Conventional cancer treatments possess limitations for solid tumours, such as lack of selectivity, thus nanomedicine is explored as an efficient tool in anticancer drug development. Therefore, this study aimed to design a novel anticancer peptide (ACP) and encapsulate it into a lipid-based nanoparticle for efficient delivery to tumours. ACP was designed using in silico methods and thereafter encapsulated into a liposomal formulation (P1CF1) using a thin-film hydration method. The shape, size, polydispersity index (PDI), and zeta potential (ZP) of the formulation were evaluated using cryo-transmission electron microscopy, and zetasizer, while the percentage drug encapsulation efficiency (%EE) and drug release were investigated using the ultra-filtration and the dialysis methods. The biocompatibility, cytotoxicity, and apoptosis activity of P1CF1 were evaluated using hemolysis, 3-[(4,5-dimethylthiazol-2-yl)-2,5-diphenyl tetrazolium bromide] and annexin V-FITC/PI assays. P1CF1 was spherical, had a size of 193.46 ± 0.10 nm, a PDI of 0.342 ± 0.12, a ZP of -7.67 ± 0.04 mV, and an %EE of 91.23 ± 0,01%. Furthermore, P1CF1 was biocompatible at low concentrations, showed a controlled in vitro drug release, induced high cytotoxic (IC50 value of 2.967 µg/ml) and apoptosis effects on the cancer cells MCF-7 and was well tolerated by non-cancer HEK293 cells (IC50 value of 135.3 µg/ml). Overall, P1CF1 showed efficient encapsulation capacity, enhanced biocompatibility, and significant anticancer activity in tumour cells with minimum effect on healthy cells. These positive characteristics indicate potential in vivo applicability. Thus, future research can include in vivo evaluation of this novel formulation. 

Cancer is a significant public health challenge as it is the second-leading cause of death in the world [1]. From the many available cancer treatment modalities, chemotherapy is regarded with greater significance in cancer treatment options and will most likely remain so for many decades to come [2]. However, though cancer therapy with conventional chemotherapeutic drugs has some benefits, the direct administration of chem-drugs is associated with severe limitations, such as rapid elimination, low bioavailability, lack of specificity, which can result in systemic toxicity and adverse consequences; and multidrug resistance (MDR), which can lead to recurring, unresponsive tumors and inadequate drug dose, which can impede certain apoptotic pathways and hinder cell death processes [3-6]. Thus, methods to address these challenges are still required. 

The use of lipid-based delivery systems such as liposomes as drug carriers in nanomedicine has gained popularity due to their sophistication and attractive traits [7]. These include flexibility, ease of synthesis, reliability, versatility, nano-size (50-500 nm), biocompatibility, enhanced bioavailability, biodegradability, the capacity to shield payloads from intracellular enzyme degradation; targeted drug delivery to tumor tissues; enhanced effectiveness and therapeutic index; the capacity to encapsulate drugs that are both hydrophilic and hydrophobic; high levels of drug loading and entrapment efficiency; regulated drug release profiles; reduced toxicity of the encapsulated drugs; enhanced pharmacokinetic effects including reduced excretion, prolonged circulation lifespans; and ability to be tailor-designed for targeted drug delivery [8-12]. Because of the leaky nature of tumor tissue capillaries, these vesicles' spherical form and micro size allow them to extravasate and passively accumulate in cancer regions, a phenomenon known as the increased permeability and retention (EPR) effect [13]. To date, numerous liposome formulations have been approved for cancer therapy, including OnivydeTM, Marqibo®, Doxil®, Visudyne®, and Depocyt® [14]. Future research and innovation in liposomal drug delivery systems hold considerable potential for the advancement of pharmaceuticals and nanomedicine.

The concept of peptide-targeted liposomes is a significant advancement in the usage of liposomal formulations in cancer treatment. Peptides have been identified as attractive and viable targeting ligands for directing liposomes to target tumours and sequentially enhancing the selectivity and specificity of drug-loaded liposomes, minimizing off-target delivery [15, 16]. This is supported by many in vivo and in vitro research that have been documented over the years [16-24]. However, there is a huge gap in the literature and drug development focussing on liposomal encapsulation of anticancer peptides (APCs) that target solid tumours. 

Recently, ACPs have vastly become a developing approach in nanomedicine, particularly, in drug development. These are a series of short amino acids, with anticancer properties which are alternatives to chemo-drugs used in cancer treatments. These ACPs cause cell death by different mechanisms, such as mediated immunity, membrane disruption apoptosis, DNA synthesis inhibition, hormonal/membrane receptors, and anti-angiogenic [25, 26]. Though these drugs exhibit great anticancer properties, limited studies have been done on them as therapeutic payloads, and this could be due to their lack of bioavailability because of intracellularly degraded following administration. Therefore, extensive research is required to establish whether liposomes can encapsulate these ACPs and deliver them safely and efficiently to tumour sites. Thus, this study aimed to design and advance a novel anti-cancer peptide using computational-aided drug design tools, encapsulate it into a liposome and evaluate its physicochemical features as well as its biological activity in vitro. 

2.1 Materials 

Novel anticancer peptides (sequence: FKKLLAKLAK) were designed in-house using the lead anticancer peptide purchased from ChinaPeptides (QYAOBIO) Ltd, China, by a solid phase peptide synthesis protocol. Cholesterol and phosphatidylcholine that were purchased from Sigma-Aldrich (USA) were used for liposomal preparations. Trifluoracetic acid (TFA) acetonitrile, phosphate-buffered saline (PBS), dichloromethane, and a dialysis bag with a molecular weight cut-off (MWCO) size of 10,000 Dalton (Da) was purchased from Sigma-Aldrich (USA). All other solvents used were of High-Performance Liquid Chromatography (HPLC) analytical grade and were used without additional purification. Distilled water was used throughout this study and was purified in the laboratory with a Milli-Q purification system (Millipore Corp., USA). 

2.2 Methods 

2.2.1 Anticancer Peptide Design 

The amino acid sequences of the purified peptides were determined using the automated online software CancerPPD (Database of Anticancer Peptides and Proteins) (http://crdd.osdd.net/raghava/cancerppd/) [27]. The peptide sequence FAKLLAKLAK with ID 1854 and a 10 amino acid chain, was used as the anticancer reference sequence that targeted solid tumours - breast cancer. The reference amino acid sequence was incorporated into CellPPD: Designing of Cell Penetrating Peptides (http://crdd.osdd.net/raghava/cellppd) online software to select mutants of the reference compound. A mutant peptide was generated with a 10 amino acid length, FKKLLAKLAK. The lead anticancer peptide was then purchased from ChinaPeptides (QYAOBIO) Ltd., China The sequence FKKLLAKLAK (Phe-Lys-Lys-Leu-Leu-Ala-Lys-Leu-Ala-Lys) was identified with a purity of 98.28% and was soluble at 1 mg/ml in 17

Over the years, therapeutic peptides have seemingly received significant attention from scientists as potential drug candidates [36]. Peptides as therapeutic agents have proved to be favourable for many diseases, including cancer, and their applications have been highly advantageous due to their size, high biocompatibility, simplicity in production, and/or modification together with their capacity to penetrate tumours [37]. Much research has been conducted on novel therapeutic ACPs using in silico tools to target various diseases. Consequently, there have been many ACPs that have entered clinical trials but still only a few have been approved. 

3.1 Identification and Design of Novel Anti-Cancer Peptides (ACPs)

The ACP was designed using the module CancerPPD database (Database of Anticancer Peptides and Proteins) which has already been validated with anticancer activity [38]. For each peptide on the CancerPPD database, adequate details on the specific assays utilized and the experimentally measured activity of the peptides against different cancer cell lines are provided [39]. During selection, the peptide sequences needed to be short, thus lengths ranged from 5 - 10 amino acids due to the consideration of synthesis and cost of the peptide. The initially selected sequence was FLAK50 T1 (FAKLLAKLAK), and the strategy for novelty was to adjust the amino acids in the given sequence. The sequence FLAK50 T1 originates from FLAK peptides that are highly rich in Phe, Leu, Ala, and Lys and were compiled in the Owen patent dataset [40]. The peptide targets breast cancer and has chirality L, the sequence was linear and the reported activity was LD50 = 615 µg/ml. The retrieved sequence from CancerPPD was then taken to CellPPD to develop mutant peptides of the sequence. CellPPD database, which is a support vector machine (SMV), assists by developing and designing cell-penetrating peptides (CPPs), which was essential when looking for mutants of the retrieved sequence [27, 41]. CellPPD allows for the design of single mutant analogues of given peptide sequences and identifies whether they are penetrating cells or not. Additionally, CellPPD also provides the physicochemical properties of the generated mutant peptide. From the retrieved sequence, Ala amino acid was substituted with Lys to generate the mutant peptide. The newly generated peptide had the sequence FKKLLAKLAK (Phe-Lys-Lys-Leu-Leu-Ala-Lys-Leu-Ala-Lys) (Fig 1). The peptide had a molecular weight of 1159.67 and a +4 net charge, which advantageously leads to the destruction of cancer cells by engaging with their anionic cell membrane components [42]. The SVM score was 0.30 and the peptide had a clear CPP prediction. For ACPs, hydrophobicity plays an important role in the peptide's ability to cross membrane barriers and hence exert their anticancer effects [43]. The ACP had a hydrophobicity of -0.17, which was ideal as the peptide showed to be mostly hydrophilic. The hydrophilicity was 0.31 and the hydropathicity was 0.22. 

Figure 1: 3D structure (a) and 2D structure (b) of the designed ACP - FKKLLAKLAK (Phe-Lys-Lys-Leu-Leu-Ala-Lys-Leu-Ala-Lys). Images drawn using ChemDraw Software

3.2 Characterisation of ACP Encapsulated Liposomes 

3.2.1 Cryo-TEM, DLS-Zeta Sizer

The physicochemical properties of the drug delivery systems have a substantial impact on their tumour permeability, biodistribution, and blood circulation half-life [44]. The ACP-loaded liposome formulations and blank liposomes (without ACP) were successfully formulated using the thin-film hydration method. A small volume of cholesterol was included in the formulation to increase the stability of the lipid bilayer in biological fluids such as blood plasma [45].

Cryo-TEM images revealed that all formulated liposomes formed spherical structures with similar homogenous sizes (Fig 2). The DLS-Zeta Sizer revealed that the sizes of the liposome formulations ranged from 164 nm – 194 nm, with the blank liposome having a mean diameter size of 164.1 ± 0.11nm, while the ACP liposome has a slightly larger mean diameter size of 193.4 ± 0.10 nm (Table 1). The increase in size could have been due to the presence of the ACP within the liposome. Studies have shown that NPs larger than 200 nm in diameter tend to activate the complement system, which causes them to leave the bloodstream fast and accumulate in the liver and spleen [44]. Therefore, the formulated liposome sizes fell within the standard size range which is between 100 nm and 200 nm needed for drug delivery through non-specific or receptor-specific endocytosis cellular uptake [46-49]. The use of probe sonication could have had a major influence in obtaining the ideal particle size, as this technique can rearrange and reassemble the lipid content of the liposome to create favourable particle sizes [50]. The PDI value which is correlated to the distribution stability of the formulation, is an important indicator of the overall liposome size distribution [51]. According to literature, PDI values, close to 1.0 are considered not ideal as they may indicate the inverse distribution of the particles and/or the presence of large particles [52]. Ideally, PDI values should range between 0.30 and low, which indicates that over 60% of the nano-formulations are within the same particle size range and are evenly distributed [51-53].     From the results obtained, the PDI of the blank was below the optimum 0.30, however, the ACP encapsulated liposomes ranged close to the optimum (0.294 ± 0,12 nm to 0.394 ± 0,12 nm) thus showing high-to-medium homogeneity of the liposome mean sizes.

To investigate the physical stability of liposomes, zeta potential was measured. Zeta potential distinguishes the particle surface charge, provides data on the repulsive forces that exist between the particles, and aids in colloidal dispersion stability estimations [54]. Good colloidal stability of nanoparticles is related to ZP values greater than +/- 25 mV [55, 56]. All the liposome formulations in the study presented negative ZP values (negative charge), which could be due to the phosphatidylcholine’s (used in the thin film) headgroup orientation located at the vesicles surface, and the position of the phosphate group above the choline group plane [57, 58]. The blank liposomes exhibited a ZP of -7.73 ± 0.47 mV, while the ACP liposomes showed a ZP of -7.67 ± 0.04 mV. These findings suggested that the ACP liposome formulation displayed low colloidal stability, which has no profound impact in in vitro studies. However, before the application of the formulation in vivo, the inclusion of a cationic phospholipid or cationic polymer, helper lipid, and stealth polymer (e.g., polyethylene glycol, PEG) is mandatory for improved colloidal stability and longer circulation half-life of this system.

Figure 2: TEM micrographs of the P1CF1 formulation at different resolutions showing efficient encapsulation of the peptide in the liposome. Scale bars = 0.2 µm; 100 nm.

Table 1: Particle size, polydispersity Index (PDI), and zeta potential (ZP) of blank liposomes and ACP encapsulated liposomes. Data displayed as mean ± SD (n = 3).

3.3 Determination of Encapsulation Efficiency Percentage

The analysis of liposome encapsulation efficiency (EE) is an extremely crucial parameter since liposomes are utilized in pharmaceuticals as drug carriers that facilitate drug absorption, targeting effects, and protection of drugs [59]. According to the literature, the EE% is calculated from the difference between the entrapped drug (the total amount of drug added to the liposome formulation) and the unentrapped drug (the amount of drug found in the supernatant of the resulting liposome formulation) divided by the total drug added (EE% = total drug added – free non-entrapped drug / total drug added) [60]. In this study, the encapsulation efficiency of the liposome was determined using RP-HPLC at wavelength 220 nm. The standard curve of the formulation was made by plotting HPLC peak areas against the concentration. The steps indirect method used for the entrapped and unentrapped values were reported by Suleiman and coworkers [61], however, the dilutions used in this study were with acetonitrile. The concentrations of the entrapped and unentrapped drugs were calculated using the calibration curve with the equation y=5045,2x. To measure the amount of active drug loaded in a liposome, the drug would have had to be fully encapsulated by the lipid of the liposome formulation. To determine the EE, the mass ratio between the amount of the drug integrated into the liposome and this ratio was employed in the liposome preparation. The overall entrapment yield was calculated to be 91.23 ± 0.01%, which is relatively high; this was expected as many liposomes have been reported to possess high EE because of their high volume to surface area ratio [9]. The relatively high EE% indicates that the ratio of the used lipids to the ACP was an optimal choice for the formulation. The obtained results from the study revealed high EE and this may be due to many factors, one of them being the effect of probe sonication. According to the literature, liposomal formulations that have been probe-sonicated for longer tend to be more homogenized and hence more susceptible to interacting with surrounding molecules, which greatly influences their enhanced encapsulation efficiency [62]. Additionally, the hydrophilic nature of the peptide used herein could have further facilitated the observed high EE%. This is according to previous studies which have shown that the encapsulation of hydrophilic peptides (as with the one used in this study) in liposomes is expected to be the most efficient due to the electrostatic interaction between the peptide and the liposome surface [61, 63].

3.4 In vitro Hemolysis Testing 

The in vivo application of drugs/drug-loaded nanoformulations entails their transportation via the bloodstream which could lead to adverse effects, such as immunological responses, complex formation with macromolecules, and cell damage [64]. Exposure to chemicals, such as drugs or drug-loaded nanoformulations, can result in an early breakdown of erythrocytes of red blood cells releasing the hemoglobin (Hb), which can disrupt normal oxygen transport and induce hemolytic anemia/hemolysis [65, 66]. The drugs/drug-loaded nanoformulations may either adsorb on the membrane of the erythrocyte, causing the membrane to distort and become damaged [67]; or they have the ability to cause osmotic lysis by causing holes in the erythrocyte membrane [68]. Thus, it is important to determine if drug-loaded nanoformulations with therapeutic effects induce hemolysis in erythrocytes in vitro [69]. To establish this, erythrocytes of sheep red blood cells were exposed to different doses of the ACP encapsulated liposome formulation (P1CF1) and evaluated for possible hemolysis [70]. 

The percentage (%) of erythrocyte hemolysis induced by the P1CF1 formulation is shown in Fig 3. Both the graph and the image insert depicted that the formulation was non-toxic to erythrocytes at doses 0.05 and 0.1 mg/ml, suggesting its biocompatibility at these low therapeutic doses. Hemolysis was neglectable at a dose of 0.2 mg/ml and increased with increasing concentration of the P1CF1 formulation from doses greater than 0.2 mg/ml up to 0.5 mg/ml. This can be credited to the agglomeration feature of ACP liposomes at higher doses. Moreover, compared to untreated cells (C2), an 88% increase in hemolysis was observed at a dose of 0.5 mg/ml post-treatment, indicating that the formulation had an adverse effect on erythrocytes at higher doses. Overall, these findings indicated that the formulation was non-toxic at low therapeutic concentrations equal to or less than 0.2 mg/ml. 

Figure 3: Hemolysis toxicity of the P1CF1 formulation towards red blood cells at varying concentrations (0.05 to 0.5 mg/ml). Data is shown as means ± SD (n = 3). The image insert shows the representation of the haemolytic behaviour of RBCs after treatment with P1CF1 formulation. C1 denotes control 1 = positive control (+) = RBCs + distilled water; C2 denotes control 2 = negative control (-) = RBCs + PBS. ∗∗∗∗p < 0>.0001 vs. C2; ns indicates non-statistical significance from C2. 

3.5 In vitro Drug Release Analysis 

To predict the quantity of a drug's accumulation into the bloodstream and tumour sites over time, in vitro drug release studies are usually conducted. The in vitro drug release profile of the ACP encapsulated liposomes (P1CF1 formulation) was assessed using a dialysis bag technique with PBS solutions at pH 6.0 and 7.4 prepared at 37°C for 72 hrs. These pH buffer solutions mimic the basic physiological state and the acidic endosomal cancer cell microenvironment (pH 4-6) [71]. Bare peptide was used as a reference or negative control. There was only 5.1% and 3.7% of 

bare peptide released at pH of 7.4 and pH 6.0 respectively at 72 hrs (Figure 8 in the Supplementary Information). This was expected since the peptide was not encapsulated in any nanoparticle, so, its large size could have restricted its release or diffusion across the semipermeable dialysis bag membrane. 

Figure 4 displays the release profile of P1CF1 formulations and the accumulative release of loaded ACP, which was sustained/controlled, acid-dependent, and exhibited a biphasic release pattern across the time studied. During the first 10 hours, there was a quick release of the APC, followed by a slow controlled release of the APC for the remaining duration of 72 hrs. A higher ACP release rate of 88% was seen at an acidic pH of 6.0 whereas a slower drug release rate of 79% was seen at a healthy pH of 7.4. The initial burst ACP release may have been caused by the release of the unentrapped ACP on the liposome periphery, while the  subsequent slow and controlled release may have been caused by the release of the ACP encapsulated inside the liposome [71, 72]. The protonation of the amine groups of the encapsulated ACP (see Fig 1) at acidic pHs results in a conformational change of the liposome, causing swelling, bursting, and releasing of the encapsulated ACP into the buffer solution [73]. Sustained drug release is of high importance for cancer therapeutics as it enables the drugs to be released for longer durations, ensuring a continuous stable dosage of the drug to the tumour sites [74]. The cholesterol present within the liposome also helps to regulate the properties of the lipid bilayer of the liposomes as well as the release of water-soluble compounds (i.e. the ACP peptide) from liposomes by influencing the fluidity and permeability of the lipid bilayer [1, 75]. The pH-dependent drug release property raises the possibility that this P1CF1 formulation could be employed to deliver anticancer drugs specifically to tumours (with an acidic microenvironment) [74, 76]. Additionally, the results indicate that most of the ACP encapsulated within the   liposome will remain intact in the plasma (pH 7.4) after injection, greatly decreasing the possibility of any systemic adverse effects on the healthy tissues. Furthermore, the results showed that once inside the tumour cells, the cytosolic release of the ACP-encapsulated liposomes/ P1CF1 from the endosome is expected to be quicker due to the lower pH than physiological pH conditions. Consequently, it is hoped that the encapsulation of ACP into liposomes will improve the effectiveness of targeted cancer treatment.

Figure 4: In vitro drug release profile of ACP encapsulated liposome in pH 6.0 and 7.4 over 72 hours. Data is shown as mean ± SD (n = 3).

3.6 In vitro Cytotoxicity Analysis 

The initial step in assessing a drug delivery system's biocompatibility usually includes the use of cell-culture-based research such as in vitro cell viability studies, which typically starts with the assessment of its cytotoxicity profile. The MTT test that was used in this study quantifies metabolically active cells by measuring yellow formazan products produced from the conversion of purple MTT salt by mitochondrial reductase enzyme found in live cells [34]. Untreated cells served as negative controls, azacitidine, a known anticancer drug [77], was used as a positive control. 

MCF-7 cancer cells and HEK293 non-cancer cells were treated with ACP-encapsulated liposomes (P1CF1), bare peptide, and azacitidine at varying doses ranging from 10 to 200 µg/ml (Figure 5-6). Following treatment, a drastic decrease in

A novel anticancer peptide-encapsulated liposome formulation (P1CF1) was successfully designed and formulated using in silico and thin-film layer rehydration methods, respectively. The formulation had optimal physiochemical characteristics for efficient drug delivery in vitro, as well as improved encapsulation efficiency. Encapsulating the novel anticancer peptide in a liposome delivery system improved its in vitro bioavailability and biocompatibility. At acidic pH conditions, the formulation also demonstrated increased controlled and sustained in vitro anticancer peptide release, implying that treatment at the target tumour sites might be prolonged. The P1CF1 formulation had higher cytotoxicity than the known anticancer drug, azacitidine in cancer cells, with a percentage cell inhibition of 98.2%. However, it was ineffective in non-cancer cells (

project conceptualization and designing, Lusanda Mtetwa, Nkeiruka Igbokwe, and Mbuso Faya.; resources, Mbuso Faya and Anil Chuturgoon.; data collection.; Lusanda Mtetwa.; data curation and software, Lusanda Mtetwa, Nkeiruka Igbokwe, Aviwe Ntsethe, Makabongwe Mazibuko and Terisha Ghazi.; writing - initial draft preparation, Lusanda Mtetwa.; writing - final draft preparation, Lusanda Mtetwa, Nkeiruka Igbokwe and Londiwe Simphiwe Mbatha.; final draft reviewing and editing, Londiwe Simphiwe Mbatha and Mbuso Faya.; funding acquisition, Mbuso Faya and Anil Chuturgoon.; project supervision and administration, Mbuso Faya. All authors have read and agreed to the published version of the manuscript.

The University of KwaZulu Natal, College of Health Science is acknowledged for financial support as well as the National Research Foundation (grant number: 138291). Any opinions, findings, conclusions, or recommendations expressed in this article are those of the authors.

Supporting data are available from the corresponding author upon request.

Conflicts of interest: The authors declare no conflict of interest.

Declarations: Ethical approval, is not applicable.

Fig 8. In vitro drug release profile of bare ACP in pH 7.4 and pH 6.0 over 72 hours. Data is shown as mean ± SD (n = 3); Fig 9(a-d). IC50 estimated values of P1CF1 (a), bare ACP (b), and azacitidine (c) in MCF-7 cell lines; IC50 estimated value of P1CF1 (d) in HEK293; Table 3. Flow cytometry analysis; gate total percentages of MCF-7 only, P1CF1, Bare ACP, and Azacitidine.