- Email: [email protected]
pH-controllable cell-penetrating polypeptide that exhibits cancer targeting
Accepted Manuscript Full length article pH-controllable cell-penetrating polypeptide possessing cancer targeting DaeYong Lee, IlKoo Noh, Jisang Yoo, N.Sanoj Rejinold, Yeu-Chun Kim PII: DOI: Reference:
S1742-7061(17)30331-8 http://dx.doi.org/10.1016/j.actbio.2017.05.040 ACTBIO 4899
To appear in:
Acta Biomaterialia
Received Date: Revised Date: Accepted Date:
17 February 2017 11 May 2017 16 May 2017
Please cite this article as: Lee, D., Noh, I., Yoo, J., Rejinold, N.Sanoj, Kim, Y-C., pH-controllable cell-penetrating polypeptide possessing cancer targeting, Acta Biomaterialia (2017), doi: http://dx.doi.org/10.1016/j.actbio. 2017.05.040
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
pH-controllable cell-penetrating polypeptide possessing cancer targeting DaeYong Lee, IlKoo Noh, Jisang Yoo, N.Sanoj Rejinold, Yeu-Chun Kim* Department of Chemical and Biomolecular engineering, Korea Advanced Institute of Science and Technology (KAIST), Daejeon 305-701, Republic of Korea. Corresponding Author: Prof. Yeu-Chun Kim, TEL: +82-42-350-3939, FAX: +82-48-3503910, E-mail: [email protected]
Abstract Helical peptides were naturally-occurring ordered conformations that mediated various biological functions essential for biotechnology. However, it was difficult for natural helical polypeptides to be applied in biomedical fields due to low bioavailability. To avoid these problems, synthetic alpha-helical polypeptides have recently been introduced by further modifying pendants in the side chain. In spite of an attractive biomimetic helical motif, these systems could not be tailored for targeted delivery mainly due to nonspecific binding events. To address these issues, we created a conformation-transformable polypeptide capable of eliciting a pH-activated cell-penetrating property solely at the cancer region. The developed novel polypeptide showed that the bare helical conformation had a function at physiological conditions while the pH-induced helical motif provided an active cell-penetrating characteristic at a tumor extracellular matrix pH. The unusual conformation-transformable system can elicit bioactive properties exclusively at mild acidic pH.
1. Introduction α-helical structures mediate protein folding or unfolding, and are the fundamental foundation for the protein assembly and bioactive functions.[1, 2] Among various α-helical peptides, cationic helical peptides (CHP) have been used in biomedical fields due to their unique cell-penetrating property. CHPs composed of sterically unhindered and lysine or arginine-rich amino acid residues strongly bind to the lipid plasma membranes via electrostatic attractions, and then distort the arrangement of lipid molecules, thereby being translocated in the cytosol.[3, 4] Nevertheless, naturally obtained helical peptides showed the innate limitations such as unsatisfactory packing with polyanions, low delivery effectiveness, and vulnerability to enzymatic degradation and temperature.[1, 3] In an effort to be applied biologically, further modification of side chains in the polypeptide effectively enables the elicitation of a bioavailable helical structure that is mainly dependent on three key factors: the length of hydrophobic chain, the bulkiness of side chains, and the surface charge density.[1, 3, 5-7] The water-affinitive helical polypeptide with proper surface charge density is rendered thermodynamically stable at physiological conditions.[1, 3, 5, 8-10] Based on previous studies, the artificial helical polypeptide (AHP) has successfully mimicked the natural helical conformation, and initiated a new platform for drug and gene delivery systems.[1, 3, 5, 8-10] AHP has been recently utilized as an effective non-viral vector for therapeutic purposes.[3, 5, 8-10] Generally, AHP, a gene-condensing material, was packed with therapeutic gene and then the complex is passively delivered to the diseased sites, demonstrating the superior therapeutic efficacy in a comparison to the previously reported polycation.[3, 5, 9, 10] The driving force of the high delivery efficiency of AHP is associated with its similar behavior as a natural cell-penetrating peptide, where the helical conformation is allowed to facilitate cell penetration into the membrane by interacting with plasma phospholipids.[3, 11-14] However, its delivery system was not well-customized to all the diseases due to the innate limitation of cell-penetrating characteristics, including nonspecific interactions, even with normal cellular plasma membranes.[11-14] The non-specific targeting strategy has caused unexpected sideeffects in drug and gene delivery systems.[15, 16] Therefore, it is imperative for targeting delivery systems to endow desirable properties only at the diseased region.[16] Comparing normal cells, a specific stumli-responsive characteristic should be introduced in the delivery systems to possess high selectivity to cancer. In contrast to the normal environments, the tumor extracellular conditions were relatively acidified due to high lactate concentrations, which might be the novel stimulus for selective delivery to cancers.[17, 18]
Herein, we suggested that pH-controllable cell-penetrating polypeptides (PCCP) regulating pH-adaptable helicity attained cells for specific cancer guidance (Figure 1). We hypothesized that PCCP was capable of undergoing pH-dependent conformational transition, thereby revealing the cell penetrating characteristic, specifically at the target site (Figure 1b). PCCP contained low helical propensity at a physiological pH through the electrostatic attractions between carboxylate and protonated amine groups in each side chain (Figure 1a). Thus, indiscriminate cellular internalization could be expeditiously circumvented by a hiding effect. In contrast, with an acidic environment, the pH-inactivated motif rapidly converted to an intact helical structure whereby the electrostatic interactions between repulsions and attractions were well-balanced throughout the side chains at pH 5~6 (Figure 1). Below pH 4, the electrostatic repulsions were so dominant that the back bone can be more elongated, bringing about a rapid reduction in helicity (Figure 1a). Consequently, the conformation of PCCP could change to a helix that possessed a superior cell-penetrating property, but only at cancer sites. In this study, we synthesized several poly-L-lysine (PLL)-based PCCPs two optimize pHactivated cell-penetrating capability exclusively in cancer cells. The primary amine moieties in PLL provided the further reaction because of the nucleophilicity. Using benzoylation, poly-4-bromobenzoyl-L-lysine (PBL) which was vulnerable to Pd-catalyzed reactions, was successfully synthesized. To determine the protein secondary structure affected by the bulkiness and the additional charge in the side chains, 4-imidazoleacrylic acid and acrylic acid were used as a building block to synthesize poly(4-(2-(imidazole-4-yl)acrylic acid)benzoyl-co-4-(4-(imidazole-2-yl)benzoyl-L-lysine) and poly(4-(3-acrylic acid)benzoylL-lysine) (PABL). Based on the synthesis of potential PCCPs. the optimized PCCPs showed two-fold increase of helicity when pH level was reduced to pH 6 (tumor extracellular pH). This design principle was grafted onto cellular conditions to strongly confirm pH-dependent cell penetration. The optimized PCCP exhibited outstanding selectivity only against cancer at low pH, driven by specific cell-penetration.
2. Materials and methods 2.1. PLL synthesis The PLL synthetic method was utilized as described in previous studies.[19-21]
2.2 PBL synthesis In a glove box, PLL (1.00 g) was solubilized in anhydrous N,N’-dimethylformamide (DMF) (10 mL) with triethylamine (TEA) (2.7 mL, 19.3 mmol), and then 4-bromobenzoyl chloride (3.40 g, 15.48 mmol) was slowly added to the solution. The reaction was allowed to occur at RT for 24 hr. A yellowish suspension was precipitated with deionized water used to remove the salt that formed, and washed with water twice. To remove the unreacted agent, the crude product was washed with 1 M NaOH solution three times, and then lypophilized. From this procedure, PBL (2.00 g) was obtained. 2.3. poly(4-(2-(imidazole-4-yl)acrylic acid)benzoyl-co-4-(4-(imidazole-2-yl)benzoyl-Llysine) (PIABL) using Heck reaction The PBL (1.00 g) that dissolved in DMF (10 mL) was poured into a DMF mixture with triphenylphosphine (16.91 mg, 64.5 µmol), Pd(OAc)2 (7.24 mg, 32.4 µmol), 4imidazoleacrylic acid (1.78 g, 12.90 mmol), and TEA (0.90 mL, 6.45 mmol). The reaction mixture was stirred at 80°C overnight, and then precipitated with diethyl ether. The crude product was isolated by centrifugation, and washed with diethyl ether three times. To remove the unreacted urocanic acid, the crude mixture was washed with deionized (DI) water five times, and then lypophilized. From this procedure, PIABL (1.45 g) was obtained. 2.4. Synthesis of poly(4-(3-acrylic acid)benzoyl-L-lysine) (PABL) The reaction was followed by a PIABL synthetic procedure. 2.5. Synthesis of 1-(2-aminoethyl)piperidine-conjugated PIABL (PIABL1), 4-(2aminoethyl)morpholine-conjugated PIABL (PIABL2), 1-(2-aminoethyl)piperazineconjugated PIABL (PIABL3), PIABL (0.20 g) was dissolved into DMF (10 mL) with TEA (0.165 mL, 1.19 mmol). Each reagent, 1-(2-aminoethyl)piperidine (0.51 mL, 3.56 mmol), 4-(2-aminoethyl)morpholine (0.46 mL, 3.56 mmol), and 1-(2-aminoethyl)piperazine (0.47 mL, 3.56 mmol), was independently added into the DMF solution. The reaction mixture was stirred at 80°C for 48 hr. Each polymer was isolated by a precipitation method using DI water, washed with DI water five times, and lypophilized. From this procedure, PIABL1 (0.23 g), PIABL2 (0.21 g), and PIABL3 (0.25 g) were obtained. 2.6. Synthesis of 1-(2-aminoethyl)piperidine-conjugated PABL (PABL1), 4-(2-
aminoethyl)morpholine-conjugated
PABL
(PABL2),
and
1-(2-
aminoethyl)piperazine-conjugated PABL (PABL3). The reaction was carried out by using the same reaction described in the synthesis of the PIABL series. 2.7. Estimation of physical properties Theoretical calculations of the molecular properties were calculated by using Marvin and JChem calculator plugins. 2.8. Demonstration of pH-dependent conformational conversion CD spectrometer (J-815 spectropolarimeter 150-L type, JASCO, Japan) was used to characterize a secondary protein structure with 0.01 cm path length of a quartz cell ranging from 190 nm to 260 nm at 20°C and 37°C. Prior to the measurement, all the polypeptides (100 mg) were protonated with 0.1 N HCl solution (10 mL), and then sonicated for one second. The protonated polypeptides were dialyzed against DI water, and then lypophilized. All the polypeptide samples were independently dissolved into different pH solutions (pH 4, pH 5, pH 6, and pH 7.4, respectively) and the concentrations were individually adjusted to 1 mg/mL. The corresponding non-buffer pH solutions were tuned by the diluted HCl and NaOH solutions. The helicity of the polypeptides was digitized by the following equation: helicity (%)= (-[θ]222nm + 3000)/39000) x 100. [22] 2.9. Intracellular delivery using calcein Prior to performing the cellular uptake study, three different types of calcein-loaded nanoparticles were prepared by a sonication method. Briefly, each polypeptide (50 mg) was solubilized in a phosphate buffer saline (PBS) solution (9 mL), and then a cacelin solution (5 mg/mL in PBS solution) was added into each polypeptide solution during sonication (5 sec, 20% amplitute, 600 W tip sonicator). The diameter and zeta-potential values were determined by dynamic light scattering and electrophoretic methods in supporting information. Both cell lines, NCI-H460 and HFL1, were seeded on 24-well plates at 70000 cells/well using Dulbecco's Modified Eagle's Medium (DMEM) (100 µL), as well as 10% heat-inactivated fetal bovine serum (FBS) and 1% antibiotic antimycotic solution, and incubated at 37°C in a 5% CO2 atmosphere. After 1 day of incubation, the medium was replaced with 4% serum DMEM (pH 7.4 and pH 6) containing polypeptide (10 µg/mL) and
calcein (1 µg/mL). After 4 hr’ incubation, cells were washed with the PBS solution three times, and then each well was treated with 0.2 (w/v) % trypan blue PBS solution to quench the extracellular fluorescence. Thereafter, the cells were rewashed with the PBS solution two times, and then trypsinized. The isolated cells were fixed by a 4% paraformaldehyde PBS solution. A flow cytometer (Becton Dickinson and Company, Cytek FACSCalibur, USA) was employed to conduct quantitative measurements.[23, 24] 2.10. Cell-penetrating mechanism To demonstrate the cell-penetrating property of PCCPs, both cell lines, NCI-H460 and HFL1, were seeded on 24-well plates at 70000 cells/well and cultured for 24 hr. The cells were preincubated with chloropromazine (10 µg/mL) as an endocytosis inhibitor for 30 min before being treated with 4% serum DMEM, including polypeptides (10 µg/mL) and calcein (1 µg/mL). The cells were then incubated at 37°C for 4 hr. To block the energy-dependent cellular internalization, the cells were incubated at 4°C during a 4 hr uptake period.[3] The measurement of uptake efficiency was performed according to the procedure mentioned above. 2.11. Confocal laser scanning microscopy (CLSM) measurement Both cells, NCI-H460 and HFL1, were seeded on 24-well plates at 70000 cells/well and incubated for 24 hr. The medium was replaced with reduced serum media containing polypeptides (50 µg/mL) and calcein (5 µg/mL), and further incubated for 4 hr. Thereafter, each well was rinsed with PBS solution three times, and 0.2 (w/v) % trypan blue PBS solution was added to each well to quench the extracellular fluorescence. After 5 min’ incubation, trypan blue was removed, and then the cells were washed with the PBS solution three times. Fixation was carried out using a 4% paraformaldehyde solution prior to stain the nuclei with DAPI (2 µg/mL, Invitogen, USA). Cells were observed by a confocal laser scanning microscope (LSM 510 META, ZEISS, Germany). The visualization of the cellpenetrating mechanism followed the procedure described above except that the cells were preincubated with the reduced serum media (4% FBS) including Lysotracker deep red (100 nM, Invitrogen, USA) for 30 min prior to the addition of each sample. NCI-H460 cells seeded on 24-well plates at 70000 cells were pretreated with reduced serum media (4% FBS) containing PCCPs (50 µg/mL) for 1 hr. Each well was then rinsed with PBS three times, and then replaced with fresh reduced serum media including calcein (1 µg/mL). After 2 hr’
incubation, cells were washed with the PBS solution three times and then 0.2 (w/v) % trypan blue solution was added. Thereafter, excess trypan blue was removed and the cells were rinsed with the PBS solution three times. The cells were fixed by a 4% paraformaldehyde solution. Uptake levels were determined by flow cytometry. 2.12. Biocompatibility confirmation To demonstrate the biocompatibility of polypeptides, a methyl thiazolyl tetrazolium (MTT) assay was employed against NCI-H460 (human lung carinoma) and HFL1 (hmuan normal lung) cell lines. Both cell lines were seeded on 96-well plates at 10000 cells and then cultured for 1 day at 37°C. The old medium was removed, washed with PBS solution once, and then both pH 7.4 and pH 6 reduced serum media (4% FBS) containing PCCP solutions at final concentrations of 200, 150, 100, 50, 20, and 10 µg/mL, respectively, were added. Thereafter, both polypeptides-treated cell lines were incubated for 1 day. MTT solution (20 µL, 5 mg/mL in PBS solution) was added to each well, and the samples were incubated further for 4 hr. To solubilize the formazine that formed in the cells, each well was treated with dimethylsulfoxide (DMSO) (100 µL) after the old medium was completly removed. The UV-Vis abosrobance of each well was detected by an UV-Vis spectrometeter (MultiSkanTM Go, Thermo Scientific, USA) at 490 nm. The cell viability was expressed as the ratio of Asample to Acontrol. 2.13. Analysis of membrane destabilization NCI-H460 cells were seeded on 24-well plates at 70000 cells/well, and incubated for 24 hr. The cells were then treated with 4% FBS-containing media including 20 µg/mL of PLL, PIABL3 and PABL3 after the removal of old media. After 1 hr of incubation, the cells were washed with PBS solution three times and calcein solution (2 µg/mL in reduced serum media) was added to each well. Followed by a further 3 hr incubation period, the cells were rinsed with PBS three times and then 0.2 (w/v) % trypan blue solution was added. Before trypsinization of cells, excess trypan blue was removed by PBS washing performed three times. The cells were fixed by a 4% paraformaldehyde solution. The uptake levels were evaluated by flow cytometry. Statistical analysis Statistical analysis was performed for an in vivo study using one-way analysis of variance (ANOVA). A p < 0.05 was considered to be statistically significant.
3. Results 3.1. Synthesis and characterization of PLL, PIABL and PABL series. Several stepwise reactions were employed for the synthetic polypeptide to endow its pHdependent cell-penetrating characteristic in the polypeptide undergoing pH-dependent conformational transitions (Figure S1-2). In this study, PLL was selected as a fundamental material due to ease of further functionalization and the relative long alkyl chain which contributes to helix formation.[1, 6] The PLL synthesized by a hexamethyldisilazanemediated living polymerization method [20] had 97 lysine mers and an almost monodisperse molecular weight distribution (Table 1). To functionalize the primary amine groups contained in the side chain of PLL, the amine moieties were completely benzoylated by 4-bromobenzoyl chloride, which can provide further reactions owing to the presence of aryl bromide selectively reacting with an α,βunsaturated carbonyl compound under a Pd catalyst. The NMR peaks appeared at 7-8 ppm, which indicated the complete conjugation of benzene moieties to the side chains of PLL (Figure S3). To compare the helical propensity affected by introducing a bulky group, both building blocks (acrylic acid and 4-imidazoleacrylic acid) were employed for the palladiumcatalyzed reaction to be able to intactly maintain Michael reaction’s position that is vulnerable to nucleophiles. The two peaks shown in Figure S3-5 occurred at 6-7 ppm, which illustrates that the α,β-unsaturated carbonyl configuration of both the acrylic acid and urocanic acid was completely preserved. Based on the NMR spectra, both intermediate polypeptides were successfully coupled to the para position of the benzene ring included in the side segment and no undesirable reaction occurred (Figures S3, S5). Lastly, to optimize the pH-dependent conformation-transformable polypeptide, three different PCCPs were synthesized and exploited three analogues which were piperdine, morpholine, and piperazine derivatives (Figure S4-5). Three different primary amine-bearing reagents were conjugated to the Michael reaction’s position in PIABL and PABL. The degree of modifications with the corresponding pendants was almost successful because the two peaks in Figures S4-5 at 6-7 ppm disappeared (Table S1). The detailed synthetic procedures were described in the methods section. The physical properties of PCCPs were predicted by using Marvin and JChem calculator
plugins (Figure 2). The amine moieties (primary, secondary and tertiary amine) were protonated at physiological conditions due to somewhat high basicity (Figure 2a). However, the carboxylate groups possessed pH-responsive characteristic at physiological pH levels because their pKa values (pKa 3~3.8) belonged to the equilibrium pH (Figure 2a). Therefore, the carboxylate groups were likely to contribute to pH-dependent conformational transitions in that the pH-sensitive carboxylate groups controlled electrostatic interactions within the side chains. Furthermore, all the log D values of PCCPs were estimated to determine watersolubility at the ambient pH levels. In regardless of pH levels, all the log D values of PCCPs were lower than log D 0, which indicated that PCCPs were highly water-soluble at the ambient pH levels (Figure 2b). 3.2. Determination of pH-dependent conformational transitions. Circular dichroism (CD) spectrometry was used to demonstrate the protein secondary structure and pH-dependent conformational transition. To determine the conformation affected by a bulky group at the side chain, Heck reaction of poly(4-bromobenzoyl-L-lysine) (PBL) was conducted using two α,β-unsaturated carbonyl reagents, which differentiated the attachment of the imidazole ring (Figure 2a). The plots of the PIABL series had a molar residue ellipticity minimum at 227 nm and a point of inflection at 217 nm, which displayed a similar tendency of β-sheet conformation in that sterically hindered imidazole rings disturbed spiral folding at pH 7.4[25, 26] (Figures 3a). On the other hand, both ellipitcity minima at 208 nm and 225 nm were observed in the PABL series, indicating that PABL series in the absence of a sterically hindered pendant exhibited a nearly helical formation[27] (Figures 3b). Based on the results, the presence of the bulky groups at the side chain had an adverse influence on the helix formation. To confirm the pH-triggered conformational transition influenced by analogues at the end of the side chain, three pendants having different physical properties were conjugated to PIABL and PABL at the outskirt of the side chain (Figure 2a). With decreasing pH conditions, both the minimum molar residue ellipticities of PIABL3 and PABL3 were deeper than those of other PIABL and PABL series, meaning that a piperazine group contributed to pH-sensitive conformational change than other pendants (Figure 3a-b). As a result, PIABL3 and PABL3 possessed the highest fluctuation with pH variation, and were used as optimal PCCPs for further study. To verify the pH-dependent conformational transition of PLL, PIABL3, and PABL3,
PCCPs were treated with four different levels of pH (Figure 3a-c). With the solution acidified, the minimum value of PIABL3 at 227 nm was significantly lowered. The pH-triggered decline of the minimum value at 227 nm indicated that the amount of β-sheet contents decreased, synchronizing a helicity increase. PIABL3 was capable of undergoing a pHinduced β-sheet to a helix conformational transition because both molar residue elliptcity values at 208 and 222 nm were lower with decreasing pH levels.[28] Therefore, helical contents were higher while beta sheet contents were lower. [25, 26] (Figure 3a). In contrast, in the absence of a sterically hindered group, PABL3 followed a different trend. Two minima molar residue ellipticities (208 nm and 225 nm) were moderately negative at pH 7.4, implying that the helical conformation was suppressed [27, 29] (Figure 3b). The two minima steeply declined at pH 5 and 6, suggesting that the intact helical structure was exhibited (Figure 3b).[29] Nevertheless, the helical formation was disrupted at pH 4 (Figure 3b). The PLL did not undergo a pH-dependent conformational change (Figure 3c). As shown in Figures S6-7 (37°C), the graph tendencies of PIABL3, PABL3, and PLL resembled those of the three polypeptides at 20°C, indicating that a pH-triggered conformational change could be elicited, even at physiological temperatures. To quantify the helical contents at each pH condition, all the -[θ]222
nm
(Molar residue
elliptcity at 222 nm) values were plotted against four different pH levels [1] and then the helicity was calculated by the equation (Table 1) because the equation of helical contents was derived from the molar residue elliptcity at 222 nm.[1, 30] The -[θ]222 nm values of PLL were kept consistent at ambient pH conditions and PLL had approximately 40% helicity (Figure 3c & Table 1). In contrast, it was found that the -[θ]222 nm values of two PCCPs, PIABL3 and PABL3 were fluctuated relatively depending on the pH environments (Figure 3d). In the case of the PIABL3 tendency, its -[θ]222 nm value increased more until pH 5, and then lowered at pH 4. The helical propensity of pH 5 (65.2% helicity) was two times higher than that of pH 7.4 (29.2%) while its helicity in pH 5 at 37°C (45.4%) was somewhat reduced (Figure 3d & Table 1). With regard to PABL3, its -[θ]222
nm
value soared from pH 7.4 to 5, and then
dramatically fell at pH 4 (Figure 3d). Its helical contents in pH 5 (92.8%) and pH 6 (78.3%) were almost three times higher and two times higher, respectively, than that of pH 7.4 (31.3%) (Table 1). The helicity of PABL3 at 37°C was almost similar to that at 20°C (Figure 3e & Table 1). 3.3. Cellular uptake study.
Based on CD spectra (Figure 3), the conformational transition of PLL did not occur with pH variation. However, the other polypeptides were allowed to transform their own conformation responding to proton concentration; non-transition in PLL, beta-sheet to helix transition in PIABL3 and dramatic helicity augmentation in PABL3. Prior to the cellular uptake study, nanoparticles were formed by electrostatic interactions to deliver fluorescent molecules. All the diameters and zeta potential values were shown in Figure S8. Flow cytometry and fluorescence uptake studies were performed to compare the internalization effectiveness influenced by different pH-dependent conformational transitions (Figure 4a, S9). In this study, NCI-H460 (human epithelial lung cancer cells) and HFL1 (normal lung fibroblast cells) were used to clearly demonstrate specific targetability, but only against carcinoma cells. Both cell lines were treated with calcein-loaded PLL, PIABL3, and PABL3, and were provided with sufficient uptake time (4 hr) to internalize the polypeptides. Compared to the uptake efficacy of PABL3 between pH 7.4 and pH 6, the uptake level of PABL3 at pH 6 (~ 60%) was two-fold higher than that at neutral pH (< 30%) (Figure 4a). On the other hand, it was determined that the rare amount of calcein molecules was retained in HFL1 cells at pH 7.4 (< 5%) (Figure 4a). Furthermore, even at pH 6 in HFL1 cells, PABL3 did not achieve considerable accumulation of calcein molecules in mostly HFL1 in comparison to NCI-H460 at pH 6 (~ 20%) (Figure 4a). Considering the general extracellular environments of normal (pH 7.4) and tumor sites (pH ~6), the appreciable difference in uptake level between HFL1 at pH 7.4 and NCI-H460 at pH 6 was observed (Figure 4b). The uptake efficiency of NCI-H460 at pH 6 was even higher than that of HFL1 at pH 7.4, which elucidated that PABL3 promoted pH-activated cellular internalization and provides tremendous selectivity solely within tumor cells (Figure 4b). In contrast to PABL3, PLL, prevalently used as delivering polycation, possessed insufficient uptake capability against carcinoma cells (Figure 4a). Furthermore, indistinguishable cellular internalization caused relatively augmented fluorescence uptake against normal cells, regardless of pH variation, which indicated that non-conformational transition was incapable of eliciting selective cancer targetability (Figure 4b). In comparison to PIABL3, PLL’s delivery efficiency was particularly low, irrespective of the types of cell lines and pH values, illustrating that pHdependent beta sheet-to-helix transition did not nearly impact specific cellular uptake (Figure 4a-b). CLSM was employed to visualize the degree of calcein uptake into both NCI-H460 and HFL1 cells under different pH conditions (Figure 4c). As far as the CLSM images of PABL3
were concerned, the green fluorescence of PABL3 at pH 6 was more intense than that at physiological pH (Figure 4c). Moreover, the fluorescence signal of PABL3 was punctuated at pH 7.4 while that was highly diffused within the cytosol at pH 6, which verified that PABL3 followed different cellular internalization mechanism via pH-sensitivity (Figure 4c). In contrast to the behavior in NCI-H460 cells, the green fluorescence was diminished in the HFL1 cell line regardless of pH levels (Figure 4c). It was confirmed that PABL3 possessed pH-activated cell penetrating capability and unusual selectivity of carcinoma cells. The appreciable difference of PLL and PIABL3 between pH 7.4 and pH 6 was observed regardless of cell lines (Figure 4c). Weak green fluorescence was exhibited in the cytosol of both cell lines, indicating that it was difficult for the behavior of conformational transition, non-transition, and β-sheet to helix transition to provide high delivery effectiveness into the cytoplasm, and also specificity (Figure 4c). 3.4. Cell-penetrating mechanism. To reveal the cell penetrating mechanism, a fluorescent uptake study was carried out using flow cytometry against NCI-H460 after the energy-dependent endocytosis pathway was completely inhibited by pretreatment of chloropromazine and incubation at 4°C (Figure 4d-e, S10). As shown in Figure 4d, the fluorescent internalization levels of all the polypeptides under endocytosis-blocking conditions were dramatically inferior. In addition, it was found that the cell penetrating property of PABL3 was capable of remaining inactive at pH 7.4, which might be the driving force of the pH-dependent cell-penetrating characteristic (Figure 4d). On the other hand, the uptake efficacy of PABL3 at pH 6 with chloropromazine treatment was comparable to that with 37°C incubation, which verified that PABL3 was not internalized via clathrin-mediated endocytosis (Figure 4e). Granted that the cellular uptake of PLL and PABL3 was nearly blocked regardless of the pH levels and cell lines at 4°C (inhibition of total energy-dependent internalization), the cellular internalization was somewhat maintained (~30%) at pH 6 (Figure 4d-e). This suggested that energy-independent cell-penetration was dominant over the other energy-required endocytosis pathways during cellular internalization. Liposome leakage and fluorescent uptake levels pretreated by PCCPs were assessed to confirm cellular membrane destabilization driven by the direct penetration (Figure 4f, S11). As shown in Figure S11, it was verified for PABL3 that liposomal membranes was immensely disrupted only at pH 6. On the other hand, it was shown that PLL and PIABL3 did not followed pH-dependent membrane disruption (Figure S11). To scrutinize
the membrane destabilizing activity in cellular conditions, the calcein uptake levels of PCCPs were determined by the pretreatment of PCCPs (Figure 4f). The uptake efficacy of PABL3 at pH 6 was almost two-fold incremental than that at pH 7.4 (Figure 4f). However, PLL and PIABL3 did not show pH-dependent membrane destabilization although they possessed a destabilizing property (Figure 4f). Based on the results of cellular internalization mechanism, it can be concluded that pH-controllable helicity provided specific cell-penetrating capability exclusively at low pH. CLSM was performed to strongly verify our hypothesis regarding direct penetration driven by cationic helix (Figure 5a-d). In figure 5d, the green fluorescence of PABL3 at pH 7.4 was considerably overlapped with red one (lysosome) while the green signal at pH 6 was rarely merged with the red, which indicated that PABL3 followed the behavior of direct penetration exclusively at pH 6 (Figure 5d). However, PLL followed endocytic pathways with low delivery efficacy because of overlapping green and red fluorescence regardless of pH levels (Figure 5b). Likewise PLL, it was found for PIABL3 that green fluorescence was overlapped with red fluorescence, meaning that PIABL3 was internalized into the cells via endocytosis pathway (Figure 5c). Therefore, pH-induced low-to-high helicity transition could endow selective cellular penetration at tumor pH. Cell viability was quantified using methyl thiazolyl tetrazolium (MTT) assay against NCI-H460 and HFL1 at pH 7.4 and pH 6 to monitor the biocompatibility of polypeptides. (Figure 5e) Cells were treated with various polypeptide concentrations for 1 day. Both PIABL3 and PABL3 showed outstanding cytocompatibility (>80% relative cell viability) against two cell lines at the ambient polypeptide concentrations (up to 200 µg/mL), indicating that the polypeptides were notably biocompatible as a delivery agent (Figure 5e). However, PLL showed decreased cell viability in both cell lines at pH 7.4 and pH 6, suggesting that it remained severely detrimental to cellular growth even at low concentration, and had inapplicability for delivering cargoes (Figure 5e).
4. Discussion Cationic helical peptides (CHP) typically possess a cell-penetrating characteristic driven by the unique ordered structure [11, 31]. For this reason, they have been intensively investigated as a promising delivery carrier in drug or gene delivery systems [32-35]. In these applications,
the use of CHPs has shown critical problems with selective arrival to the target site. More specifically, CHP nanocomplexes used for cancer targeting systems have almost been dependent on enhanced permeation and retention effects pending their arrival at the tumor sites, which have caused unexpected interactions with the undesirable regions [36, 37]. For cancer targeting systems, it is imperative that the targeting capability be endowed exclusively at tumor sites, thereby circumventing non-specific binding events [36, 38, 39].
In the present work, we developed novel PCCP systems that were capable of achieving specific cellular penetration solely at cancer cells. We speculated that the helical propensity of PCCPs were modulated by the pH-induced conversion of electrostatic interactions within the side chains, thereby affecting the unusual selective penetration. To demonstrate the specific cellular penetration influenced by the structural factors, PCCPs undergoing different conformational transitions were successfully synthesized by differentiating their chemical structures. The PLL did not undergo any conformational transitions while the PIABL3 and PABL3 demonstrated different behavior, a beta sheet-to-helix transition, and an increase in helicity, respectively, as evidenced by the CD spectra (Figure 6). We envisaged that the PCCPs undergoing different pH-induced conformational transitions could significantly influence the selective penetration. This assumption was supported by a cellular uptake study, which indicated that the unique characteristic of pH-controllable helicity in PABL3 allowed specific cell-penetration against carcinoma cells at pH 6, which resulted from the rapid augmented helical propensity. On the other hand, it was found that non-transition and beta sheet-to-low helical transitions were incapable of exploiting specific cellular uptake due to low helicity. Furthermore, we conducted pH-dependent cell-penetrating mechanism studies to underpin our hypothesis. It was confirmed that PABL3 followed direct penetration behavior and selectively penetrated the cell membrane in cancer cells (Figure 4). Additionally, we postulated that the pH-activated cationic helical conformation could destabilize the plasma membranes when it was transformed to intact helices at the cell membranes. In conclusion, we designed a pH-dependent conformation-transformable polypeptide endowing specific cell-penetrating characteristics only at tumor sites. The pH-controllable electrostatic interactions between the side chains resulted in adaptable helicity, thereby inducing specificity at the desirable target site. The flexible conformational transition systems can provide a new platform for a selective targeting strategy.
5. Acknowledgements This work was financially supported by the KAIST High Risk High Return project (Project No. N10150052) and Ministry of Science, ICT, and Future Planning (Project No. NRF2014M3A9E4064580, NRF-2016R1A2B4009619).We thank the Korea Research Institute of Bioscience and Biotechnology (KRIBB) for the confocal laser scanning microscopy and flow cytometry facilities.
6. References [1] H. Lu, J. Wang, Y.G. Bai, J.W. Lang, S.Y. Liu, Y. Lin, J.J. Cheng, Ionic polypeptides with unusual helical stability, Nat Commun 2 (2011). [2] N.E. Zhou, C.T. Mant, R.S. Hodges, Effect of preferred binding domains on peptide retention behavior in reversed-phase chromatography: amphipathic alpha-helices, Peptide research 3(1) (1990) 8-20. [3] H.Y. Tang, L.C. Yin, K.H. Kim, J.J. Cheng, Helical poly(arginine) mimics with superior cellpenetrating and molecular transporting properties, Chem Sci 4(10) (2013) 3839-3844. [4] M. Zorko, U. Langel, Cell-penetrating peptides: mechanism and kinetics of cargo delivery, Adv Drug Deliver Rev 57(4) (2005) 529-545. [5] N.P. Gabrielson, H. Lu, L.C. Yin, D. Li, F. Wang, J.J. Cheng, Reactive and Bioactive Cationic aHelical Polypeptide Template for Nonviral Gene Delivery, Angew Chem Int Edit 51(5) (2012) 11431147. [6] P.C. Lyu, M.I. Liff, L.A. Marky, N.R. Kallenbach, Side-Chain Contributions to the Stability of Alpha-Helical Structure in Peptides, Science 250(4981) (1990) 669-673. [7] P.C. Lyu, J.C. Sherman, A. Chen, N.R. Kallenbach, Alpha-Helix Stabilization by Natural and Unnatural Amino-Acids with Alkyl Side-Chains, P Natl Acad Sci USA 88(12) (1991) 5317-5320. [8] L.C. Yin, Z.Y. Song, Q.H. Qu, K.H. Kim, N. Zheng, C. Yao, I. Chaudhury, H.Y. Tang, N.P. Gabrielson, F.M. Uckun, J.J. Cheng, Supramolecular Self-Assembled Nanoparticles Mediate Oral Delivery of Therapeutic TNF-alpha siRNA against Systemic Inflammation, Angew Chem Int Edit 52(22) (2013) 5757-5761. [9] L.C. Yin, H.Y. Tang, K.H. Kim, N. Zheng, Z.Y. Song, N.P. Gabrielson, H. Lu, J.J. Cheng, LightResponsive Helical Polypeptides Capable of Reducing Toxicity and Unpacking DNA: Toward Nonviral Gene Delivery, Angew Chem Int Edit 52(35) (2013) 9182-9186. [10] N. Zheng, Z. Song, Y. Liu, R. Zhang, R. Zhang, C. Yao, F.M. Uckun, L. Yin, J. Cheng, Redox-
responsive, reversibly-crosslinked thiolated cationic helical polypeptides for efficient siRNA encapsulation and delivery, Journal of controlled release : official journal of the Controlled Release Society (2015). [11] K.A. Brogden, Antimicrobial peptides: Pore formers or metabolic inhibitors in bacteria?, Nat Rev Microbiol 3(3) (2005) 238-250. [12] B. Gupta, T.S. Levchenko, V.P. Torchilin, Intracellular delivery of large molecules and small particles by cell-penetrating proteins and peptides, Adv Drug Deliver Rev 57(4) (2005) 637-651. [13] J.E. Johnson, R.B. Cornell, Membrane-Binding Amphipathic Alpha-Helical Peptide Derived from Ctp-Phosphocholine Cytidylyltransferase, Biochemistry-Us 33(14) (1994) 4327-4335. [14] J.P. Richard, K. Melikov, E. Vives, C. Ramos, B. Verbeure, M.J. Gait, L.V. Chernomordik, B. Lebleu, Cell-penetrating peptides - A reevaluation of the mechanism of cellular uptake, J Biol Chem 278(1) (2003) 585-590. [15] D.W. Pack, A.S. Hoffman, S. Pun, P.S. Stayton, Design and development of polymers for gene delivery, Nat Rev Drug Discov 4(7) (2005) 581-593. [16] E. Fleige, M.A. Quadir, R. Haag, Stimuli-responsive polymeric nanocarriers for the controlled transport of active compounds: Concepts and applications, Adv Drug Deliver Rev 64(9) (2012) 866884. [17] J.Z. Du, T.M. Sun, W.J. Song, J. Wu, J. Wang, A Tumor-Acidity-Activated Charge-Conversional Nanogel as an Intelligent Vehicle for Promoted Tumoral-Cell Uptake and Drug Delivery, Angew Chem Int Edit 49(21) (2010) 3621-3626. [18] E.K. Rofstad, B. Mathiesen, K. Kindem, K. Galappathi, Acidic extracellular pH promotes experimental metastasis of human melanoma cells in athymic nude mice, Cancer Res 66(13) (2006) 6699-6707. [19] J.R. Hernandez, H.A. Klok, Synthesis and ring-opening (Co) polymerization of L-lysine Ncarboxyanhydrides containing labile side-chain protective groups, J Polym Sci Pol Chem 41(9) (2003) 1167-1187. [20] H. Lu, J. Cheng, Hexamethyldisilazane-mediated controlled polymerization of alpha-amino acid N-carboxyanhydrides, Journal of the American Chemical Society 129(46) (2007) 14114-5. [21] D. Lee, K. Choe, Y. Jeong, J. Yoo, S.M. Lee, J.H. Park, P. Kim, Y.C. Kim, Establishment of a controlled insulin delivery system using a glucose-responsive double-layered nanogel, Rsc Adv 5(19) (2015) 14482-14491. [22] K.S. Krannig, H. Schlaad, pH-Responsive Bioactive Glycopolypeptides with Enhanced Helicity and Solubility in Aqueous Solution, Journal of the American Chemical Society 134(45) (2012) 1854218545. [23] I. Noh, H.O. Kim, J. Choi, Y. Choi, D.K. Lee, Y.M. Huh, S. Haam, Co-delivery of paclitaxel and gemcitabine via CD44-targeting nanocarriers as a prodrug with synergistic antitumor activity against
human biliary cancer, Biomaterials 53 (2015) 763-774. [24] J. Yoo, D. Lee, V. Gujrati, N.S. Rejinold, K.M. Lekshmi, S. Uthaman, C. Jeong, I.K. Park, S. Jon, Y.C. Kim, Bioreducible branched poly(modified nona-arginine) cell-penetrating peptide as a novel gene delivery platform, J Control Release (2016). [25] B. Davidson, G.D. Fasman, The conformational transitions of uncharged poly-L-lysine. Alpha helix-random coil-beta structure, Biochemistry-Us 6(6) (1967) 1616-29. [26] I. Satake, J.T. Yang, Effect of chain length and concentration of anionic surfactants on the conformational transitions of poly(L-ornithine) and poly(L-lysine) in aqueous solution, Biochem Biophys Res Commun 54(3) (1973) 930-6. [27] K. Park, A. Perczel, G.D. Fasman, Differentiation between Transmembrane Helices and Peripheral Helices by the Deconvolution of Circular-Dichroism Spectra of Membrane-Proteins, Protein Sci 1(8) (1992) 1032-1049. [28] N.J. Greenfield, Using circular dichroism spectra to estimate protein secondary structure, Nat Protoc 1(6) (2006) 2876-2890. [29] E.G. Baker, G.J. Bartlett, M.P. Crump, R.B. Sessions, N. Linden, C.F.J. Faul, D.N. Woolfson, Local and macroscopic electrostatic interactions in single alpha-helices, Nat Chem Biol 11(3) (2015) 221-U92. [30] S. Padmanabhan, S. Marqusee, T. Ridgeway, T.M. Laue, R.L. Baldwin, Relative Helix-Forming Tendencies of Nonpolar Amino-Acids, Nature 344(6263) (1990) 268-270. [31] M. Zasloff, Antimicrobial peptides of multicellular organisms, Nature 415(6870) (2002) 389-395. [32] J. Farrera-Sinfreu, E. Giralt, S. Castel, F. Albericio, M. Royo, Cell-penetrating cis-gamma-aminoL-proline-derived peptides, J Am Chem Soc 127(26) (2005) 9459-9468. [33] I. Nakase, Y. Konishi, M. Ueda, H. Saji, S. Futaki, Accumulation of arginine-rich cell-penetrating peptides in tumors and the potential for anticancer drug delivery in vivo, J Control Release 159(2) (2012) 181-188. [34] H. Park, H. Tsutsumi, H. Mihara, Cell penetration and cell-selective drug delivery using alphahelix peptides conjugated with gold nanoparticles, Biomaterials 34(20) (2013) 4872-9. [35] K. Rittner, A. Benavente, A. Bompard-Sorlet, F. Heitz, G. Divita, R. Brasseur, E. Jacobs, New basic membrane-destabilizing peptides for plasmid-based gene delivery in vitro and in vivo, Mol Ther 5(2) (2002) 104-14. [36] R. Langer, Drug delivery and targeting, Nature 392(6679) (1998) 5-10. [37] R. Mo, Q. Sun, J.W. Xue, N. Li, W.Y. Li, C. Zhang, Q.N. Ping, Multistage pH-Responsive Liposomes for Mitochondrial-Targeted Anticancer Drug Delivery, Adv Mater 24(27) (2012) 36593665. [38] G. Mikhaylov, U. Mikac, A.A. Magaeva, V.I. Itin, E.P. Naiden, I. Psakhye, L. Babes, T. Reinheckel, C. Peters, R. Zeiser, M. Bogyo, V. Turk, S.G. Psakhye, B. Turk, O. Vasiljeva, Ferri-
liposomes as an MRI-visible drug-delivery system for targeting tumours and their microenvironment, Nat Nanotechnol 6(9) (2011) 594-602. [39] L. Rajendran, H.J. Knolker, K. Simons, Subcellular targeting strategies for drug design and delivery, Nat Rev Drug Discov 9(1) (2010) 29-42.
Figure and table captions Figure 1. Design of PCCPs, and proposed mechanism of pH-controllable helicity a nd selective cellular penetration. (a) Proposed mechanism of PCCP possessing a pHactivated cell penetrating property exclusively at the tumor extracellular matrix. 2° and 3° indicate “secondary” and “tertiary”, respectively. (b) Schematic illustration of the PCCP undergoing pH-dependent conformational transition induced by the charge balances of two opposite ions. Figure 2. Estimation of physical properties. (a) Prediction of pKa values of PCCPs. (b) Estimation of log D value with pH variation from pH 0 to pH 14. Theoretical calculations of the molecular properties were calculated by using Marvin and JChem calculator plugins. All the values were estimated by Marvin Sketch software. Lograthim distribution-coefficient (log D) is correlated to water solubility. (log D < 0, water-soluble) Figure 3. Determination of different conformational transitions using CD spectrometry. (a) CD spectra of PIABL series with different pH levels at 20°C. (b) CD spectra of PABL series with different pH levels at 20°C. (c) CD spectra of PLL with different pH levels at 20°C. Molar residue ellipticity at 222 nm plotted against different pH levels at (d) 20°C, and at (e) 37°C. Molar residue ellipticity = (mdeg x molecular weight)/(pathlength in millimeter x concentration). Figure 4. Calcein uptake and cell-penetrating mechanism studies at different pH levels. (a) The uptake levels of naked calcein, calcein-loaded PLL, PIABL3, and PABL3 using NCI and HFL1 cell lines at pH 7.4 and pH 6. (b) The uptake efficacy of all the controls against NCI-H460 at pH 6 and HFL1 at pH 7.4 to confirm the pH-depedent cell-penetrating characteristic. (c) CLSM images of naked calcein, calcein-loaded PLL, PIABL3, and PABL3 against NCI-H460 and HFL1 both at pH 7.4 and pH 6. 4',6-diamidino-2-phenylindole (DAPI): blue; Calcein: green; Scale bar: 25 µm) A cell-penetrating mechanism study was performed using NCI-H460 with endocytsis-blocking conditions, chloropromazine preincubation, and
4°C incubation, (d) at pH 7.4 and (e) at pH 6. (f) Uptake efficiency against NCI-H460 cells pretreated with none, PLL, PIABL3, and PABL3 prior to the addition of calcein as evaluated by flow cytometry. All the error bars are expressed as S.D. (n=4). Figure 5. Visualization of cell-penetrating mechanism and cytotoxicity study. Visualization of cellular uptake mechanism of (a) naked calcein, (b) PLL, (c) PIABL3 and (d) PABL3 against NCI-H460 cells at pH 7.4 and pH 6 by CLSM. (4',6-diamidino-2phenylindole (DAPI): blue; Lysotraker: red; Calcein: green; Scale bar: 50 µm) (e) Relative cell viability of polypeptides, PLL, PIABL3, and PABL3 at pH 7.4 and pH 6, toward NCIH460 and HFL1 cell lines after 1 day treatment, as evaluated by MTT assay. The error bars are expressed as S.D. (n=3). Figure 6. Schematic illustration of different conformational transitions. Non-transition in regardless of pH levels in PLL. Beta sheet-to-helix transition and low-to-high helix transition observed in PIABL3 and PABL3, respectively, with decreasing pH.
Table 1. Characterization of PLL, PIABL series and PABL series.
Table 1. Characterization of PLL, PIABL series and PABL series.
a
Helicity (%)d at pH 7.5 (20 / 37°C)
Helicity (%) at pH 6 (20 / 37°C)
Helicity (%) at pH 5 (20 / 37°C)
Helicity (%) at pH 4 (20 / 37°C)
~100 b
<1.1
39.5 / 44.6
48.3 / 46.3
42.0 / 43.9
37.1 / 44.9
97
~100b,c
<1.1
30.9 / 29.1
33.4 / 13.2
28.4 / 18.3
27.4 / 17.6
97
~100b,c
<1.1
40.9 / 30.2
43.8 / 27.6
31.3 / 19.4
26.3 / 21.0
PIABL3
97
~100
b,c
<1.1
55.8 / 45.1
67.4 / 45.4
33.5 / 31.3
29.2 / 29.6
PABL1
97
~100b,c
<1.1
44.1 / 40.1
44.2 / 44.6
80.0 / 62.7
73.1 / 54.8
PABL2
97
~100b,c
<1.1
28.5 / 17.7
62.5 / 44.3
59.7 / 45.5
48.9 / 43.9
PABL3
97
~100b,c
<1.1
29.6 / 23.5
78.3 / 68.4
92.8 / 90.6
31.3 / 23.5
Repeating unita (Lys)
Degree of modification (%)
PLL
97
PIABL1 PIABL2
Polypeptide Entry
a
PDI
The number of lysine units and polydiserpsity index (PDI) were determined by gel permeation chromatography (GPC) as a standard of PTFL. b Degree of modification was quantified by NMR spectrometry. c Degree of modification was calculated by elemental analysis. d Helicity was determined by the equation; helicity (%)= (-[θ]222 nm+3000)/39000 x 100
We developed pH-controllable cell-penetrating polypeptides (PCCPs) undergoing pHinduced conformational transitions. Unlike natural cell-penetrating peptides, PCCPs was capable of penetrating the plasma membranes dominantly at tumor pH, driven by pHcontrolled helicity. The conformation of PCCPs at neutral pH showed low helical propensity because of dominant electrostatic attractions within the side chains. However, the helicity of PCCPs was considerably augmented by the balance of electrostatic interactions, thereby inducing
selective
cellular
penetration.
Three
polypeptides
undergoing
different
conformational transitions were prepared to verify the selective cellular uptake influenced by their structures. The PCCP undergoing low-to-high helical conformation provided the tumor specificity and enhanced uptake efficiency. pH-induced conformation-transformable polypeptide might provide a novel platform for stimuli-triggered targeting systems.
*Graphical Abstract
S1742-7061(17)30331-8 http://dx.doi.org/10.1016/j.actbio.2017.05.040 ACTBIO 4899
To appear in:
Acta Biomaterialia
Received Date: Revised Date: Accepted Date:
17 February 2017 11 May 2017 16 May 2017
Please cite this article as: Lee, D., Noh, I., Yoo, J., Rejinold, N.Sanoj, Kim, Y-C., pH-controllable cell-penetrating polypeptide possessing cancer targeting, Acta Biomaterialia (2017), doi: http://dx.doi.org/10.1016/j.actbio. 2017.05.040
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
pH-controllable cell-penetrating polypeptide possessing cancer targeting DaeYong Lee, IlKoo Noh, Jisang Yoo, N.Sanoj Rejinold, Yeu-Chun Kim* Department of Chemical and Biomolecular engineering, Korea Advanced Institute of Science and Technology (KAIST), Daejeon 305-701, Republic of Korea. Corresponding Author: Prof. Yeu-Chun Kim, TEL: +82-42-350-3939, FAX: +82-48-3503910, E-mail: [email protected]
Abstract Helical peptides were naturally-occurring ordered conformations that mediated various biological functions essential for biotechnology. However, it was difficult for natural helical polypeptides to be applied in biomedical fields due to low bioavailability. To avoid these problems, synthetic alpha-helical polypeptides have recently been introduced by further modifying pendants in the side chain. In spite of an attractive biomimetic helical motif, these systems could not be tailored for targeted delivery mainly due to nonspecific binding events. To address these issues, we created a conformation-transformable polypeptide capable of eliciting a pH-activated cell-penetrating property solely at the cancer region. The developed novel polypeptide showed that the bare helical conformation had a function at physiological conditions while the pH-induced helical motif provided an active cell-penetrating characteristic at a tumor extracellular matrix pH. The unusual conformation-transformable system can elicit bioactive properties exclusively at mild acidic pH.
1. Introduction α-helical structures mediate protein folding or unfolding, and are the fundamental foundation for the protein assembly and bioactive functions.[1, 2] Among various α-helical peptides, cationic helical peptides (CHP) have been used in biomedical fields due to their unique cell-penetrating property. CHPs composed of sterically unhindered and lysine or arginine-rich amino acid residues strongly bind to the lipid plasma membranes via electrostatic attractions, and then distort the arrangement of lipid molecules, thereby being translocated in the cytosol.[3, 4] Nevertheless, naturally obtained helical peptides showed the innate limitations such as unsatisfactory packing with polyanions, low delivery effectiveness, and vulnerability to enzymatic degradation and temperature.[1, 3] In an effort to be applied biologically, further modification of side chains in the polypeptide effectively enables the elicitation of a bioavailable helical structure that is mainly dependent on three key factors: the length of hydrophobic chain, the bulkiness of side chains, and the surface charge density.[1, 3, 5-7] The water-affinitive helical polypeptide with proper surface charge density is rendered thermodynamically stable at physiological conditions.[1, 3, 5, 8-10] Based on previous studies, the artificial helical polypeptide (AHP) has successfully mimicked the natural helical conformation, and initiated a new platform for drug and gene delivery systems.[1, 3, 5, 8-10] AHP has been recently utilized as an effective non-viral vector for therapeutic purposes.[3, 5, 8-10] Generally, AHP, a gene-condensing material, was packed with therapeutic gene and then the complex is passively delivered to the diseased sites, demonstrating the superior therapeutic efficacy in a comparison to the previously reported polycation.[3, 5, 9, 10] The driving force of the high delivery efficiency of AHP is associated with its similar behavior as a natural cell-penetrating peptide, where the helical conformation is allowed to facilitate cell penetration into the membrane by interacting with plasma phospholipids.[3, 11-14] However, its delivery system was not well-customized to all the diseases due to the innate limitation of cell-penetrating characteristics, including nonspecific interactions, even with normal cellular plasma membranes.[11-14] The non-specific targeting strategy has caused unexpected sideeffects in drug and gene delivery systems.[15, 16] Therefore, it is imperative for targeting delivery systems to endow desirable properties only at the diseased region.[16] Comparing normal cells, a specific stumli-responsive characteristic should be introduced in the delivery systems to possess high selectivity to cancer. In contrast to the normal environments, the tumor extracellular conditions were relatively acidified due to high lactate concentrations, which might be the novel stimulus for selective delivery to cancers.[17, 18]
Herein, we suggested that pH-controllable cell-penetrating polypeptides (PCCP) regulating pH-adaptable helicity attained cells for specific cancer guidance (Figure 1). We hypothesized that PCCP was capable of undergoing pH-dependent conformational transition, thereby revealing the cell penetrating characteristic, specifically at the target site (Figure 1b). PCCP contained low helical propensity at a physiological pH through the electrostatic attractions between carboxylate and protonated amine groups in each side chain (Figure 1a). Thus, indiscriminate cellular internalization could be expeditiously circumvented by a hiding effect. In contrast, with an acidic environment, the pH-inactivated motif rapidly converted to an intact helical structure whereby the electrostatic interactions between repulsions and attractions were well-balanced throughout the side chains at pH 5~6 (Figure 1). Below pH 4, the electrostatic repulsions were so dominant that the back bone can be more elongated, bringing about a rapid reduction in helicity (Figure 1a). Consequently, the conformation of PCCP could change to a helix that possessed a superior cell-penetrating property, but only at cancer sites. In this study, we synthesized several poly-L-lysine (PLL)-based PCCPs two optimize pHactivated cell-penetrating capability exclusively in cancer cells. The primary amine moieties in PLL provided the further reaction because of the nucleophilicity. Using benzoylation, poly-4-bromobenzoyl-L-lysine (PBL) which was vulnerable to Pd-catalyzed reactions, was successfully synthesized. To determine the protein secondary structure affected by the bulkiness and the additional charge in the side chains, 4-imidazoleacrylic acid and acrylic acid were used as a building block to synthesize poly(4-(2-(imidazole-4-yl)acrylic acid)benzoyl-co-4-(4-(imidazole-2-yl)benzoyl-L-lysine) and poly(4-(3-acrylic acid)benzoylL-lysine) (PABL). Based on the synthesis of potential PCCPs. the optimized PCCPs showed two-fold increase of helicity when pH level was reduced to pH 6 (tumor extracellular pH). This design principle was grafted onto cellular conditions to strongly confirm pH-dependent cell penetration. The optimized PCCP exhibited outstanding selectivity only against cancer at low pH, driven by specific cell-penetration.
2. Materials and methods 2.1. PLL synthesis The PLL synthetic method was utilized as described in previous studies.[19-21]
2.2 PBL synthesis In a glove box, PLL (1.00 g) was solubilized in anhydrous N,N’-dimethylformamide (DMF) (10 mL) with triethylamine (TEA) (2.7 mL, 19.3 mmol), and then 4-bromobenzoyl chloride (3.40 g, 15.48 mmol) was slowly added to the solution. The reaction was allowed to occur at RT for 24 hr. A yellowish suspension was precipitated with deionized water used to remove the salt that formed, and washed with water twice. To remove the unreacted agent, the crude product was washed with 1 M NaOH solution three times, and then lypophilized. From this procedure, PBL (2.00 g) was obtained. 2.3. poly(4-(2-(imidazole-4-yl)acrylic acid)benzoyl-co-4-(4-(imidazole-2-yl)benzoyl-Llysine) (PIABL) using Heck reaction The PBL (1.00 g) that dissolved in DMF (10 mL) was poured into a DMF mixture with triphenylphosphine (16.91 mg, 64.5 µmol), Pd(OAc)2 (7.24 mg, 32.4 µmol), 4imidazoleacrylic acid (1.78 g, 12.90 mmol), and TEA (0.90 mL, 6.45 mmol). The reaction mixture was stirred at 80°C overnight, and then precipitated with diethyl ether. The crude product was isolated by centrifugation, and washed with diethyl ether three times. To remove the unreacted urocanic acid, the crude mixture was washed with deionized (DI) water five times, and then lypophilized. From this procedure, PIABL (1.45 g) was obtained. 2.4. Synthesis of poly(4-(3-acrylic acid)benzoyl-L-lysine) (PABL) The reaction was followed by a PIABL synthetic procedure. 2.5. Synthesis of 1-(2-aminoethyl)piperidine-conjugated PIABL (PIABL1), 4-(2aminoethyl)morpholine-conjugated PIABL (PIABL2), 1-(2-aminoethyl)piperazineconjugated PIABL (PIABL3), PIABL (0.20 g) was dissolved into DMF (10 mL) with TEA (0.165 mL, 1.19 mmol). Each reagent, 1-(2-aminoethyl)piperidine (0.51 mL, 3.56 mmol), 4-(2-aminoethyl)morpholine (0.46 mL, 3.56 mmol), and 1-(2-aminoethyl)piperazine (0.47 mL, 3.56 mmol), was independently added into the DMF solution. The reaction mixture was stirred at 80°C for 48 hr. Each polymer was isolated by a precipitation method using DI water, washed with DI water five times, and lypophilized. From this procedure, PIABL1 (0.23 g), PIABL2 (0.21 g), and PIABL3 (0.25 g) were obtained. 2.6. Synthesis of 1-(2-aminoethyl)piperidine-conjugated PABL (PABL1), 4-(2-
aminoethyl)morpholine-conjugated
PABL
(PABL2),
and
1-(2-
aminoethyl)piperazine-conjugated PABL (PABL3). The reaction was carried out by using the same reaction described in the synthesis of the PIABL series. 2.7. Estimation of physical properties Theoretical calculations of the molecular properties were calculated by using Marvin and JChem calculator plugins. 2.8. Demonstration of pH-dependent conformational conversion CD spectrometer (J-815 spectropolarimeter 150-L type, JASCO, Japan) was used to characterize a secondary protein structure with 0.01 cm path length of a quartz cell ranging from 190 nm to 260 nm at 20°C and 37°C. Prior to the measurement, all the polypeptides (100 mg) were protonated with 0.1 N HCl solution (10 mL), and then sonicated for one second. The protonated polypeptides were dialyzed against DI water, and then lypophilized. All the polypeptide samples were independently dissolved into different pH solutions (pH 4, pH 5, pH 6, and pH 7.4, respectively) and the concentrations were individually adjusted to 1 mg/mL. The corresponding non-buffer pH solutions were tuned by the diluted HCl and NaOH solutions. The helicity of the polypeptides was digitized by the following equation: helicity (%)= (-[θ]222nm + 3000)/39000) x 100. [22] 2.9. Intracellular delivery using calcein Prior to performing the cellular uptake study, three different types of calcein-loaded nanoparticles were prepared by a sonication method. Briefly, each polypeptide (50 mg) was solubilized in a phosphate buffer saline (PBS) solution (9 mL), and then a cacelin solution (5 mg/mL in PBS solution) was added into each polypeptide solution during sonication (5 sec, 20% amplitute, 600 W tip sonicator). The diameter and zeta-potential values were determined by dynamic light scattering and electrophoretic methods in supporting information. Both cell lines, NCI-H460 and HFL1, were seeded on 24-well plates at 70000 cells/well using Dulbecco's Modified Eagle's Medium (DMEM) (100 µL), as well as 10% heat-inactivated fetal bovine serum (FBS) and 1% antibiotic antimycotic solution, and incubated at 37°C in a 5% CO2 atmosphere. After 1 day of incubation, the medium was replaced with 4% serum DMEM (pH 7.4 and pH 6) containing polypeptide (10 µg/mL) and
calcein (1 µg/mL). After 4 hr’ incubation, cells were washed with the PBS solution three times, and then each well was treated with 0.2 (w/v) % trypan blue PBS solution to quench the extracellular fluorescence. Thereafter, the cells were rewashed with the PBS solution two times, and then trypsinized. The isolated cells were fixed by a 4% paraformaldehyde PBS solution. A flow cytometer (Becton Dickinson and Company, Cytek FACSCalibur, USA) was employed to conduct quantitative measurements.[23, 24] 2.10. Cell-penetrating mechanism To demonstrate the cell-penetrating property of PCCPs, both cell lines, NCI-H460 and HFL1, were seeded on 24-well plates at 70000 cells/well and cultured for 24 hr. The cells were preincubated with chloropromazine (10 µg/mL) as an endocytosis inhibitor for 30 min before being treated with 4% serum DMEM, including polypeptides (10 µg/mL) and calcein (1 µg/mL). The cells were then incubated at 37°C for 4 hr. To block the energy-dependent cellular internalization, the cells were incubated at 4°C during a 4 hr uptake period.[3] The measurement of uptake efficiency was performed according to the procedure mentioned above. 2.11. Confocal laser scanning microscopy (CLSM) measurement Both cells, NCI-H460 and HFL1, were seeded on 24-well plates at 70000 cells/well and incubated for 24 hr. The medium was replaced with reduced serum media containing polypeptides (50 µg/mL) and calcein (5 µg/mL), and further incubated for 4 hr. Thereafter, each well was rinsed with PBS solution three times, and 0.2 (w/v) % trypan blue PBS solution was added to each well to quench the extracellular fluorescence. After 5 min’ incubation, trypan blue was removed, and then the cells were washed with the PBS solution three times. Fixation was carried out using a 4% paraformaldehyde solution prior to stain the nuclei with DAPI (2 µg/mL, Invitogen, USA). Cells were observed by a confocal laser scanning microscope (LSM 510 META, ZEISS, Germany). The visualization of the cellpenetrating mechanism followed the procedure described above except that the cells were preincubated with the reduced serum media (4% FBS) including Lysotracker deep red (100 nM, Invitrogen, USA) for 30 min prior to the addition of each sample. NCI-H460 cells seeded on 24-well plates at 70000 cells were pretreated with reduced serum media (4% FBS) containing PCCPs (50 µg/mL) for 1 hr. Each well was then rinsed with PBS three times, and then replaced with fresh reduced serum media including calcein (1 µg/mL). After 2 hr’
incubation, cells were washed with the PBS solution three times and then 0.2 (w/v) % trypan blue solution was added. Thereafter, excess trypan blue was removed and the cells were rinsed with the PBS solution three times. The cells were fixed by a 4% paraformaldehyde solution. Uptake levels were determined by flow cytometry. 2.12. Biocompatibility confirmation To demonstrate the biocompatibility of polypeptides, a methyl thiazolyl tetrazolium (MTT) assay was employed against NCI-H460 (human lung carinoma) and HFL1 (hmuan normal lung) cell lines. Both cell lines were seeded on 96-well plates at 10000 cells and then cultured for 1 day at 37°C. The old medium was removed, washed with PBS solution once, and then both pH 7.4 and pH 6 reduced serum media (4% FBS) containing PCCP solutions at final concentrations of 200, 150, 100, 50, 20, and 10 µg/mL, respectively, were added. Thereafter, both polypeptides-treated cell lines were incubated for 1 day. MTT solution (20 µL, 5 mg/mL in PBS solution) was added to each well, and the samples were incubated further for 4 hr. To solubilize the formazine that formed in the cells, each well was treated with dimethylsulfoxide (DMSO) (100 µL) after the old medium was completly removed. The UV-Vis abosrobance of each well was detected by an UV-Vis spectrometeter (MultiSkanTM Go, Thermo Scientific, USA) at 490 nm. The cell viability was expressed as the ratio of Asample to Acontrol. 2.13. Analysis of membrane destabilization NCI-H460 cells were seeded on 24-well plates at 70000 cells/well, and incubated for 24 hr. The cells were then treated with 4% FBS-containing media including 20 µg/mL of PLL, PIABL3 and PABL3 after the removal of old media. After 1 hr of incubation, the cells were washed with PBS solution three times and calcein solution (2 µg/mL in reduced serum media) was added to each well. Followed by a further 3 hr incubation period, the cells were rinsed with PBS three times and then 0.2 (w/v) % trypan blue solution was added. Before trypsinization of cells, excess trypan blue was removed by PBS washing performed three times. The cells were fixed by a 4% paraformaldehyde solution. The uptake levels were evaluated by flow cytometry. Statistical analysis Statistical analysis was performed for an in vivo study using one-way analysis of variance (ANOVA). A p < 0.05 was considered to be statistically significant.
3. Results 3.1. Synthesis and characterization of PLL, PIABL and PABL series. Several stepwise reactions were employed for the synthetic polypeptide to endow its pHdependent cell-penetrating characteristic in the polypeptide undergoing pH-dependent conformational transitions (Figure S1-2). In this study, PLL was selected as a fundamental material due to ease of further functionalization and the relative long alkyl chain which contributes to helix formation.[1, 6] The PLL synthesized by a hexamethyldisilazanemediated living polymerization method [20] had 97 lysine mers and an almost monodisperse molecular weight distribution (Table 1). To functionalize the primary amine groups contained in the side chain of PLL, the amine moieties were completely benzoylated by 4-bromobenzoyl chloride, which can provide further reactions owing to the presence of aryl bromide selectively reacting with an α,βunsaturated carbonyl compound under a Pd catalyst. The NMR peaks appeared at 7-8 ppm, which indicated the complete conjugation of benzene moieties to the side chains of PLL (Figure S3). To compare the helical propensity affected by introducing a bulky group, both building blocks (acrylic acid and 4-imidazoleacrylic acid) were employed for the palladiumcatalyzed reaction to be able to intactly maintain Michael reaction’s position that is vulnerable to nucleophiles. The two peaks shown in Figure S3-5 occurred at 6-7 ppm, which illustrates that the α,β-unsaturated carbonyl configuration of both the acrylic acid and urocanic acid was completely preserved. Based on the NMR spectra, both intermediate polypeptides were successfully coupled to the para position of the benzene ring included in the side segment and no undesirable reaction occurred (Figures S3, S5). Lastly, to optimize the pH-dependent conformation-transformable polypeptide, three different PCCPs were synthesized and exploited three analogues which were piperdine, morpholine, and piperazine derivatives (Figure S4-5). Three different primary amine-bearing reagents were conjugated to the Michael reaction’s position in PIABL and PABL. The degree of modifications with the corresponding pendants was almost successful because the two peaks in Figures S4-5 at 6-7 ppm disappeared (Table S1). The detailed synthetic procedures were described in the methods section. The physical properties of PCCPs were predicted by using Marvin and JChem calculator
plugins (Figure 2). The amine moieties (primary, secondary and tertiary amine) were protonated at physiological conditions due to somewhat high basicity (Figure 2a). However, the carboxylate groups possessed pH-responsive characteristic at physiological pH levels because their pKa values (pKa 3~3.8) belonged to the equilibrium pH (Figure 2a). Therefore, the carboxylate groups were likely to contribute to pH-dependent conformational transitions in that the pH-sensitive carboxylate groups controlled electrostatic interactions within the side chains. Furthermore, all the log D values of PCCPs were estimated to determine watersolubility at the ambient pH levels. In regardless of pH levels, all the log D values of PCCPs were lower than log D 0, which indicated that PCCPs were highly water-soluble at the ambient pH levels (Figure 2b). 3.2. Determination of pH-dependent conformational transitions. Circular dichroism (CD) spectrometry was used to demonstrate the protein secondary structure and pH-dependent conformational transition. To determine the conformation affected by a bulky group at the side chain, Heck reaction of poly(4-bromobenzoyl-L-lysine) (PBL) was conducted using two α,β-unsaturated carbonyl reagents, which differentiated the attachment of the imidazole ring (Figure 2a). The plots of the PIABL series had a molar residue ellipticity minimum at 227 nm and a point of inflection at 217 nm, which displayed a similar tendency of β-sheet conformation in that sterically hindered imidazole rings disturbed spiral folding at pH 7.4[25, 26] (Figures 3a). On the other hand, both ellipitcity minima at 208 nm and 225 nm were observed in the PABL series, indicating that PABL series in the absence of a sterically hindered pendant exhibited a nearly helical formation[27] (Figures 3b). Based on the results, the presence of the bulky groups at the side chain had an adverse influence on the helix formation. To confirm the pH-triggered conformational transition influenced by analogues at the end of the side chain, three pendants having different physical properties were conjugated to PIABL and PABL at the outskirt of the side chain (Figure 2a). With decreasing pH conditions, both the minimum molar residue ellipticities of PIABL3 and PABL3 were deeper than those of other PIABL and PABL series, meaning that a piperazine group contributed to pH-sensitive conformational change than other pendants (Figure 3a-b). As a result, PIABL3 and PABL3 possessed the highest fluctuation with pH variation, and were used as optimal PCCPs for further study. To verify the pH-dependent conformational transition of PLL, PIABL3, and PABL3,
PCCPs were treated with four different levels of pH (Figure 3a-c). With the solution acidified, the minimum value of PIABL3 at 227 nm was significantly lowered. The pH-triggered decline of the minimum value at 227 nm indicated that the amount of β-sheet contents decreased, synchronizing a helicity increase. PIABL3 was capable of undergoing a pHinduced β-sheet to a helix conformational transition because both molar residue elliptcity values at 208 and 222 nm were lower with decreasing pH levels.[28] Therefore, helical contents were higher while beta sheet contents were lower. [25, 26] (Figure 3a). In contrast, in the absence of a sterically hindered group, PABL3 followed a different trend. Two minima molar residue ellipticities (208 nm and 225 nm) were moderately negative at pH 7.4, implying that the helical conformation was suppressed [27, 29] (Figure 3b). The two minima steeply declined at pH 5 and 6, suggesting that the intact helical structure was exhibited (Figure 3b).[29] Nevertheless, the helical formation was disrupted at pH 4 (Figure 3b). The PLL did not undergo a pH-dependent conformational change (Figure 3c). As shown in Figures S6-7 (37°C), the graph tendencies of PIABL3, PABL3, and PLL resembled those of the three polypeptides at 20°C, indicating that a pH-triggered conformational change could be elicited, even at physiological temperatures. To quantify the helical contents at each pH condition, all the -[θ]222
nm
(Molar residue
elliptcity at 222 nm) values were plotted against four different pH levels [1] and then the helicity was calculated by the equation (Table 1) because the equation of helical contents was derived from the molar residue elliptcity at 222 nm.[1, 30] The -[θ]222 nm values of PLL were kept consistent at ambient pH conditions and PLL had approximately 40% helicity (Figure 3c & Table 1). In contrast, it was found that the -[θ]222 nm values of two PCCPs, PIABL3 and PABL3 were fluctuated relatively depending on the pH environments (Figure 3d). In the case of the PIABL3 tendency, its -[θ]222 nm value increased more until pH 5, and then lowered at pH 4. The helical propensity of pH 5 (65.2% helicity) was two times higher than that of pH 7.4 (29.2%) while its helicity in pH 5 at 37°C (45.4%) was somewhat reduced (Figure 3d & Table 1). With regard to PABL3, its -[θ]222
nm
value soared from pH 7.4 to 5, and then
dramatically fell at pH 4 (Figure 3d). Its helical contents in pH 5 (92.8%) and pH 6 (78.3%) were almost three times higher and two times higher, respectively, than that of pH 7.4 (31.3%) (Table 1). The helicity of PABL3 at 37°C was almost similar to that at 20°C (Figure 3e & Table 1). 3.3. Cellular uptake study.
Based on CD spectra (Figure 3), the conformational transition of PLL did not occur with pH variation. However, the other polypeptides were allowed to transform their own conformation responding to proton concentration; non-transition in PLL, beta-sheet to helix transition in PIABL3 and dramatic helicity augmentation in PABL3. Prior to the cellular uptake study, nanoparticles were formed by electrostatic interactions to deliver fluorescent molecules. All the diameters and zeta potential values were shown in Figure S8. Flow cytometry and fluorescence uptake studies were performed to compare the internalization effectiveness influenced by different pH-dependent conformational transitions (Figure 4a, S9). In this study, NCI-H460 (human epithelial lung cancer cells) and HFL1 (normal lung fibroblast cells) were used to clearly demonstrate specific targetability, but only against carcinoma cells. Both cell lines were treated with calcein-loaded PLL, PIABL3, and PABL3, and were provided with sufficient uptake time (4 hr) to internalize the polypeptides. Compared to the uptake efficacy of PABL3 between pH 7.4 and pH 6, the uptake level of PABL3 at pH 6 (~ 60%) was two-fold higher than that at neutral pH (< 30%) (Figure 4a). On the other hand, it was determined that the rare amount of calcein molecules was retained in HFL1 cells at pH 7.4 (< 5%) (Figure 4a). Furthermore, even at pH 6 in HFL1 cells, PABL3 did not achieve considerable accumulation of calcein molecules in mostly HFL1 in comparison to NCI-H460 at pH 6 (~ 20%) (Figure 4a). Considering the general extracellular environments of normal (pH 7.4) and tumor sites (pH ~6), the appreciable difference in uptake level between HFL1 at pH 7.4 and NCI-H460 at pH 6 was observed (Figure 4b). The uptake efficiency of NCI-H460 at pH 6 was even higher than that of HFL1 at pH 7.4, which elucidated that PABL3 promoted pH-activated cellular internalization and provides tremendous selectivity solely within tumor cells (Figure 4b). In contrast to PABL3, PLL, prevalently used as delivering polycation, possessed insufficient uptake capability against carcinoma cells (Figure 4a). Furthermore, indistinguishable cellular internalization caused relatively augmented fluorescence uptake against normal cells, regardless of pH variation, which indicated that non-conformational transition was incapable of eliciting selective cancer targetability (Figure 4b). In comparison to PIABL3, PLL’s delivery efficiency was particularly low, irrespective of the types of cell lines and pH values, illustrating that pHdependent beta sheet-to-helix transition did not nearly impact specific cellular uptake (Figure 4a-b). CLSM was employed to visualize the degree of calcein uptake into both NCI-H460 and HFL1 cells under different pH conditions (Figure 4c). As far as the CLSM images of PABL3
were concerned, the green fluorescence of PABL3 at pH 6 was more intense than that at physiological pH (Figure 4c). Moreover, the fluorescence signal of PABL3 was punctuated at pH 7.4 while that was highly diffused within the cytosol at pH 6, which verified that PABL3 followed different cellular internalization mechanism via pH-sensitivity (Figure 4c). In contrast to the behavior in NCI-H460 cells, the green fluorescence was diminished in the HFL1 cell line regardless of pH levels (Figure 4c). It was confirmed that PABL3 possessed pH-activated cell penetrating capability and unusual selectivity of carcinoma cells. The appreciable difference of PLL and PIABL3 between pH 7.4 and pH 6 was observed regardless of cell lines (Figure 4c). Weak green fluorescence was exhibited in the cytosol of both cell lines, indicating that it was difficult for the behavior of conformational transition, non-transition, and β-sheet to helix transition to provide high delivery effectiveness into the cytoplasm, and also specificity (Figure 4c). 3.4. Cell-penetrating mechanism. To reveal the cell penetrating mechanism, a fluorescent uptake study was carried out using flow cytometry against NCI-H460 after the energy-dependent endocytosis pathway was completely inhibited by pretreatment of chloropromazine and incubation at 4°C (Figure 4d-e, S10). As shown in Figure 4d, the fluorescent internalization levels of all the polypeptides under endocytosis-blocking conditions were dramatically inferior. In addition, it was found that the cell penetrating property of PABL3 was capable of remaining inactive at pH 7.4, which might be the driving force of the pH-dependent cell-penetrating characteristic (Figure 4d). On the other hand, the uptake efficacy of PABL3 at pH 6 with chloropromazine treatment was comparable to that with 37°C incubation, which verified that PABL3 was not internalized via clathrin-mediated endocytosis (Figure 4e). Granted that the cellular uptake of PLL and PABL3 was nearly blocked regardless of the pH levels and cell lines at 4°C (inhibition of total energy-dependent internalization), the cellular internalization was somewhat maintained (~30%) at pH 6 (Figure 4d-e). This suggested that energy-independent cell-penetration was dominant over the other energy-required endocytosis pathways during cellular internalization. Liposome leakage and fluorescent uptake levels pretreated by PCCPs were assessed to confirm cellular membrane destabilization driven by the direct penetration (Figure 4f, S11). As shown in Figure S11, it was verified for PABL3 that liposomal membranes was immensely disrupted only at pH 6. On the other hand, it was shown that PLL and PIABL3 did not followed pH-dependent membrane disruption (Figure S11). To scrutinize
the membrane destabilizing activity in cellular conditions, the calcein uptake levels of PCCPs were determined by the pretreatment of PCCPs (Figure 4f). The uptake efficacy of PABL3 at pH 6 was almost two-fold incremental than that at pH 7.4 (Figure 4f). However, PLL and PIABL3 did not show pH-dependent membrane destabilization although they possessed a destabilizing property (Figure 4f). Based on the results of cellular internalization mechanism, it can be concluded that pH-controllable helicity provided specific cell-penetrating capability exclusively at low pH. CLSM was performed to strongly verify our hypothesis regarding direct penetration driven by cationic helix (Figure 5a-d). In figure 5d, the green fluorescence of PABL3 at pH 7.4 was considerably overlapped with red one (lysosome) while the green signal at pH 6 was rarely merged with the red, which indicated that PABL3 followed the behavior of direct penetration exclusively at pH 6 (Figure 5d). However, PLL followed endocytic pathways with low delivery efficacy because of overlapping green and red fluorescence regardless of pH levels (Figure 5b). Likewise PLL, it was found for PIABL3 that green fluorescence was overlapped with red fluorescence, meaning that PIABL3 was internalized into the cells via endocytosis pathway (Figure 5c). Therefore, pH-induced low-to-high helicity transition could endow selective cellular penetration at tumor pH. Cell viability was quantified using methyl thiazolyl tetrazolium (MTT) assay against NCI-H460 and HFL1 at pH 7.4 and pH 6 to monitor the biocompatibility of polypeptides. (Figure 5e) Cells were treated with various polypeptide concentrations for 1 day. Both PIABL3 and PABL3 showed outstanding cytocompatibility (>80% relative cell viability) against two cell lines at the ambient polypeptide concentrations (up to 200 µg/mL), indicating that the polypeptides were notably biocompatible as a delivery agent (Figure 5e). However, PLL showed decreased cell viability in both cell lines at pH 7.4 and pH 6, suggesting that it remained severely detrimental to cellular growth even at low concentration, and had inapplicability for delivering cargoes (Figure 5e).
4. Discussion Cationic helical peptides (CHP) typically possess a cell-penetrating characteristic driven by the unique ordered structure [11, 31]. For this reason, they have been intensively investigated as a promising delivery carrier in drug or gene delivery systems [32-35]. In these applications,
the use of CHPs has shown critical problems with selective arrival to the target site. More specifically, CHP nanocomplexes used for cancer targeting systems have almost been dependent on enhanced permeation and retention effects pending their arrival at the tumor sites, which have caused unexpected interactions with the undesirable regions [36, 37]. For cancer targeting systems, it is imperative that the targeting capability be endowed exclusively at tumor sites, thereby circumventing non-specific binding events [36, 38, 39].
In the present work, we developed novel PCCP systems that were capable of achieving specific cellular penetration solely at cancer cells. We speculated that the helical propensity of PCCPs were modulated by the pH-induced conversion of electrostatic interactions within the side chains, thereby affecting the unusual selective penetration. To demonstrate the specific cellular penetration influenced by the structural factors, PCCPs undergoing different conformational transitions were successfully synthesized by differentiating their chemical structures. The PLL did not undergo any conformational transitions while the PIABL3 and PABL3 demonstrated different behavior, a beta sheet-to-helix transition, and an increase in helicity, respectively, as evidenced by the CD spectra (Figure 6). We envisaged that the PCCPs undergoing different pH-induced conformational transitions could significantly influence the selective penetration. This assumption was supported by a cellular uptake study, which indicated that the unique characteristic of pH-controllable helicity in PABL3 allowed specific cell-penetration against carcinoma cells at pH 6, which resulted from the rapid augmented helical propensity. On the other hand, it was found that non-transition and beta sheet-to-low helical transitions were incapable of exploiting specific cellular uptake due to low helicity. Furthermore, we conducted pH-dependent cell-penetrating mechanism studies to underpin our hypothesis. It was confirmed that PABL3 followed direct penetration behavior and selectively penetrated the cell membrane in cancer cells (Figure 4). Additionally, we postulated that the pH-activated cationic helical conformation could destabilize the plasma membranes when it was transformed to intact helices at the cell membranes. In conclusion, we designed a pH-dependent conformation-transformable polypeptide endowing specific cell-penetrating characteristics only at tumor sites. The pH-controllable electrostatic interactions between the side chains resulted in adaptable helicity, thereby inducing specificity at the desirable target site. The flexible conformational transition systems can provide a new platform for a selective targeting strategy.
5. Acknowledgements This work was financially supported by the KAIST High Risk High Return project (Project No. N10150052) and Ministry of Science, ICT, and Future Planning (Project No. NRF2014M3A9E4064580, NRF-2016R1A2B4009619).We thank the Korea Research Institute of Bioscience and Biotechnology (KRIBB) for the confocal laser scanning microscopy and flow cytometry facilities.
6. References [1] H. Lu, J. Wang, Y.G. Bai, J.W. Lang, S.Y. Liu, Y. Lin, J.J. Cheng, Ionic polypeptides with unusual helical stability, Nat Commun 2 (2011). [2] N.E. Zhou, C.T. Mant, R.S. Hodges, Effect of preferred binding domains on peptide retention behavior in reversed-phase chromatography: amphipathic alpha-helices, Peptide research 3(1) (1990) 8-20. [3] H.Y. Tang, L.C. Yin, K.H. Kim, J.J. Cheng, Helical poly(arginine) mimics with superior cellpenetrating and molecular transporting properties, Chem Sci 4(10) (2013) 3839-3844. [4] M. Zorko, U. Langel, Cell-penetrating peptides: mechanism and kinetics of cargo delivery, Adv Drug Deliver Rev 57(4) (2005) 529-545. [5] N.P. Gabrielson, H. Lu, L.C. Yin, D. Li, F. Wang, J.J. Cheng, Reactive and Bioactive Cationic aHelical Polypeptide Template for Nonviral Gene Delivery, Angew Chem Int Edit 51(5) (2012) 11431147. [6] P.C. Lyu, M.I. Liff, L.A. Marky, N.R. Kallenbach, Side-Chain Contributions to the Stability of Alpha-Helical Structure in Peptides, Science 250(4981) (1990) 669-673. [7] P.C. Lyu, J.C. Sherman, A. Chen, N.R. Kallenbach, Alpha-Helix Stabilization by Natural and Unnatural Amino-Acids with Alkyl Side-Chains, P Natl Acad Sci USA 88(12) (1991) 5317-5320. [8] L.C. Yin, Z.Y. Song, Q.H. Qu, K.H. Kim, N. Zheng, C. Yao, I. Chaudhury, H.Y. Tang, N.P. Gabrielson, F.M. Uckun, J.J. Cheng, Supramolecular Self-Assembled Nanoparticles Mediate Oral Delivery of Therapeutic TNF-alpha siRNA against Systemic Inflammation, Angew Chem Int Edit 52(22) (2013) 5757-5761. [9] L.C. Yin, H.Y. Tang, K.H. Kim, N. Zheng, Z.Y. Song, N.P. Gabrielson, H. Lu, J.J. Cheng, LightResponsive Helical Polypeptides Capable of Reducing Toxicity and Unpacking DNA: Toward Nonviral Gene Delivery, Angew Chem Int Edit 52(35) (2013) 9182-9186. [10] N. Zheng, Z. Song, Y. Liu, R. Zhang, R. Zhang, C. Yao, F.M. Uckun, L. Yin, J. Cheng, Redox-
responsive, reversibly-crosslinked thiolated cationic helical polypeptides for efficient siRNA encapsulation and delivery, Journal of controlled release : official journal of the Controlled Release Society (2015). [11] K.A. Brogden, Antimicrobial peptides: Pore formers or metabolic inhibitors in bacteria?, Nat Rev Microbiol 3(3) (2005) 238-250. [12] B. Gupta, T.S. Levchenko, V.P. Torchilin, Intracellular delivery of large molecules and small particles by cell-penetrating proteins and peptides, Adv Drug Deliver Rev 57(4) (2005) 637-651. [13] J.E. Johnson, R.B. Cornell, Membrane-Binding Amphipathic Alpha-Helical Peptide Derived from Ctp-Phosphocholine Cytidylyltransferase, Biochemistry-Us 33(14) (1994) 4327-4335. [14] J.P. Richard, K. Melikov, E. Vives, C. Ramos, B. Verbeure, M.J. Gait, L.V. Chernomordik, B. Lebleu, Cell-penetrating peptides - A reevaluation of the mechanism of cellular uptake, J Biol Chem 278(1) (2003) 585-590. [15] D.W. Pack, A.S. Hoffman, S. Pun, P.S. Stayton, Design and development of polymers for gene delivery, Nat Rev Drug Discov 4(7) (2005) 581-593. [16] E. Fleige, M.A. Quadir, R. Haag, Stimuli-responsive polymeric nanocarriers for the controlled transport of active compounds: Concepts and applications, Adv Drug Deliver Rev 64(9) (2012) 866884. [17] J.Z. Du, T.M. Sun, W.J. Song, J. Wu, J. Wang, A Tumor-Acidity-Activated Charge-Conversional Nanogel as an Intelligent Vehicle for Promoted Tumoral-Cell Uptake and Drug Delivery, Angew Chem Int Edit 49(21) (2010) 3621-3626. [18] E.K. Rofstad, B. Mathiesen, K. Kindem, K. Galappathi, Acidic extracellular pH promotes experimental metastasis of human melanoma cells in athymic nude mice, Cancer Res 66(13) (2006) 6699-6707. [19] J.R. Hernandez, H.A. Klok, Synthesis and ring-opening (Co) polymerization of L-lysine Ncarboxyanhydrides containing labile side-chain protective groups, J Polym Sci Pol Chem 41(9) (2003) 1167-1187. [20] H. Lu, J. Cheng, Hexamethyldisilazane-mediated controlled polymerization of alpha-amino acid N-carboxyanhydrides, Journal of the American Chemical Society 129(46) (2007) 14114-5. [21] D. Lee, K. Choe, Y. Jeong, J. Yoo, S.M. Lee, J.H. Park, P. Kim, Y.C. Kim, Establishment of a controlled insulin delivery system using a glucose-responsive double-layered nanogel, Rsc Adv 5(19) (2015) 14482-14491. [22] K.S. Krannig, H. Schlaad, pH-Responsive Bioactive Glycopolypeptides with Enhanced Helicity and Solubility in Aqueous Solution, Journal of the American Chemical Society 134(45) (2012) 1854218545. [23] I. Noh, H.O. Kim, J. Choi, Y. Choi, D.K. Lee, Y.M. Huh, S. Haam, Co-delivery of paclitaxel and gemcitabine via CD44-targeting nanocarriers as a prodrug with synergistic antitumor activity against
human biliary cancer, Biomaterials 53 (2015) 763-774. [24] J. Yoo, D. Lee, V. Gujrati, N.S. Rejinold, K.M. Lekshmi, S. Uthaman, C. Jeong, I.K. Park, S. Jon, Y.C. Kim, Bioreducible branched poly(modified nona-arginine) cell-penetrating peptide as a novel gene delivery platform, J Control Release (2016). [25] B. Davidson, G.D. Fasman, The conformational transitions of uncharged poly-L-lysine. Alpha helix-random coil-beta structure, Biochemistry-Us 6(6) (1967) 1616-29. [26] I. Satake, J.T. Yang, Effect of chain length and concentration of anionic surfactants on the conformational transitions of poly(L-ornithine) and poly(L-lysine) in aqueous solution, Biochem Biophys Res Commun 54(3) (1973) 930-6. [27] K. Park, A. Perczel, G.D. Fasman, Differentiation between Transmembrane Helices and Peripheral Helices by the Deconvolution of Circular-Dichroism Spectra of Membrane-Proteins, Protein Sci 1(8) (1992) 1032-1049. [28] N.J. Greenfield, Using circular dichroism spectra to estimate protein secondary structure, Nat Protoc 1(6) (2006) 2876-2890. [29] E.G. Baker, G.J. Bartlett, M.P. Crump, R.B. Sessions, N. Linden, C.F.J. Faul, D.N. Woolfson, Local and macroscopic electrostatic interactions in single alpha-helices, Nat Chem Biol 11(3) (2015) 221-U92. [30] S. Padmanabhan, S. Marqusee, T. Ridgeway, T.M. Laue, R.L. Baldwin, Relative Helix-Forming Tendencies of Nonpolar Amino-Acids, Nature 344(6263) (1990) 268-270. [31] M. Zasloff, Antimicrobial peptides of multicellular organisms, Nature 415(6870) (2002) 389-395. [32] J. Farrera-Sinfreu, E. Giralt, S. Castel, F. Albericio, M. Royo, Cell-penetrating cis-gamma-aminoL-proline-derived peptides, J Am Chem Soc 127(26) (2005) 9459-9468. [33] I. Nakase, Y. Konishi, M. Ueda, H. Saji, S. Futaki, Accumulation of arginine-rich cell-penetrating peptides in tumors and the potential for anticancer drug delivery in vivo, J Control Release 159(2) (2012) 181-188. [34] H. Park, H. Tsutsumi, H. Mihara, Cell penetration and cell-selective drug delivery using alphahelix peptides conjugated with gold nanoparticles, Biomaterials 34(20) (2013) 4872-9. [35] K. Rittner, A. Benavente, A. Bompard-Sorlet, F. Heitz, G. Divita, R. Brasseur, E. Jacobs, New basic membrane-destabilizing peptides for plasmid-based gene delivery in vitro and in vivo, Mol Ther 5(2) (2002) 104-14. [36] R. Langer, Drug delivery and targeting, Nature 392(6679) (1998) 5-10. [37] R. Mo, Q. Sun, J.W. Xue, N. Li, W.Y. Li, C. Zhang, Q.N. Ping, Multistage pH-Responsive Liposomes for Mitochondrial-Targeted Anticancer Drug Delivery, Adv Mater 24(27) (2012) 36593665. [38] G. Mikhaylov, U. Mikac, A.A. Magaeva, V.I. Itin, E.P. Naiden, I. Psakhye, L. Babes, T. Reinheckel, C. Peters, R. Zeiser, M. Bogyo, V. Turk, S.G. Psakhye, B. Turk, O. Vasiljeva, Ferri-
liposomes as an MRI-visible drug-delivery system for targeting tumours and their microenvironment, Nat Nanotechnol 6(9) (2011) 594-602. [39] L. Rajendran, H.J. Knolker, K. Simons, Subcellular targeting strategies for drug design and delivery, Nat Rev Drug Discov 9(1) (2010) 29-42.
Figure and table captions Figure 1. Design of PCCPs, and proposed mechanism of pH-controllable helicity a nd selective cellular penetration. (a) Proposed mechanism of PCCP possessing a pHactivated cell penetrating property exclusively at the tumor extracellular matrix. 2° and 3° indicate “secondary” and “tertiary”, respectively. (b) Schematic illustration of the PCCP undergoing pH-dependent conformational transition induced by the charge balances of two opposite ions. Figure 2. Estimation of physical properties. (a) Prediction of pKa values of PCCPs. (b) Estimation of log D value with pH variation from pH 0 to pH 14. Theoretical calculations of the molecular properties were calculated by using Marvin and JChem calculator plugins. All the values were estimated by Marvin Sketch software. Lograthim distribution-coefficient (log D) is correlated to water solubility. (log D < 0, water-soluble) Figure 3. Determination of different conformational transitions using CD spectrometry. (a) CD spectra of PIABL series with different pH levels at 20°C. (b) CD spectra of PABL series with different pH levels at 20°C. (c) CD spectra of PLL with different pH levels at 20°C. Molar residue ellipticity at 222 nm plotted against different pH levels at (d) 20°C, and at (e) 37°C. Molar residue ellipticity = (mdeg x molecular weight)/(pathlength in millimeter x concentration). Figure 4. Calcein uptake and cell-penetrating mechanism studies at different pH levels. (a) The uptake levels of naked calcein, calcein-loaded PLL, PIABL3, and PABL3 using NCI and HFL1 cell lines at pH 7.4 and pH 6. (b) The uptake efficacy of all the controls against NCI-H460 at pH 6 and HFL1 at pH 7.4 to confirm the pH-depedent cell-penetrating characteristic. (c) CLSM images of naked calcein, calcein-loaded PLL, PIABL3, and PABL3 against NCI-H460 and HFL1 both at pH 7.4 and pH 6. 4',6-diamidino-2-phenylindole (DAPI): blue; Calcein: green; Scale bar: 25 µm) A cell-penetrating mechanism study was performed using NCI-H460 with endocytsis-blocking conditions, chloropromazine preincubation, and
4°C incubation, (d) at pH 7.4 and (e) at pH 6. (f) Uptake efficiency against NCI-H460 cells pretreated with none, PLL, PIABL3, and PABL3 prior to the addition of calcein as evaluated by flow cytometry. All the error bars are expressed as S.D. (n=4). Figure 5. Visualization of cell-penetrating mechanism and cytotoxicity study. Visualization of cellular uptake mechanism of (a) naked calcein, (b) PLL, (c) PIABL3 and (d) PABL3 against NCI-H460 cells at pH 7.4 and pH 6 by CLSM. (4',6-diamidino-2phenylindole (DAPI): blue; Lysotraker: red; Calcein: green; Scale bar: 50 µm) (e) Relative cell viability of polypeptides, PLL, PIABL3, and PABL3 at pH 7.4 and pH 6, toward NCIH460 and HFL1 cell lines after 1 day treatment, as evaluated by MTT assay. The error bars are expressed as S.D. (n=3). Figure 6. Schematic illustration of different conformational transitions. Non-transition in regardless of pH levels in PLL. Beta sheet-to-helix transition and low-to-high helix transition observed in PIABL3 and PABL3, respectively, with decreasing pH.
Table 1. Characterization of PLL, PIABL series and PABL series.
Table 1. Characterization of PLL, PIABL series and PABL series.
a
Helicity (%)d at pH 7.5 (20 / 37°C)
Helicity (%) at pH 6 (20 / 37°C)
Helicity (%) at pH 5 (20 / 37°C)
Helicity (%) at pH 4 (20 / 37°C)
~100 b
<1.1
39.5 / 44.6
48.3 / 46.3
42.0 / 43.9
37.1 / 44.9
97
~100b,c
<1.1
30.9 / 29.1
33.4 / 13.2
28.4 / 18.3
27.4 / 17.6
97
~100b,c
<1.1
40.9 / 30.2
43.8 / 27.6
31.3 / 19.4
26.3 / 21.0
PIABL3
97
~100
b,c
<1.1
55.8 / 45.1
67.4 / 45.4
33.5 / 31.3
29.2 / 29.6
PABL1
97
~100b,c
<1.1
44.1 / 40.1
44.2 / 44.6
80.0 / 62.7
73.1 / 54.8
PABL2
97
~100b,c
<1.1
28.5 / 17.7
62.5 / 44.3
59.7 / 45.5
48.9 / 43.9
PABL3
97
~100b,c
<1.1
29.6 / 23.5
78.3 / 68.4
92.8 / 90.6
31.3 / 23.5
Repeating unita (Lys)
Degree of modification (%)
PLL
97
PIABL1 PIABL2
Polypeptide Entry
a
PDI
The number of lysine units and polydiserpsity index (PDI) were determined by gel permeation chromatography (GPC) as a standard of PTFL. b Degree of modification was quantified by NMR spectrometry. c Degree of modification was calculated by elemental analysis. d Helicity was determined by the equation; helicity (%)= (-[θ]222 nm+3000)/39000 x 100
We developed pH-controllable cell-penetrating polypeptides (PCCPs) undergoing pHinduced conformational transitions. Unlike natural cell-penetrating peptides, PCCPs was capable of penetrating the plasma membranes dominantly at tumor pH, driven by pHcontrolled helicity. The conformation of PCCPs at neutral pH showed low helical propensity because of dominant electrostatic attractions within the side chains. However, the helicity of PCCPs was considerably augmented by the balance of electrostatic interactions, thereby inducing
selective
cellular
penetration.
Three
polypeptides
undergoing
different
conformational transitions were prepared to verify the selective cellular uptake influenced by their structures. The PCCP undergoing low-to-high helical conformation provided the tumor specificity and enhanced uptake efficiency. pH-induced conformation-transformable polypeptide might provide a novel platform for stimuli-triggered targeting systems.
*Graphical Abstract