Bacterial strains, cell lines, plasmids and reagents: Top 10 and BL21 (DE3) Escherichia coli (E. coli) strains and cell lines including HeLa (Human cervical cancer cell line) and Human Umbilical Vein Endothelial Cell (HUVEC) were provided from Pasteur Institute of Iran (Tehran, Iran). Ampicillin and IPTG was purchased from Sigma (San Diego, California, USA). Chitin resin and pTWIN-1 vector were purchased from New England Biolabs (USA). The mentioned vector was transferred to Biomatik Company (Canada) for sub-cloning of desired genes including TP4-LYC1, TP4, and LYC1. All buffers for the purification were prepared according to the IMPACT^TM manual, otherwise mentioned.

Soluble expression of TP4-LYC1, TP4, and LYC1 molecules: The selection of E. coli BL21 (DE3) colonies harboring the recombinant vectors including pTWIN-1-TP4-LYC1, pTWIN-1-TP4, and pTWIN-1-LYC1 was performed on LB-agar plates containing 100 µg/mL ampicillin. After the overnight cultivation, fresh cultures were inoculated and upon reaching an OD600 of 0.6, total expression of peptides in fusion to intein 1 and 2 of pTWIN-1 vector was induced by 1 mM IPTG for 4 hr at 37°C. Furthermore, in order to produce the mentioned proteins in soluble form, the expression was optimized at various IPTG concentrations (0.1, 0.3, and 0.5 mM) in different incubation temperatures as 7, 22, and 37°C for 17 hr. In each stage, the cells were harvested via centrifugation at 7000×g for 10 min at 4°C. Finally, evaluation of the protein expression was performed by 12% SDS-PAGE.

Peptide purification: The IMPACT^TM purification system (New England Biolabs, Massachusetts, USA) was used to purify the recombinant peptides. Based on the N- and C- terminal cloning, the self-cleavage activity of inteins was induced by lowering the pH level from 8.5 to 6.5 for intein 1 and Dithiotritol addition to the related buffer for intein 2.

In detail, induction of intein cleavage from all fusion proteins was performed as follows: First, B1 buffer (Tris-HCl 20 mM, NaCl 500 mM, & EDTA 1 mM, pH=8.5) was added to the bacterial pellets and mixed using a vortex mixer. Then, for cell lysis, sonication was performed followed by the samples centrifuge at 7000 rpm for 30 min. The supernatant was poured into a column containing chitin resin and incubated for 30 min. The sample was passed through the column and then, the column was washed, at least 5 times, with B1 buffer. After that, 5 ml of B3 buffer (Tris-HCl 20 mM, NaCl 500 mM, EDTA 1 mM, 50 mM DTT, Triton-X100 0.3%, & Tween-20 0.2%, pH=8.5) was added to the column and left at room temperature for 24 hr. In the next day, the column supernatant was discharged and 5 ml of B2 buffer [Phosphate-Buffered Saline-(PBS), pH=6.5] was added to the column and put at room temperature for 24 hr. Finally, the purified column samples were collected and analyzed by Tris-Tricine SDS-PAGE.

Cytotoxic activity evaluation: The sterilization of each peptide was performed by 0.22 µM filtration and serial dilution of each peptide sample by the addition of PBS. Evaluation of the cytotoxic effects of recombinant peptides including TP4-LYC1, TP4, and LYC1 was performed by MTT assay on HeLa cancer cells compared with HUVEC as normal cells. Briefly, cells were cultured in RPMI 1640 culture medium enriched with 10% (v/v) Fetal Bovine Serum (FBS) and antibiotics (100 IU/ml penicillin and 100 µg/ml streptomycin) and incubated in the incubator at 37°C and 5% CO₂. In this test, 180 µl of RPMI medium containing cells (in final concentration as 3×10⁴ cell/ml) was added to each well of a 96-well plate and incubated at 37°C for 24 hr. The next day, 0.62, 1.25, 2.5, 5, and 10 µM of each peptide were added to the wells in the 20 μl final volume and incubated at 37°C for the other 48 hr. Exactly 20 μl of MTT solution (5 mg/ml) was added to each well and the plate was incubated in a CO₂ incubator for 3 hr. Then, 150 μl of Dimethyl Sulfoxide (DMSO) was added to each well and slowly pipetted to dissolve the formazan crystals and the OD was measured at 570 nm using a microplate reader (Bio-Rad, US). PBS treated cells were used as the negative control and the cell free medium culture used as the blank.

Statistical analysis: In the biological assay stage, MTT test was repeated triplicate during independent experiments with four repeated wells for each concentration of peptides. The results were expressed as cell viability percent±SD. Data analysis was performed by SPSS 23 software. ANOVA and Tukey’s post hoc test was used to distinguish the differences between or among groups. The significance was assumed as p<0.05.

Expression and purification of peptides: SDS-PAGE analysis confirmed the expression of TP4-LYC1, TP4, LYC1 fused to intein 1 and 2 with bands at 60, 57, and 57 kDa, respectively. On the other hand, the optimum conditions for the soluble expression of each peptide was similar [0.5 mM IPTG at 22°C (Figures 1 and 2)]. After the induction of cleavage, the inteins 1 and 2 from the recombinant proteins, the appearance of bands at approximately 6, 3, and 3 kDa for TP4-LYC1, TP4, and LYC1, respectively on Tricine SDS-PAGE proved successful purification by the IMPACT system (Figure 3).

Finally, according to the Bradford method, in the optimized condition, the final yield of the purified peptides was calculated as 432, 571 and 477.5 μg/L of bacterial culture medium for TP4-LYC1, TP4, and LYC1, respectively.

Biological assay: The results of statistical analysis showed that approximately all peptides reveal their cytotoxic effects in a concentration-dependent manner. Actually, except LYC1 when used for the treatment of HUVEC cell line, all other peptides showed these effects. For HeLa cells, there was significant differences between the cytotoxic effects of the chimeric peptide in comparison to the negative control. Furthermore, the significant difference between the cytotoxicity of TP4-LYC1 and TP4 was established in the same concentrations. Finally, in the case of LYC1, the observed toxic effects were statistically lower than the two other peptide. However, a significant toxic effect was shown only in the higher concentration (10 µM) in comparison to the negative control (p-value=0.02) (Figure 4).

Regarding the HUVEC cells, the first finding is this fact that LYC1 showed no significant toxicity even in the higher concentration (10 µM) when compared to the negative control. Furthermore, in the same molar ratios there was no significant differences between the cytotoxicity of TP4 and TP4-LYC1. However, there was significant difference in the toxicity between TP4-LYC1 and the targeting moiety (LYC1) in the same concentrations (Figure 4).

Finally, based on the concentration vs. survival percent graph, the calculated IC50 of each surveyed peptides against two cell lines are reported as bellow:

For HeLa cells, the IC50 value was about 0.83, 2.75, and higher than 10 µM for TP4-LYC1, TP4 and LYC1. This value was not calculable for HUVEC cell line in the surveyed concentrations.

The aim of the present study was to produce a new recombinant fusion peptide containing TP4 with established cytotoxic effects and LYC1 as an AMP with safe profile against normal cell lines and successfulness in the usage as a drug delivery vehicle, in order to enhance the penetration ability of TP4 to cancer cells and diminish its toxic effects on the body normal cells and tissues.

In the first stage, the recombinant DNA technology was used as a suitable strategy to earn a fusion peptide containing 49 amino acid residues instead of its chemical synthesize in order to overcome the drawbacks of this method to access the enough amounts of this peptide. In the attempts to produce this chimeric peptide in soluble form and consequently with correct folding, we used intein tags and besides that, we used the optimization of the expression conditions in order to obtain the maximum amount of soluble peptides. The main concern in the production procedure of these AMP was their potency to kill the bacterial host cell after the production. We used several solutions to overcome this problem. First, we used lower post-induction temperature in order to diminish the activity of these antimicrobial peptides against E. coli expression system. Although lowering the temperature after the expression induction leads to diminish the protein expression, the chance of soluble expression increases. This strategy was used for the soluble expression of various peptides and proteins. For example, our previous fusion protein, DFF40-iRGD, was only produced at 7°C ¹⁷.

The other solution we used in this project was the fusion of the AMO to the intein tags, the large amino acid sequences in order to cover the alpha-helical structure of the peptides responsible for the bacterial membrane penetration and their antibacterial effects. In our pervious study, in an attempt to produce BR2 AMP, the fusion to the gyr A intein in its C-terminal was performed. Generally, production of various AMP in fusion to the inteins with auto-cleavage ability is a suitable approach that not only prevents their cell lysis activity, but also facilitates their purification stage. The other examples in this regard are human β-defensin 2 and LL-37 which are fused to the ΔI-CM mini-intein, expressed in aggregate form and after the self-cleavage induction, they were produced in soluble form with antimicrobial activities ¹⁸. In this example, it is obvious that in the expression stage, the produced AMP are safe for the expression of the host because of aggregation.

As shown in the results section, lowering the inducer concentration also leads to the soluble expression. Based on this fact, IMPACT protocol advises to use 0.4 mM IPTG as an inducer to improve the expression in soluble form. However, we used lower IPTG concentrations in order to diminish the cost of peptide production as well as improve the soluble expression. In our previous study, BIF1-iRGD fusion protein was only produced in soluble form by 0.05 mM of IPTG ¹⁹. In the present study 0.1 mM of IPTG leads to the soluble expression.

In the purification stage, on the other hand, we had to use two independent strategies in order to induce the auto-cleavage activity of two various inteins. The covering of two sites of peptides not only helps to prevent their bacterial lysis properties but also leads to the more soluble expression in comparison to the situations in which only one intein tag is used. However, the usage of two inteins complicates the self-cleavage induction. We have to use B2 buffer for inducing the intein 1 with pH=6.5 and B3 buffer containing DTT in order to induce the self-cleavage activity of intein 2. pH-shift in order to delete the intein 1 is a simple strategy for the cleavage of inteins but considering that the change in this item can easily occur at any stage of expression and purification, it leads to the induction of self-cleavage and reducing the final yield of peptide/protein product. One of the reasons for using low temperature in the expression stage is to reduce this process because the induction of self-cleavage occurs more likely at room temperature and the use of low temperatures in the expression stage prevents this process before the attachment of fusion protein to the chitin column. Various pH values were surveyed to determine the best one for more self-cleavage induction and it was shown that it is dependent to the protein type. For example, in our previous work for the recombinant production of a mutant form of IL-6R, the results showed that between three different pH, 4, 5.5, and 6.5, the most self-cleavage induction occurred at pH=4 ²⁰. While for the other protein, IL-1Ra, we produced more protein with a B2 buffer pH=6.5 ²¹.

For the intein 2, on the other hand, the self-cleavage induction is a more complicated procedure than the intein 1; in this case, this process is related to the amino acid in the C-terminal of the target protein. Actually, for all produced peptides, (TP4 with an Arginine and TP4-LYC1 and LYC1 with a leucine at its C-terminal), down-stream intein (intein 2) was separated after 24 hr of incubation at room temperature. For BR2, terminated with an arginine residue, the same condition was used in our previous study ²².

Finally, in the case of biological study, as mentioned in the introduction section, TP4 was investigated for its anti-tumor effects in several studies. For example, Ting et al, surveyed the selective necrosis induction of this peptides against breast cancer cell lines including MDA-MB231, MDA-MB453, and MCF7 and their results showed that concentration of 7.5 µM of this peptide led to the significant toxicity after 24 hr of treatment in all cell lines especially MCF-7 and MDA-MB231 ¹². The concentration used in the mentioned study was similar to the concentrations in our study. However, we saw more toxicity of this peptide probably due to the more incubation time as 48 hr.

The other example in this regard is the investigation of antitumor activity of this peptide against A549 cancer cell. When this peptide was used in 18 µg/ml (6 µM) for 3 hr, microtubule network in this cell was disrupted. So, we can conclude that the concentrations which were surveyed in this study can be used for the other biological tests such as determination the cell death mechanism, microtubule disruption and so on. In the mentioned study, the penetration ability of this peptide was analyzed; TP4 showed penetration efficacy around the cancer cell membrane. However, there is no data about targeted penetration using normal cell lines ¹⁰. The other study in this regard, used normal lung cell lines as the control, and when compared to the normal cells, cancer cell lined showed more toxicity than normal cells but it seems that this difference is not significant based on their calculated IC50 for 3 hr incubation. However, over the time, this difference became more significant and this indicates that the cell entry of the peptide requires a minimum time. Another finding of this study was that this peptide activated the caspase-3 and induced apoptosis in 6.71 µM ¹¹.

In our study, to further ensure the specific and targeted function of TP4, another peptide was added to its structure, which had been evaluated till now for the targeted delivery of gold nanoparticles to different cancer cells and showed no toxicity on normal cells such erythrocytes and HEK293 cells ¹⁶. It was the only study in this regard for targeting a therapeutic agent for cancer cells and as reported in the results section, LYC1 did not show any toxicity either on normal cells or on cancer cells. However, compared to TP4, the amount of toxicity against cancer cells increased in the chimeric peptide in equal molar ratios and this can confirm more penetration of the fusion protein than native TP4.

In the present study, we tried to produce a recombinant form of chimeric peptide in order to decrease the chance of cytotoxic effects of TP4 against normal cells of the body. Although the production of these three peptides occurred in low yield, and also because of the suitable potential to act as a targeted therapy for the eradication of cancer cell, attempt to increase the final yield of production is extremely advised. In this regard, the usage of various auto-cleavage condition such as pH, different types of reducing agents, other additives such as Triton-×100 and tween-20 and finally changing the incubation time and temperature can be suggested for ongoing studies. However, other strategies for increasing the soluble protein production and purification are urgent stages in order to continue the other biological assays on the fusion peptide. Thus, this fusion peptide can act as a safe bio-molecule with potent cytotoxic effects against cancer cells. Meanwhile, in the first stage for the future studies in this regard, determination of penetration ability and the cell death mechanism must be performed in order to have more precise view in its action as a tumor targeted agent.

The authors would like to appreciate valuable technical assistance of laboratory experts in molecular biotechnology and cell culture laboratories. The content of this paper was extracted from the grant with number as 198278, financially supported by Research Deputy of Isfahan University of Medical Sciences and approved by the ethics committee of IUMS with code IR.MUI. RESEARCH.REC.1398.820.

Funding: The content of this paper was extracted from the grant with number as 198278, financially supported by Research Deputy of Isfahan University of Medical Sciences.

This article does not contain any studies with human participants or animals performed by any of the authors. The Ethics Committee of Isfahan University of Medical Sciences approved this research with the code of IR.MUI.RESEARCH.REC.1398.820.

All authors confirm that they have no conflict of interest.