Destabilization or disruption of cell membrane	Direct interaction with negatively charged membrane components	HEK293, HeLa, OS143B, HEL299		^Citation18
(sC18)2	Destabilization or disruption of cell membrane	Direct interaction with negatively charged membrane components	HEK293, MCF-7, PC-3, HeLa, HCT-15, A2058, M24, HT168-M1/9, A549, H1975, H1650,MDA‐MB‐231, DU145, HT‐29, HepG2, PANC‐1, HT1080, PE/CA‐PJ15, PE/CA‐PJ41		^Citation80,Citation81
Azurine p28	Inhibition of proliferation (G2/M phase cell cycle arrest), neoangiogenesis and induction of apoptosis	1) p53-mediated: blocking COP1 (E3 ubiquitin ligase) and inhibition of proteosomal degradation of p53 via HDM2 – independent pathway 2) Non-p53-mediated: noncompetitive inhibition of VEGFR-2 and FGFR-1 kinases with the resultant reduction of the phosphorylation of their downstream targets Fak and Akt	HUVEC, UISO-Mel-2, UISO-Mel-6; HEK-293, MDA-MB-231, HCT-116, UISO-Mel-23, MCF-10A, LNCaP (p53wt, AR+), DU145 (p53mut, AR−), PC-3 (p53null, AR−), IMR-32 (p53wt), SK-N-BE2 (p53mut), ZR-75 (p53wt), U87 (p53wt), LN229, HK-2 (p53mut),SK- Mel-29 (p53wt), SK-Mel-23 (p53mut)	Melanoma xenograft athymic mice, virus-free CD-1 [Crl:CD1(ICR)] mice, BALB/c mice, Cynomolgous sp.	^Citation34,Citation82–84
Buforin IIb	Induction of mitochondrial pathway of apoptosis	Activation of pro-caspase-9 and pro-caspase-3	HeLa, Jurkat, human fibroblasts, mouse embryonic fibroblasts + 60 human tumor cell lines (data from NCI testing program)	Lung carcionoma xenograft BALB/c (nu/nu) mice	^Citation85
BR2	Induction of mitochondrial pathway of apoptosis	Activation of pro-caspase-9 and pro-caspase-3	HeLa, HCT116, B16/F10, NIH 3T3, HaCat, BJ		^Citation86
Lactofericin B	Induction of mitochondrial pathway of apoptosis and destabilization or disruption of cell membrane	1) Activation of caspases and cathepsin B 2) Direct interaction with negatively charged membrane components	B16-BL6, L5178Y-ML25, CCRF-CEM, MDA-MB-435, Bcl-2, CAL27, Het-1A, MCF-7, CRL-3271, MDA-MB-468, MDA-MB-231	Melanoma xenograft C57BL/6 mice, CDF1 and C57BL/6 mice, Syrian golden hamsters (Mesocricetus auratus)	^Citation87–91
LTX-315	Destabilization or disruption of cell membrane, induction of immunogenic cell death	1) Direct interaction with negatively charged membrane components 2) Infiltration with antitumor CD8 + T-cells	U2OS, B16F1, C57BL/6, MRC-5, HUVEC, A375, A375,	KP mice, BP mice, fibrosarcoma and melanoma xenografted C57BL/6 wild type mice, melanoma xenograft Rag2–/–mice, sarcoma xenograft PVG rats	^Citation92–96
(R/W)16	1) Reversal of the tumoral phenotype 2) Reduction in the ability of malignant cells to grow in anchorage independence 3) Reorganization of intracellular actin dynamics in stress-fibers	Interaction with extra-and intracellular protein targets	NIH3T3, NIH3T3 EF		^Citation36
Legend:
Human cancerous cell lines: head and neck squamous: PE/CA‐PJ15, PE/CA‐PJ41; liver: HepG2, CAL27; pancreatic: Panc-1; colon: HT-29, HCT-116, HCT-15; breast: MDA-MB-231, MDA-MB-468, MCF-7, ZR-75;
Cervix: HeLa; prostate: LNCaP, DU145, PC-3; lung: A549, NCI-H1975, NCI-H1650; melanoma: A375, SK-Mel-23, SK-Mel-29, HT-168-M1/9, M24, UISO-MEL-2, UISO-MEL-6, UISO-MEL-23, B16/F10,
MDA-MB-435; brain: IMR-32, SK-N-BE2, U87, LN-229; bone: OS143B, U2OS; blood: CCRF-CEM, Jurkat, A2058; fibrous connective tissue: HT1080
Murine cancerous cell lines: melanoma: B16F1, B16-BL6; blood: L5178Y-ML25
Human noncancerous cell lines: fibrous connective tissue: MRC-5, BJ; skin tissue: HaCat; umbilical cord: HUVEC; embryonic lung tissue: HEL299; kidney tissue: HEK-293, HK-2; breast tissue: MCF10A;
Esophagus: Het-1A; placenta: CRL-3271
Murine noncancerous cell lines:	Fibrous connective tissue: NIH3T3, Bcl-2; embryonic stem cell: C57BL/6

4.2.1. CPPs with strong cell membrane activity

Primary amphipathic CPPs like TP10 and sC18 form a helix in the presence of membrane phospholipids, which leads to an intense attack against negatively charged membrane components. As a consequence, destabilization or disruption of the cellular membrane (forming, e.g. transient prepores or barrel or toroidal pores) takes place.^48,51 This membrane activity is responsible for the high cell-penetrating capacity of both peptides. On the other hand, the interaction between the CPPs in question and the membrane may be intense enough to cause a necrotic effect.^{18,80,97–99}

TP10 is a natural 21-residue chimeric CPP obtained by engineering mastoparan (a peptide derived from wasp venom) and the neuropeptide galanin linked through an extra lysine residue.¹⁰⁰ Unlike arginine-rich peptides, TP10 possesses four lysines which together with an N-terminus produce a charge of 5⁺ under neutral pH conditions. Many studies performed in vitro demonstrated an evident cytotoxic action of this peptide on versatile cancerous cell lines (HeLa, OS143B) at concentrations that were only negligibly harmful for nonmalignant cells.¹⁸

The second amphipathic peptide (sC18) is derived from the cationic antimicrobial peptide CAP18. To obtain an enhanced cellular uptake, its dimeric version was designed in which the second unit was introduced at the side chain of Lys residue, ensuring that both N-termini of sC18 sequences remain free.⁸¹ Owing to this procedure, which results in bringing the net positive charge to the value of 17, and the linear topology into the branched one (sC18)₂ acquires a high cell-penetrating capacity and a rather selective action on tumorous cells. Also, its biodistribution within them changes with preference for the nuclei, where the peptide accumulates.⁸⁰

Indeed, further research on (sC18)₂ confirmed its above-presented qualities, which are expressed not only by a broader spectrum of anticancer activity but also low toxicity. This was found on the basis of an experimental protocol that comprised 16 tumor cell lines obtained from distinct tissues such as, e.g. skin, lung, liver, breast, colon, prostate. Moreover, the values of IC50 for certain types of cancers were even more than 10-times lower (e.g. melanoma A2058, lung cancer H1650, liver cancer HepG2) in comparison to those calculated for healthy cells.^80,81

The importance of branching and arginine number for activity has also been described for other CPPs.^40,101

4.2.2. CPPs with pro-apoptotic activity

The most known CPPs within this group are p28-azurin derived,^33,34 buforin IIb and its derivative BR2.^85,86

The CPP which has gained considerable attention is p28, a 2.9-kDa fragment of bacterial protein – azurin (residues of 50–77). This CPP binds to the nuclear suppressor protein p53, inhibiting its proteasomal degradation, irrespective of whether it is of wild or mutated type. The main consequence of restoration of p53 function in cancerous cells is upregulation of p21 and p27 (cyclin-dependent kinase inhibitors), which leads to a decrease in CDK2-cyclin A complex (essential protein for mitosis) as well as FoxM1 (a transcription factor for G2/M progression). Thus, p28 is considered to be a cell cycle–specific cytostatic agent that induces apoptotic cell death via a p53-mediated block at the G2/M phase.^34,39,102

Latest findings report that this peptide also functions in a non-p53-mediated mechanism to inhibit tumor growth. Namely, thanks to the ability to enter endothelial cells, it also diminishes tumor neoangiogenesis by inhibition of the activity of VEGFR-2 and FGFR-1 kinases.⁸²

The promising preclinical anticancer activity and pharmacokinetics of p28 together with lack of immunogenicity and toxicity⁸³ were an incentive for entering two phase I clinical trial studies designated NCT00914914 and NCT01975116. The former one comprised adults (at age of ≥18) with refractory solid tumors, and the latter – children and adults (at age of 3–21) with versatile brain malignancies. Both studies indicated a good tolerance and safety profile of p28, even in children who were treated with the adult suggested phase II dose. This favorable outcome is not surprising since p28 has been characterized as a tumor-homing peptide that preferentially enters tumor cells, while sparing normal ones.³³ Although in pediatric tumors, this CPP was not effective enough as a single agent, a significant improvement in the survival of adults with versatile resistant solid tumors was encouraging researchers to perform combinatorial studies in which additive cell kill with chemotherapeutics could be exploited.^103,104

Other examples of CPPs include synthetic peptides like buforin IIb and its derivative BR2. The first one indicates strong anticancer activity against a broad spectrum of cancerous cell strains, which implies that the molecular target(s) for buforin IIb is/are universally present in these cells. After being internalized via endocytosis (preceded by electrostatic interaction with cell surface gangliosides and sialic acid), this peptide is primarily accumulated in the nucleus. The cell killing effect is due to the induction of apoptosis by a mitochondria-dependent pathway.⁸⁵

Because of the α-helical structure at buforin IIb C-terminus, this peptide is also cytotoxic to normal cells, although at high concentrations. This cytotoxic effect was minimized by a procedure, which involved α-helicity reduction in the molecule of buforin IIb. As a result, another CPP, i.e. BR2 (17 amino acids) was obtained with the maintained strong anticancer activity of the parent peptide, but without toxicity to normal untransformed cells.^86,105 These features of BR2 have been confirmed in experimental protocols in which this peptide was fused with a single chain variable fragment (scFv) of the antibody against mutated KRAS⁸⁶ or with a recombinant fusion protein with truncated diphtheria toxin (DT386)¹⁰⁶ or was incorporated onto the surface of a liposome loaded with the anticancer agent cantharidin.¹⁰⁷ Importantly, BR2 displayed greater membrane translocation membrane efficacy and induced a higher degree of apoptosis in KRAS mutated cells than Tat.^86,108 BR2 as a cancer-homing peptide has a great potential to become a useful delivery system for versatile anticancer drugs.

4.2.3. CPPs with pro-apoptotic and strong membrane activities

This group of CPPs includes lactoferricins (Lfcins) derived from bovine (LfcinB) and human (LfcinH) lactoferrins. They are amphipathic, highly positively charged but differ with respect to the lengths of the peptide and the amino acid sequence. Although both possess anticancer activity, the research focusing on LfcinB and its derivatives has been more fruitful, and therefore a short outline with the encouraging results is presented below.

LfcinB is a multifunction short cyclic cationic peptide (25 residues) derived from whey protein lactoferrin by its acid-pepsin hydrolysis. Its net charge is 8⁺ due to the presence of five arginines and three lysines in the molecule. LfcinB has been reported to be cytotoxic against human cell lines obtained from versatile cancers (breast, colon, stomach, melanoma, fibrosarcoma).^87,88 Furthermore, the peptide inhibits liver and lung metastases of both murine melanomas and lymphomas.^87,109,110

The cytotoxic activity of this peptide has been attributed to the core sequence 20-RRWQWR-25 (LfcinB_20-25), which is an amphipathic and positively charged (3⁺) motif.¹¹¹ Additionally, an increase in polyvalence resulting in enhanced hydrophobicity of the molecule improves much the selective toxicity against tumor cells, and this becomes a hallmark of the tetrameric peptide LfcinB_(20–25)4.^89–91

Tetramerization of the minimal LfcinB motif results in the net positive charge of 12⁺ and a dendrimeric structure, which has been shown to maximize the efficacy of the peptide.⁸⁹ The improved selective anticancer effect of LfcinB_(20–25)4 has become visible on several cancer cell lines as MCF7 (breast cancer), SCC15, CAL27 (oral cavity squamous-cell carcinomas – OSCCs) with 80%, 93% and 96% inhibition of cell viability, respectively.

The high antitumor activity of the peptide has been confirmed on the OSCC hamster experimental model, which mimics the tumorous pathogenic processes involved in human OSCC. The cytotoxic effect against tumor cells was executed by two different types of cellular death: apoptosis (activation of caspases or cathepsin B) and necrosis (membrane lytic effect). Specifically, apoptosis was the main mechanism of cell death at lower concentrations of LfcinB_(20–25)4 while necrosis was found only at higher ones.^89,90,112

It is worth stressing that the peptide, due to the intrinsical bioactivity and solubility, meets the requirements for intratumoral administration, which as a local route produces a superior antitumor effect with less severe adverse reactions in comparison to that induced by systemic application.

The favorable features of LfcinB led to a vast number of studies on the structure–activity relationship of Lfcin-derived⁹¹ and synthetic model peptides, which resulted in designing and synthesizing an amphipathic helical nanopeptide LTX-315 whose residues have been optimized for antitumor activity.^92,110,113 Its cytotoxicity and the underlying mechanisms of action have been explored on several cancer cell lines as well as on tumor animal models that feature key aspects of human disease. As these studies indicated, LTX-315 possessed anticancer activity against a broad spectrum of malignancies, including melanoma, fibrosarcoma, mesenchymal subcutaneous sarcoma, breast cancer, etc.^93,94,114 The results obtained after a direct administration of this peptide into the tumor (e.g. mouse/rat established melanoma, sarcoma, triple-negative breast cancer) are particularly encouraging because of the complete tumor regression (fibrosarcoma in rats, melanoma in mice) or growth inhibition with a strong abscopal effect (tumor regression in nontreated distal lesions).^92,94,95 It is worth emphasizing that in humans, the above-mentioned tumors do not benefit much from the available drug therapy.

As far as the mechanisms involved in the anticancer activity of LTX315 are concerned, at least dual sequential phases are triggered by the peptide, i.e. a rapid direct cytotoxic effect (cell lysis of the vasculature) and a long-lasting one characterized by alterations of the tumor environment that involves infiltration with CD8⁺ T cells that display antitumor functions. The latter phase is particularly important in all these types of cancers that are poorly infiltrated with T cells as happens in melanoma, and therefore their conversion into highly immunogenic ones (transition from a “cold” to a “hot” tumor environment) provides a long-term immunity, preventing tumor regrowth after treatment.^{92,96,110,115}

Generally, LTX-315 is a locally acting oncolytic agent of high efficacy and proper safety profile. The urgent need for new antitumor therapies and the promising pharmacological features of the compound in question have caused it to enter different settings of phase I and II clinical trials [NCT03725605, NCT01986426, NCT01058616, NCT01223209]. Those which involve combinations with the approved anticancer drug are described in Section 6.1.1.

4.2.4. CPPs with other activities

This Subsection focuses on synthetic CPPs composed entirely of arginine amino acids (R)_n as, e.g. R8, R9, R12, R16. The minimal sequence necessary for cellular uptake is six arginines, and the peptide penetrating capacity of the plasma membrane increases with their number in the molecule.^116,117 Much research has been devoted to the oligoarginine mechanisms of cellular entry, considering of which may be worthful in the context of anticancer activity of these peptides.

It is commonly accepted that internalization of oligoarginine peptides involves diverse types of endocytotic processes with the dominant role of macropinocytosis accompanied by actin cytoskeleton rearrangement.^{116,118–120} However, there is also ample evidence for the importance of direct passing through the cellular membrane.^55,119,121

Much attention has been attracted to R8, which is even considered to have the potential of a prototypic tumor-targeting vector due to selective accumulation in malignant cells with endocytosis playing a key role. This phenomenon was observed in experiments performed on tumor xenografted mice.¹²²

Considering the mechanism of anticancer activity of arginine-rich CPPs, the research concerning their usage with cargo or toxicity hinted at the involvement of cellular DNA and RNA. Putting it more precisely, the peptides indicate not only high affinity to nucleic acids but are also able to cause widespread defects in their metabolism.^116,121,122

5. Conjugation of CPPs with a cargo – two different strategies

The association of the CPPs with their cargo may be divided into covalently or noncovalently bound. Both formulation approaches have pros and cons. The advantage of the covalent one (chemical cross-linking between the CPP and cargo) is obtaining a final product with a well-defined chemical structure and reproducibility of the procedure.^123,124 Of great importance is the cargo coupling position because it may impact the nascent conjugate with respect to its chemical stability (e.g. fragmentation), pharmacokinetics (e.g. uptake pathway) and/or pharmacodynamics (e.g. activity, cytotoxicity).^125–127 Moreover, the risk of unfavorable alteration of the biologic activity of the cargo must be considered.^128,129

On the other hand, noncovalent interaction, i.e. physical complexation involves a simple bulk-mixing procedure of the compounds (CPP and cargo with or without a linker). This strategy, in comparison to the above-described one, is more simple and much easier to perform. Furthermore, it enables the usage of versatile cargos (with the preservation of their functionality) and low concentrations for induction of the biological response. However, the difficulty to control the final ratio and orientation of the peptide and cargo remains an important limitation of this method.¹²⁹

6. The anticancer activity improvement of the standard chemotherapeutics by CPPs

Classical cytotoxic anticancer drugs are known for targeting nonspecifically any rapidly dividing cells (irrespective of whether they are tumorous or not) due to their small size and lipophilicity. Also, loss of efficacy (caused by acquired drug resistance) and their inadequate pharmacokinetic characteristic with limited biodistribution often contribute to a poor final outcome of cancer management.

Recently, it has been found that the pharmacological properties of the chemotherapeutics may be improved by conjugating them with the CPPs, and this results in an increased efficacy of the transported drugs. Thanks to the presence of the CPP in the conjugate, the drug internalization is promoted with the consequence of high concentrations within the tumorous cells. What is more, certain areas in the body (e.g. the brain), which are frequently inaccessible for the therapeutics, become the target.⁷⁷

It is noteworthy that the improved pharmacokinetics is not the only factor that impacts the activity profile of the conjugate. There is also evidence for possible pharmacodynamic interactions between its constituents since certain CPPs possess anticancer action. An additional beneficial feature of coupling a chemotherapeutic with a CPP is overcoming MDR, which frequently occurs after repeated exposure of tumor cells to the same drug.

Exemplary conjugates of anticancer drugs with CPPs and the experimental models on which they were investigated are presented in Table 2.

Table 2. Representatives of conjugates of anticancer drugs with CPPs investigated on versatile experimental models

Anticancer drug	CPP	Conjugation strategy	Experimental model		References
			In vitro (cell lines)	In vivo (animal species)
Doxorubicin	p28	Non-covalent	LNCaP (p53^wt, AR^+), DU145 (p53^mut, AR^−), and PC-3 (p53^null, AR^−), IMR-32 (p53^wt), SK-N-BE2 (p53^mut), ZR-75 (p53^wt), MDA-MB-231 (p53^mut), U87 (p53^wt), LN229 (p53^mut), SK-Mel-29 (p53^wt), SK-Mel-23 (p53^mut)	Breast and melanoma cancer xenograft athymic mice	^Citation84
	R8	Covalent	MCF-7, HT-29		^Citation130
			HeLa, CHO, pgsA-745	Ovary cancer xenograft BALB/c nu/nu mice	^Citation122
	R9	Covalent	MCF-7, HT-1080		^Citation131
	Poly-L-arginine	Non-covalent	DU145		^Citation132
	[W(RW)4]	Covalent	SK-OV-3, CCRF-CEM, HCT-116, MDA-MB-468		^Citation133
		Covalent	KB-3-1, KB-V1		^Citation134
			CT-26	Colon adrenocarcinoma xenograft BALB mice	^Citation135
			MDA-MB-231, CHO, EMT6		^Citation136
			MDA-MB-231		^Citation137
			CHO, HUVEC, NG108.15, MCF-7, MDA-MB231		^Citation138
			HepG2		^Citation139
			MCF-7, MCF-7/ADR, AT3B-1		^Citation140
			U87, bEnd.3, glial cells	Nude mice (nu/J)	^Citation141
		Non-covalent	A549, MDA-MB-231, 4T1, RAW 264.7	Breast cancer xenograft BALB/c mice	^Citation142
		Pop-up	MCF-7		^Citation143
			A2780, A2780/AD, MCF-7, HL-60, HL-60/MX2, NCI-H69/AR, A549, KB	Breast cancer xenograft BALB/c nu/nu mice	^Citation144
	Penetratin	Covalent	MDA-MB-231, CHO, EMT6		^Citation136
			MDA-MB-231		^Citation137
			CHO, HUVEC, NG108.15, MCF-7, MDA-MB231		^Citation138
				Sprague Dawley rats, nude mice	^Citation145
			CHO, H9c2		^Citation146
	Maurocalcine	Covalent	MDA-MB-231, CHO, EMT6		^Citation136
			MDA-MB-231		^Citation137
			U87		^Citation147
	Angiopep2	Covalent	U87MG, SK-Hep-1, NCI-H460	CF-1 (Mdr-1a (-/- or +/+) and CD-1 (wild-type mice)	^Citation148
		Non-covalent	U87-MG		^Citation149
	KRP	Covalent	HUVEC, MCF-7, MDA-MB-435, MCF-10A		^Citation150
			HUVEC	Xenograft mice	^Citation151
	CADY-1	Non-covalent	HeLa, MCF-7	Lymphocytic leukemia xenograft DBA/2 mice	^Citation152
	SynB1	Covalent		Sprague Dawley rats, nude mice	^Citation145,Citation153
	SynB3	Covalent		Sprague Dawley rats, nude mice	^Citation145,Citation153
	QLPVM	Covalent	U87, bEnd.3, glial cells	Nude mice	^Citation141
Paclitaxel	p28	Non-covalent	LNCaP (p53^wt, AR^+), DU145 (p53^mut, AR^−), and PC-3 (p53^null, AR^−), IMR-32 (p53^wt), SK-N-BE2 (p53^mut), ZR-75 (p53^wt), MDA-MB-231 (p53^mut), U87 (p53^wt), LN229 (p53^mut), SK-Mel-29 (p53^wt), SK-Mel-23 (p53^mut)	Breast and melanoma cancer xenograft athymic mice	^Citation84
	R8	Covalent	C6, B16, HeLa, bEnd.3	Glioma xenograft BALB/c mice	^Citation154
			OVCA-429, OVCA-429 T	Transgenic mice ubiquitously expressing firefly luciferase	^Citation155
	Transportan	Non-covalent	A375		^Citation156
	TAT	Non-covalent	A375		^Citation156
		Covalent	A549, A549T	Non-small cell lung cancer xenograft BALB/c nude mice	^Citation157
	Penetratin	Covalent	A549, MCF-7		^Citation158
		Non-covalent	A375		^Citation156
	AngioPep2	Covalent		Human solid tumors with brain metastases	^Citation159
			NCIH460, U87MG	CD‐1 mice, CF‐1 mice (mdr1a (+/+) and (−/−)), athymic nude mice (Crl:Nu/NunuBR), glioblastoma and lung carcinoma xenograft CD-1 nude mice	^Citation160
				Human with reccurent malignant glioma	^Citation161
			U87-MG, SK-Hep-1, NCI-H460	CF-1 (Mdr-1a (-/- or +/+) and CD-1 (wild-type mice)	^Citation148
	K237	Pop-up	HUVEC	Breast cancer xenograft BALB/c nude mice	^Citation162
	LWMP	Covalent	A549, A549T	Non-small cell lung cancer xenograft BALB/c nude mice	^Citation157
	8V1, 6V1	Covalent	FaDu		^Citation163
			A549, Jurkat cell		^Citation164
Docetaxel	p28	Non-covalent	LNCaP (p53^wt, AR^+), DU145 (p53^mut, AR^−), and PC-3 (p53^null, AR^−), IMR-32 (p53^wt), SK-N-BE2 (p53^mut), ZR-75 (p53^wt), MDA-MB-231 (p53^mut), U87 (p53^wt), LN229 (p53^mut), Mel-29 (p53^wt), Mel-23 (p53^mut)	Breast and melanoma cancer xenograft athymic mice	^Citation84
	R9	Non-covalent	KB, HT-1080, MCF-7, A549	Cervical xenograft BALB/c mice	^Citation165
Metotrexate	YTA4	Covalent	MDA-MB-231, EA.hy 926	Breast cancer carcinoma BALB/c nu/nu mice	^Citation166
	YTA2, YTA4	Covalent	MDA-MB-231, MCF-7		^Citation167
	Penetratin	Covalent	MDA-MB-231, MCF-7		^Citation168
	R8	Covalent	MDA-MB-231, MCF-7		^Citation168
Gemcitabine	Penetratin	Covalent	MKN-28, Caco-2, HT-29		^Citation169,Citation170
	pVEC	Covalent	MKN-28, Caco-2, HT-29		^Citation169,Citation170
Asparaginase	LMWP	Covalent		Lymphocytic leukemia xenograft DBA/2 mice	^Citation171
	TAT	Covalent	MOLT-4		^Citation172
Chlorambucil	pVEC	C	MCF-7	Breast cancer carcinoma BALB/c nu/nu mice	^Citation173
	(sc18)	Covalent	HeLa, MCF-7, TDA		^Citation174
Irinotecan	DPV1047	Covalent		Human with solid refractory tumors	^Citation175
Etoposide	AngioPep 2	Covalent	U87MG, SK-Hep-1, NCI-H460	CF-1 (Mdr-1a (-/- or +/+) and CD-1 (wild-type mice)	^Citation148
Camptothecin	TAT	Covalent	HeLa, MDA-MB-231		^Citation176
Camptothecin	TAT	Non-covalent	C6	Glioma xenograft Sprague Dawley rats	^Citation177
Daunorubicin	(sC18)	Covalent	U87, MCF-7, HT-29		^Citation178
Actinomycin D	(sC18)2	Covalent	MCF-7, HEK-293, PC-3, HeLa, HCT-15, A2058, HT-168-M1/9, M24, HepG2, DU145, Panc-1, MDA-MB-231, A549, H1975, H1650, HT1080, PE/CA-PJ15, PE/CA-PJ41, HT-29, MRC-5		^Citation80
Bleomycin	R8	Non-covalent	HeLa, 4T1	Nammary carcinoma xenograft BALB/c mice	^Citation179
Cisplatin	TP10	Non-covalent	HeLa, HEL299, HEK293, OS143B		^Citation18
Dacarbazin	p28	Non-covalent	LNCaP (p53^wt, AR^+), DU145 (p53^mut, AR^−), and PC-3 (p53^null, AR^−), IMR-32 (p53^wt), SK-N-BE2 (p53^mut), ZR-75 (p53^wt), MDA-MB-231 (p53^mut), U87 (p53^wt), LN229 (p53^mut), SK-Mel-29 (p53^wt), SK-Mel-23 (p53^mut)	Breast and melanoma cancer xenograft athymic mice	^Citation84
Temozolomide	p28	Non-covalent	LNCaP (p53^wt, AR^+), DU145 (p53^mut, AR^−), and PC-3 (p53^null, AR^−), IMR-32 (p53^wt), SK-N-BE2 (p53^mut), ZR-75 (p53^wt), MDA-MB-231 (p53^mut), U87 (p53^wt), LN229 (p53^mut), SK-Mel-29 (p53^wt), SK-Mel-23 (p53^mut)	Breast and melanoma cancer xenograft athymic mice	^Citation84
Erlotinib	TAT	Covalent	U87, bEnd.3, glial cells	Nude mice	^Citation141
Legend:
Human cancerous cell lines: head and neck: FaDu, PE/CA‐PJ15, PE/CA-PJ41; liver: HepG2, SK-Hep-1, MKN-28; pancreatic: Panc-1; colon: HT-29, HCT-116, Caco-2, HCT-15; breast: MDA-MB-231, MDA-MB-468, MCF-7, ZR-75; ovarian: SK-OV-3, A2780, OVCA429; cervix: HeLa, KB, KB-3-1, KB-V1; prostate: LNCaP, DU145, PC-3; lung: A549, NCI-H460, NCI-H69/AR, NCI-H1975, NCI-H1650; melanoma: A375, SK-Mel-23, SK-Mel-29, HT-168-M1/9, M24, MDA-MB-435; brain: IMR-32, SK-N-BE2, U87, U87MG, LN-229; bone: OS143B; blood: CCRF-CEM, HL-60, HL-60/MX2, Jurkat, MOLT-4, A2058; fibrous connective tissue: HT1080
Murine cancerous cell lines: colon: CT26; breast: EMT6, 4T1; melanoma: B16; blood: RAW264.7;
Rat cancerous cell lines: prostate: AT3B-1, brain: C6
Human noncancerous cell lines: fibrous connective tissue: MRC-5; umbilical cord: HUVEC; embryonic lung tissue: HEL299; embryonic kidney tissue: HEK-293; breast tissue: MCF10A
Murine noncancerous cell lines: brain tissue: bEnd.3
Rat noncancerous cell lines: embroynic mycardium tissue: H9c2
Hamster noncancerous cell lines: ovarian tissue: CHO, pgsA-745
Human hybrid (HUVEC, A549): EA.hy926
Hybrid of mouse and rat brain cancer cell line: NG108-15

6.1. Combinations of a chemotherapeutic with a CPP possessing additional anticancer activity

At first, attention will be focused on such combinations whose CPP constituent itself succeeded with the resultant phase I clinical trials (LTX-315 and p28). Apart from these, research on this topic comprises preclinical studies, both in vitro and in vivo. Next, other conjugates with a CPP possessing anticancer activity (R8, (sC18)₂, TP10) will be shortly considered.

6.1.1. Anticancer drugs+LTX-315 and conjugates of cytotoxic chemotherapeutics with p28

Although this subsection concentrates on the CPPs conjugated with the chemotherapeutics, an exception has been made for LTX-315 peptide, which is administered not in the form of a conjugate with an anticancer drug, but separately (intratumorally) with one of the two distinct MABs (iv). As has been mentioned before, monotherapy with this CPP disintegrates cytoplasmic organelles and induces immunogenic cell death. Because of these properties, LTX-315 entered phase I clinical trials, which aimed at demonstrating the safety of the CPP itself and the potential to enhance the systemic anticancer activity of such compounds as ipilimumab and pembrolizumab.^115,180 The results of the trials confirmed its antitumoral efficacy, also when used in combination with each of the aforementioned drugs. The response to either ipilimumab or pembrolizumab (both are immune checkpoint inhibitors, ICIs) in the presence of LTX-315 (intratumoral injection) was increased, leading to an improved prognosis in patients with melanoma or triple-negative breast cancer (NCT01986426).¹⁸⁰

Moreover, there are promising results of preclinical studies with LTX-315 (intratumoral) and doxorubicin (DOX; iv) obtained on a triple-negative breast cancer model in mice. The usage of both compounds in combination resulted in an additive anticancer effect manifested by a complete regression of the tumor in the majority of treated animals.¹¹⁴ This additive effect together with that established for LTX-315 and ICIs has been interpreted in view of the changes in the immune system of the tumor environment induced by each constituent of the therapy. Such interaction of the peptide and ICIs could have been predicted, but rather not in the case of DOX. However, recent reports demonstrated that DOX, apart from its intercalating action, has the potential to stimulate anticancer immune responses by a selective elimination of the immunosuppressive cells and is able to induce immunogenic cell death.^114,181,182

Research on the combinations of p28 with anticancer drugs includes preclinical studies with non-covalent conjugates. This peptide after mixing with a representative of DNA-damaging (i.e. doxorubicin, dacarbazine, temozolomide) or antimitotic (i.e. paclitaxel, docetaxel) drugs was tested on a variety of cancer cells expressing wild or mutated types of p53. The experiments were performed in vitro (cell lines of myriad cancer types) and in vivo (mice of both sexes). It is worth noticing that p53 protein is inactivated or malfunctioning in about 50% of human cancers, and therefore strategies have been undertaken to restore its guardian function of the cell cycle. One of these strategies includes an experimental protocol with the above-presented p28 mixtures. Their components produced an additive effect expressed by improved efficacy and lower toxicity, as compared to those observed after each single chemotherapeutic. The potential therapeutic value of the mixtures has been further analyzed in light of the possible mechanisms.⁸⁴

It has been demonstrated that p28 and the tested anticancer drugs produce an increase in p53, but by utilizing distinct pathways. The CPP by attaching DNA-binding domain of this protein, blocks the E3 ligase Cop1 and thereby induces a post-translational increase in the level and activity of p53.^183,184 By contrast, DNA damaging drugs produce this effect by activation of the ATM (Ataxia Telangiectasia Mutated protein) pathway in response to the double-strand breaks in the DNA, whereas the antimitotic agents do this via post-transcriptional modifications within the p53 protein itself. Reestablishing p53 in cancer cells, irrespective of the mechanism involved, brings about cell cycle arrest, DNA repair or apoptosis. However, the recovery of p53 function induced by chemotherapeutics is not always effective enough since the drugs secondarily trigger checkpoint signaling and initiate DNA repair pathways, including the p53-independent ones.⁸⁴ Their abrogation is a proper move to the improvement of the anticancer activity with a simultaneous drop in the risk of resistance. It has been found that such a phenomenon occurs if p28 is added to the chemotherapeutic. As has been stressed this CPP is a non-genotoxic specific cell cycle inhibitor at G2-M phase, which activates the p53/p21 axis significantly reducing CDK2 levels.^184,185

6.1.2. Conjugates of cytotoxic chemotherapeutics with other CPPs possessing anticancer activity

Versatile anticancer chemotherapeutics have been conjugated with oligoarginine peptides. The conjugates mostly explored are those whose CPP component is pure R8. This CPP has been coupled with such drugs as, e.g. DOX,¹²² paclitaxel,¹⁵⁵ bleomycin,¹⁷⁹ methotrexate¹⁶⁸ and other.¹⁸⁶ The majority of the formulations were reported to be effective in vitro as well as in vivo (tumor xenograft bearing mice), and some of them showed several other merits in comparison to the unconjugated drugs. The most important ones comprise the potential to overcome MDR and the improvement in physicochemical as well as pharmacokinetic qualities, as has been observed in the case of a paclitaxel covalent conjugate with R8. As has been demonstrated, this formulation presented higher efficacy over that of paclitaxel alone, but only against paclitaxel-resistant tumors (overexpressed P-gp efflux pumps). This quality of the conjugate became visible in vitro (lower IC50 values) on a panel of resistant ovarian and breast cancer cell lines (OVCA429T, OVCA433T, MCF7-Pgp) and in vivo (prolongation of the survival rate in mice) on an ovarian cancer mouse model.¹⁵⁵

The resistance to many chemotherapeutics, including that of paclitaxel, develops because their hydrophobic nature makes them appropriate substrates for P-gp efflux pumps.¹⁸⁷ Attachment of the drug to R8 decreases hydrophobicity in favor of hydrophilicity and thereby the conjugate is able to avoid P-gp-based resistance. Furthermore, its high aqueous solubility facilitates the administration and provides more constant concentrations (minimizing peak-trough effects) due to sustained release of the free drug from the conjugate.¹⁵⁵

The potential of R8 to improve anticancer activity was also assessed after its conjugation with DOX as a model compound. Despite its hydrophilic nature, DOX in the neutral form has the capacity to enter the intracellular compartment by itself and therefore, does not require the assistance of a CPP for this purpose. Theoretically, the response to the conjugate in DOX-sensitive cancers should be comparable to that after the free drug administration. However, the experimental data do not always confirm this assumption.

There are reports indicating an improvement of the conjugated-DOX bioactivity, but this effect was usually limited to a particular concentration/dose and/or the type of cancer cell line.^122,130 For example, R8-DOX conjugate at the concentration of 10 μM showed higher cytotoxicity against HeLa cells in comparison to that produced by the free drug, even at the concentration of 20 μM.¹²² Interestingly, the increased cytotoxicity was a characteristic feature of the DOX covalent formulation with R8, and not of the non-covalent one. This experimental outcome is compatible with the idea that due to covalent conjugation a new entity is established which often presents more favorable qualities than those of the free drug. What is more, the above-mentioned covalent conjugate demonstrated anticancer activity also in tumor xenografted mice. Both R8-DOX and the free drug induced the same percentage decrease in the tumor size, although the conjugate did so at a lower dose and without loss of mouse weight. The improved activity of the conjugate has been assigned to the R8-induced increase in the affinity of DOX to its targets within the cell.¹²²

Drug delivery potential together with a lytic effect on cell membranes is characteristic of (sC18)₂ and TP10. Each CPP has been utilized in the form of the conjugate with standard anticancer chemotherapeutics, i.e. (sC18)₂ as a covalent formulation with chlorambucil⁸¹ or actinomycin D,⁸⁰ and TP10¹⁸ as a mixture with cisplatin. The data on their action are scarce and comes from in vitro studies on diverse tumor cell lines.

For example, the conjugated form of chlorambucil or actinomycin D with (sC18)₂ demonstrated evident cytotoxicity on the breast cancer (MCF7) cell line. Interestingly, the unconjugated drugs were not cytotoxic at all, and this lack of effect has been attributed to their poor cellular internalization. Undoubtedly, the transport of chlorambucil and actinomycin D into the cancer cells has been evidently improved in the presence of (sC18)₂, and thereby the drugs gained access to the intracellular targets, e.g. DNA, which becomes alkylated (chlorambucil) or intercalated (actinomycin D). Whether the increase in cytotoxicity of the conjugates results exclusively from a much better entrance of the drug into the cancer cells still remains an open issue; obviously, the membrane lytic effect of the CPP component cannot be excluded.^80,81

Similarly to the covalently conjugated drugs, the mixture of TP10 and cisplatin indicated a strong cytotoxic effect on two tumor cell lines as HeLa – cervical cancer and OS143B – osteosarcoma. Further analysis of this effect demonstrated that the magnitude of the response equals at least the sum of that produced by the CPP and the drug alone. This additive action could have been expected since TP10 and cisplatin have a distinct anticancer mechanism of action, i.e. lytic effect on the cell membrane and cytotoxicity induced by alkylation of DNA, respectively.¹⁸

6.2. Other combinations of a chemotherapeutic with a CPP

At present, formulations of anticancer chemotherapeutics with CPPs that exploit passive targeting will be considered. The chosen CPPs are cationic and fall into natural (Tat, penetratin, maurocalcine (MCa), DPV1047, SynB family, Angiopep-2, LMWP, pVec) and synthetic (CADY, KRP) categories. They have been conjugated with representatives of anticancer drugs indicating diverse mechanisms of action. Their description starts with conjugates whose preclinical outcome made them eligible for clinical studies, i.e. SN38 (the active form of irinotecan)/DPV1047 and paclitaxel/Angiopep-2.

6.2.1. SN38/DPV1047 and paclitaxel/Angiopep-2 conjugates

DPV1047 is a highly charged 19 amino acid oligopeptide that partially originates from an anti-DNA antibody and is internalized via clathrin-and caveolae-independent endocytosis.¹⁸⁸ The peptide belongs to the collection of human-derived peptides denominated Vectocell. Thanks to the antibody sequence, DPV1047 can enter the cell and be transported to the nucleus.

DPV1047 has been covalently bound to SN38 via a specific esterase-sensitive cross-linker, and the product is known as DTS-108. Its significant clinical advantage is derived from the circumvention of the limitations of irinotecan which is a prodrug of SN38. Irinotecan is a topoisomerase I inhibitor, and the drug itself has a very weak intrinsic antitumor activity since its biotransformation into the active form (SN38) in humans is variable with the final yield of only 2–8% of the administered drug dose.¹⁸⁹ Also, due to the poor solubility and toxicity profile (high incidence of life-threatening diarrhea, bone marrow suppression), SN38 is unsuitable as a therapeutic.¹⁹⁰ On the other hand, its conjugation with a CPP renders the compound soluble for iv administration in purely aqueous vehicles and makes it not reliant on hepatic activation and metabolism. Instead, the conjugate linker is cleaved by the blood esterases with the resultant release of a high pool of active SN38.¹⁹¹

The clinical trials with DTS-108 were preceded by preclinical studies estimating its efficacy and toxicokinetics on versatile models (three colon, a lung and a mammary adenocarcinomas) and species (dog, mouse, rat).¹⁹¹ The conjugate was shown to be cytotoxic in the above-mentioned panel of cancer cell lines with IC50 values even 1000 times lower than those calculated for irinotecan. For example, IC50 of DTS-108 for LS174T colon cancer cell line was 2 nmol/L and that of irinotecan 35 μmol/L. An analysis of the amount of SN38, as a result of the conversion of the prodrugs, was performed on the basis of the experimental data obtained from the administration of DTS-108 (three different doses including maximum tolerated dose-MTD) and irinotecan (MTD) in dogs (the dog is a relevant model that mirrors the human pharmacokinetics of irinotecan). The conjugate, irrespective of the dose, always generated much bigger amounts of the active metabolite, as compared to those induced by irinotecan. It is worth stressing that the two smaller doses of DTS-108 did not produce any intestinal toxicity since a negligible amount of SN38 is present in the intestine due to the conjugate non-hepatic mode of activation and weaker liver metabolism in comparison to those of irinotecan. The modified pharmacokinetics of DTS-108 improved antitumoral efficacy, which was indicated on the nude mice or rats bearing versatile human cancer xenografts. Moreover, the conjugate produced a synergistic or an additive effect if used together with 5-fluorouracil or bevacizumab, respectively.¹⁹¹

A phase I clinical trial comprised patients with advanced/metastatic carcinomas and its outcome has been summarized in a study delivered by Coriat et al.¹⁷⁵ Its purpose was to ascertain the dose-limiting toxicities (DLTs), MTD and the recommended Phase II dose (RP2D) in patients (male, female) with different tumor types (mostly of gastrointestinal origin) and multiple-site metastases who experienced relapse after prior chemotherapy regimen(s). DTS-108 at the RP2D (313 mg/m²) permitted four times higher SN38 concentrations during the tested time intervals than irinotecan at the equivalent dose. The objective antitumor activity was manifested by maintaining the status of the stable disease for six DTS-108 cycles or longer, and proceeded with only moderate digestive toxicity and without clinically significant hematologic toxicity.¹⁷⁵

Angiopep-2 is a 19-mer CPP derived from the family of Kunitz domain peptides. It is able to attach up to three molecules onto the peptide backbone. Its primary sequence and charge (2⁺) are crucial for the passage across the BBB. This peptide possesses an extraordinary potential for stimulating brain-targeted drug delivery through LRP-1-mediated transcytosis. LRP-1 is a member of the low-density lipoprotein receptor family, which is highly represented on the BBB and, more importantly, it is also over-expressed in human glioma cells.^149,192

The chemical properties of Angiopep-2 permitted its covalent conjugation with three molecules of paclitaxel, a broad spectrum antimitotic agent. The resultant product – Ang1005 – shows superiority over paclitaxel regarding the access to the brain. By contrast, the conjugate being a prodrug (Ang1005 is cleaved by esterases to release paclitaxel) is not a substrate of drug efflux transporters (P-gp and other) and guarantees therapeutic concentrations of paclitaxel within the brain tissue. Additionally, Regina et al.¹⁶⁰ found that the conjugate with paclitaxel inhibited growth of human glioblastoma (U87 MG) more potently than paclitaxel and significantly increased mouse survival rates.

The promising outcome of the preclinical research concerning Angiopep-2/paclitaxel (ANG1005)¹⁶⁰ has brought about many clinical trials and the most encouraging are presented below.^161,193

Kurzrock et al.¹⁹⁴ and Drappatz et al.¹⁶¹ summarized phase I clinical trial results (NCT00539383; NCT00539344) that covered the experimental material obtained after ANG1005 (also known under the name GNR1005) treatment of patients with recurrent glioma or brain metastasis due to breast and lung cancers. In both cases, the compound was well tolerated and effective enough to initiate subsequent phases of clinical trials. Phase II comprised the usage of the conjugate and a mAb. Bevacizumab was chosen for glioma (recurrent WHO grade III) and trastuzumab for brain metastatic changes of breast cancer (NCT01967810; NCT01480582; NCT01679743). Also, an open-label ANG1005 – phase III study has started in patients with HER2-negative breast cancer with newly diagnosed leptomeningeal carcinomatosis and previously treated brain metastases (NCT03613181). The recruitment has already been finished, and the end of the study is projected for December 2022.

6.2.2. Preclinical studies of anticancer drugs with CPPs

They include research on conjugates of DOX, L-asparaginase, gemcitabine, methotrexate with the CPPs brought up in Subsection 6.2. The majority of the conjugates are covalent formulations which have been tested both in vitro and in vivo. Considering the drug component, most records concern doxorubicin. This is not surprising since this chemotherapeutic has a wide antitumor spectrum (hematological malignancies, solid tumors), compared with other anticancer drugs. Therefore, the presentation of the conjugates will start with those which possess DOX in the construct.

6.2.2.1. Conjugates of DOX with a CPP

The purpose of the preclinical studies is to get over the drawbacks of DOX, which limit its clinical usage. The most significant among them are poor BBB penetration, chemoresistance and numerous serious adverse reactions, including dose-dependent cardiotoxicity. Thus, by conjugation of DOX with myriads of CPPs, an attempt has been made to improve its pharmacological profile, especially in the context of the presented limitations. It is noteworthy that the improvement of the chemotherapeutic cellular entrance is here of minor importance since DOX very easily transfers the cell membrane. Hence, the outline of the preclinical trials will be as follows.

6.2.2.1.1. Overriding DOX resistance by its conjugation with a CPP

Overriding DOX resistance has become visible when its covalent forms with Tat, penetratin and MCa were tested on several DOX-resistant cancer cell lines.^{136–138,146,147,194,195} At first, a brief characteristic of these CPPs follows.

The peptides are derived from human immunodeficiency virus type 1 (Tat), Antennapedia homeodomain (penetratin) and the Tunisian chactid scorpion toxin (MCa). Tat contains six arginines (therefore it is also classified into arginine-rich CPPs) and two lysine residues in the sequence that provides a total 8⁺ charge of the molecule. Penetratin contains three arginines and four lysisns (total charge 7⁺) in its molecule, and MCa is a 33-mer highly polar basic peptide containing 12 positively charged residues (seven Lys, four Arg and one Gly). Both latter peptides possess amphipathic properties.^40,196,197

The processes involved in Tat, penetratin and MCa internalizations comprise distinct types of direct translocation (inverted micelle formation, transient pore formation) and endocytosis (clathrin-mediated endocytosis, lipid raft-dependent macropinocytosis).^196–199

Aroui et al. tested DOX conjugates with the above presented CPPs on several cell lines with different sensitivity to DOX, i.e. human triple-negative breast adenocarcinoma – MDA-MB-231, Chinese hamster ovary – CHO, human umbilical venous endothelial cells – HUVEC, differentiated NG108.15 neuronal cells, MCF7. As could be expected, conjugation of DOX to CPPs increases the efficacy of DOX both in terms of effective concentrations and maximum effect only in resistant cell lines (MDA-MB-231, CHO, HUVEC). For instance, in the MDA-MB-231 cell line, the calculated values of EC50 with the maximum effects of DOX and the conjugates (determined at the concentration of 10 μM) are as follows: DOX – 2.7 μM/~70%, DOX-MCa – 0.32 μM/~90%, DOX-Tat – 0.25 μM/~81%, DOX-penetratin – 0.41 μM/~95%. The improvement in the activity of the conjugates has been attributed to a much higher level of accumulation of the conjugates in the resistant cell lines, as compared to free DOX. This effect does not result from an increased internalization (on the contrary, it may even decrease) but from evading the P-gp efflux pump which constitutes one of the principal mechanisms of resistance to DOX.^{136–138,146,147,194,195}

Also, the reports focus on the processes engaged in apoptotic MDA-MB-231 cell death due to treatment with DOX-CPPs conjugates. As compared to free DOX, apoptosis appears at concentrations 5 times lower and involves different signaling pathways. Apart from the BAX-dependent mitochondrial one (a characteristic feature of DOX), the conjugates trigger an additional non-mitochondrial apoptotic pathway (via TRIAL) which is intrinsically inactive in, e.g. MDA-MB 231 breast cancer cell line with TRIAL expression.^{136–138,194} Also, Zhang et al. has noticed overriding DOX-resistance by Tat-DOX conjugates, but in MDR-human cervical cancer cell line (KB-V1).¹³⁴

It is worth adding that DOX-sensitive cancer cell lines responded to CPP formulations only when a large amount of DOX in the construct was present (‘pop-up’ strategy with 15 wt% of DOX in multifunctional micelle with Tat).^139,144

6.2.2.1.2. Reduction of DOX toxicity by its conjugation with a CPP

It is obvious that any targeted therapy of DOX will lead to a reduction of its global toxicity. One of them was the introduction of the liposomal form of the drug, which improved the safety profile of the chemotherapeutic. Liposomal DOX, similarly to DOX-CPP constructs, accumulates in solid tumor tissues via an enhanced EPR effect. Furthermore, due to its encapsulation within the liposome, the drug acquires different pharmacokinetic qualities, i.e. clearance and volume of distribution decrease, and half-life extends up to 72 h. Thus, Li et al.¹⁵² performed a comparative analysis of pharmacological activities of three forms of DOX – the conjugated, the liposomal and the free one. The chosen CPP for conjugation with DOX was CADY-1, which is a synthetic, cationic peptide.

CADY-1 is a chimeric 24-residue secondary amphipathic peptide, which is internalized by direct translocation. Being a representative of self-assembled peptides, this CPP is capable of forming a stable complex with a drug by simple bulk mixing.^152,200

As has been indicated in mice, the conjugate and the liposomal form of DOX possess similar blood residence time and toxicity with LD50 ca. 30 mg/kg (LD50 of free DOX ca. 17 mg/kg). However, the complex possessed higher potency and efficacy than the liposomal form or the free drug. This has been indicated on the basis of in vitro (cell viability, cell-penetrating activity) and in vivo (HeLa tumor xenografted mice) experiments. The level of the conjugate internalization was 3 times higher than that determined for the liposomal form or the free drug. The tumor xenografted mice responded with a 5-d extension of survival to CADY1-DOX complex treatment, compared to that induced by liposomal DOX.¹⁵²

Taking into account the characteristic of CADY-1/DOX complex, it could be regarded as a replacement for the liposomal form of DOX in future treatment. It is worth stressing that the synthesis of the CADY-1 peptide is not complicated and easy to scale up with current technology.¹⁵²

Also, an improvement of the pharmacological profile has been obtained in the case of covalent DOX conjugation with another synthetic, cationic CPP, i.e. KRP. This peptide is a novel lysine-rich basic CPP whose internalization involves endocytosis. Linking KRP with the drug by two stable bonds (thioether and amide) blocks the release of free DOX from the conjugate in the systemic circulation and reduces the risk of its toxic effect on normal tissues. Thanks to the latter feature, the therapeutic index of DOX increases. Additionally, the presence of two nuclear localization sequences in the KRP molecule ensures DOX a specific targeting to its site of action reflected by an augmented tumor-killing effect.¹⁵⁰

6.2.2.2. CPP as a vector for DOX transport through the BBB

An essential aspect of the CPP role in cancer treatment concerns the ability to serve as a vector for drug transport through the BBB without compromising its integrity. A vast number of anticancer drugs including DOX has poor access to the brain tissue, even though their capacity to transfer the cell membrane is high. This is caused by a high rate of active removal from the brain tissue by the efflux pumps. Bypassing these mechanisms would enable the drug to attain appropriate concentrations indispensable for the brain tumor treatment. In this context, covalent conjugates of DOX with BBB penetrable CPPs have been thoroughly evaluated with respect to the healthy brain tissue (penetratin-, SynB family-DOX conjugates), and next to the tumor transformed one (Angiopep2-DOX conjugates).

Considering the CPP component of the conjugates, the one which requires a short description is SynB family, i.e. SynB1 and SynB3 peptides (a truncated derivative of SynB1). They are derived from the porcine antimicrobial peptide – protegrin 1 and possess six (SynB1) or five (SynB3) arginines in the molecule. Their prevalent uptake mechanism is endocytosis.¹⁵³

Rousselle et al.^145,153 deliver research on BBB penetration of DOX-penetratin and DOX-SynB1 conjugates performed on rats (in situ) and DOX-SynB1/SynB3 conjugates in mice (in situ, in vivo). The in situ rat and mouse brain perfusion methods indicated that vectorization of DOX brought about a 20- or 30-fold increase in the brain uptake of the conjugates, respectively. This increase was caused by the avoidance of the efflux activity of P-gp at the level of BBB. The type of transport involved in the conjugate internalization points to adsorptive-mediated endocytosis.^145,153 Moreover, the in vivo study showed an altered tissue distribution profile with much lower concentration of the CPP-modified DOX in the heart, which could potentially reduce the cardiotoxicity concerns.¹⁴⁵

The anticancer activity of DOX (3 molecules) conjugates with Angiopep-2 have been tested on glioma U87 cell line and in mice bearing U87 glioma tumors. The in vitro study indicated a comparable cytotoxicity of the conjugate to that produced by the parent drug. On the other hand, in the in vivo experiments, the conjugate was present in the tumor implanted hemisphere in the amount two times bigger than that of DOX. Also, there was a dramatically higher BBB influx rate of the conjugate in comparison to the free drug (mouse brain perfusion method) which was ascribed to the bypass of DOX-induced efflux mechanisms. This has been further verified on Mdr-1a (-/-) knockout mice, which demonstrated a much higher brain penetration of the conjugate in comparison to that of the parent drug, implicating also other efflux pumps that limit the entry of the unconjugated compound into the brain (e.g. BCRP, ABCG2).¹⁴⁸

To increase the therapeutic result of glioma therapy, an effective option is the usage of polytherapy. With this in mind, two anticancer therapeutics – DOX and erlotinib have been loaded into dual functionalized liposomes containing on their surface a CPP (Tat or QLPVM) and transferrin (active targeting due to expression of transferrin receptors on endothelial and glioblastoma cells). However, this construct, not being a conjugate, has been added because it constitutes a very innovative targeted delivery system for a CPP and more than one drug. The construct was tested on an in vitro brain tumor model as well as in vivo and ex vivo after iv administration in mice. The experiments revealed an efficient co-transfer of DOX and erlotinib across the BBB (co-culture endothelial barrier), which led to a significant increase in DOX and erlotinib content in mice brain, as compared to the free drugs. Simultaneously, no signs of organ toxicity such as necrosis, nuclei enlargement, inflammation or diffuse fibrosis of myocardium were observed. The brain accumulation of both drugs enabled them to act more efficaciously, resulting in tumor regression, which was reflected by a significant decrease in the percentage of tumor cell viability and a higher number of dead tumor cells.¹⁴¹

6.2.2.3. Conjugates of L-asparaginase with Tat or LMWP

As a short description of Tat has been presented previously, only that of LMWP is provided. It can be readily produced in mass quantities by enzymatic digestion of native protamine with the final result of a peptide fragment (MW: about 1.8 kDa) with 10 arginine residues. The peptide displays a significant similarity to Tat as far as the structure, transcellular localization behavior and kinetics are concerned. LMWP efficiently transfers the cargo into the nucleus and cytoplasm in a short time period.²⁰¹

Tat and LMWP have been covalently conjugated with L-asparaginase, which is the primary drug for the treatment of acute lymphoblastic leukemia (ALL). Each of the conjugates constitutes a component of a complex construct, i.e. the Tat conjugate of heparin/protamine regulated delivery system known as ATTEMPTS,¹⁷² and the LMWP conjugate of the encapsulated red blood cell (RBC) carrier system.¹⁷¹ These strategies aim at reducing the pitfalls of L-asparaginase treatment (immunological reactions, liver dysfunctions and other toxicities, protection from proteolytic degradation and metabolic clearance).

The primary significance of the ATTEMPTS method lies in its capability of allowing only the tumor cells to carry their self-destructing agent, i.e. the cellularly delivered asparaginase by Tat. By this means the cancer cells are deprived of asparagine, while its nearly normal concentration is maintained in the circulation. What is more, thanks to this approach, a higher efficacy (leukemia MOLT-4 cell line) and a prolonged mean survival time (leukemia bearing mice) have been achieved in comparison to those of the free drug.¹⁷²

As regards the LMWP-mediated RBC encapsulation of asparaginase, this method ensures on the one hand, retention of the complete structural and functional integrity of the erythrocytes, which may serve as a drug reservoir, and, on the other, protection of the loaded drug against the physiological environment. Thus, the pharmacokinetic and pharmacodynamic parameters of asparaginase have been changed with resultant prolongation of its plasma half-life (4.5 ± 0.5 d whereas that of RBCs loaded with asparaginase via a hypotonic method was 2.4 ± 0.7 d), safety of iv usage (erythrocytes help avoiding immune response toward the encapsulated drugs), increase in efficacy with the median survival time of the lymphoma tumor bearing mice by ca. 44% in comparison to the control group.¹⁷¹

6.2.2.4. Conjugates of cytotoxic antimetabolites with CPPs

The conjugates taken into account comprise gemcitabine+pVEC, gemcitabine+penetratin, and methotrexate+penetratin. pVEC, being a peptide of 18 amino acids, is derived from the cadherin murine vascular endothelium. It enters the cell by direct and endocytic mechanisms of internalization.^202,203

Gemcitabine is an antimetabolite of pyrimidine nucleosides, which is used in various types of solid tumors comprising first-line treatment of pancreatic, metastatic bladder or non-small cell lung cancers. One of the drawbacks is its poor pharmacokinetic profile including hydrophilicity, low oral bioavailability, transporter dependent cellular uptake (nucleoside transporter deficiency is related to different forms of drug resistance) and rapid inactivation by cytidine deaminase in plasma and liver. In order to diminish the process of inactivation, gemcitabine has been conjugated with pVEC or penetratin. Conjugation of the drug with the CPP masks its aniline moiety, which results in higher availability of gemcitabine for cellular internalization, facilitated additionally by the presence of the peptide. What is more, both gemcitabine conjugates acquired prolonged half-lives (9.6 d-gemcitabine/penetratin or 42 h-gemcitabine/pVEC) and enhanced antitumor activity, which was determined on three human cancer cell lines (KN-28-gastric cancer, Caco-2 – colorectal adenocarcinoma and HT-29 – colon cancer).¹⁶⁹

Methotrexate (MTX) is an effective anticancer drug which is widely used in several types of solid tumors as well as hematological malignancies. As a folic acid analogue, it inhibits dihydrofolate reductase and, by this means, the synthesis of tetrahydrofolate. The latter one is a key factor involved in the production of thymidylate and purine nucleotides, thereby interfering with the formation of DNA and RNA that finally results in cell death. Inside the cell, MTX undergoes polyglutamation by folylpolyglutamate synthetase (FPGS), and this process is crucial for the therapeutic action. Polyglutamylated forms of the drug retain within the cancer cells and determine the cytotoxic action of this chemotherapeutic. Furthermore, their decreased formation is one of the mechanisms responsible for MTX resistance.

The process of cellular polyglutamylation is not always efficient enough, and even absent in resistant tumor cells which leads to a dramatic drop in the therapeutic activity of MTX. To overcome this unfortunate impediment, Szabo et al.¹⁶⁸ undertook a strategy with a “ready-made” pentaglutamylated MTX coupled covalently to penetratin. The presence of the CPP was indispensable for the cellular entrance of this form of drug.

The conjugates were tested in vitro (cellular uptake and cytotoxicity profiles) on MCF7-sensitive or MDA-MB-231-resistant breast cancer cell lines. As could have been expected, the pentaglutamylated MTX–penetratin conjugate indicated activity exclusively against drug-resistant MDA-MB-231 cells, and the efficacy was higher than that of the free MTX on sensitive cells. The cytotoxicity on this resistant cell line, expressed by the IC50 value, was 0.1 μM (IC50 of MTX >100).¹⁶⁸

Generally, there is no direct correlation between the degree of internalization and activity. Despite the high efficacy of the pentaglutamylated MTX–penetratin conjugate, its uptake in both MCF7 and MDA-MB-231 cell lines was smaller compared with that of non-glutamylated form of the conjugate. The reason for this is an electrostatic interaction between the anionic (mainly glutamic acid residues) and cationic (arginine residues) domains within the construct. Introducing a spacer (tetrapeptide) between these domains attenuates the interaction, resulting in improved cellular internalization. It is intriguing that the pentaglutamylated MTX–penetratin conjugate equipped with such a spacer revealed cytotoxic activity also on MTX-sensitive cells (IC50 = 0.3 μM).¹⁶⁸

7. Conclusions

Concerning CPPs

1. As potential vectors for cellular delivery of small molecule drugs, they indicate a certain level of selectivity to cancer cells, called passive selectivity.

2. Besides drug transporting activity, certain peptides possess anticancer activity per se, accomplished mainly by membrane lytic effect, proapoptotic activity, or both actions.

3. Among the anticancer peptides, the most promising are p28 and LTX-315. Of great interest is particularly the latter one, the first locally acting oncolytic agent that reorganizes the tumor microenvironment. It entered several settings of phase I and II clinical trials.

Concerning combinations of anticancer drugs with CPPs

1. The research includes preclinical studies carried out both in vitro (versatile cancer cell lines) and in vivo (usually tumor xenograft bearing mice). They indicated an improved pharmacologic profile of the conjugates in comparison to the unconjugated drugs. It concerned mainly better cellular internalization (in case a drug showed a poor one) and circumvention of tissue barriers (e.g. access to the brain tissue), overcoming drug resistance and lowering toxicity. These qualities of the conjugates brought about superior anticancer activity over that of the anticancer drugs alone. Most records concern conjugates of DOX with a CPP.

2. A few preclinical studies were the basis for entering clinical trials. The first phase entered four entities: LTX 315+ ipilimumab, LTX-315+ pembrolizumab as well as two conjugates, i.e. DTS-108 and Ang1005. The trials with the latter one comprise also phase II and the unfinished phase III. So far, their outcome has been favorable.

Disclosure statement

No potential conflict of interest was reported by the author(s).