2022 Volume 70 Issue 11 Pages 812-817
Amphipathic cell-penetrating peptides based on the pep-1 sequence were synthesized by replacing the three hydrophilic glutamic acid residues with disubstituted, non-proteinogenic, hydrophobic amino acids. These substitutions facilitated maintenance of the peptides’ secondary structure in a helical conformation, even in aqueous solution. Stability against enzymatic degradation was improved through the use of disubstituted amino acids. The resultant peptides exhibited high membrane permeability that remained relatively stable during prolonged incubation times. The results of this study indicate that the use of non-proteinogenic amino acids may be an effective approach to improve the cell membrane permeability for existing amphiphilic peptides.
The barrier properties of cell membranes prevent hydrophilic compounds with large molecular weights, such as proteins and antibodies, from penetrating the membrane. Several studies have investigated methods for permeating the cell membrane to enable transport of these macromolecules into cells; one approach is the use of cell penetrating peptides (CPPs).1–5) CPPs are classified into three types based on their properties. These include cationic peptides, containing abundant basic amino acids; amphipathic peptides, consisting of a balanced combination of hydrophilic and hydrophobic amino acids; and hydrophobic peptides, containing abundant hydrophobic amino acids. Among those, amphipathic peptides can be further classified into several types depending on the way the hydrophilic and hydrophobic moieties are combined.6) First-generation amphipathic peptides adopt a helical structure in which the hydrophobic moiety, containing abundant hydrophobic amino acids, is linked by a linker to a hydrophilic moiety, containing abundant hydrophilic (cationic) amino acids.7–10) Second-generation amphipathic peptides also form a helical structure in which hydrophilic groups are concentrated on one side of the helix with hydrophobic groups on the opposite side.11–14) This is characteristic of many antimicrobial peptides. In addition, various proline-rich amphipathic peptides and others have been developed, making use of a polyproline helical structure rather than an α-helical one.
In the present study, we focused on a first-generation amphiphilic peptide, pep-1. This peptide was first reported by M. C. Morris in 2001 and is a chimera consisting of a “hydrophobic domain” (12 residues: KETWWETWWTEW) containing six tryptophan residues and a “hydrophilic domain” (6 residues: KKKRKV) derived from the nuclear localization sequence (NLS) of the simian virus, joined by a “spacer domain” (3 residues: SQP).15) Although this peptide is classified as being amphipathic with linked hydrophilic and hydrophobic domains, the hydrophobic domain consists of seven hydrophilic and five hydrophobic residues. If this domain forms a helical structure, it could be classified as a second-generation amphiphilic peptide. It was expected that converting the hydrophilic amino acids of this domain to hydrophobic amino acids would increase the hydrophobicity of the domain and that subsequent investigation of this modified peptide 1 would provide new insights.
Non-proteinogenic α,α-disubstituted amino acids (dAAs) were used as the introduced hydrophobic amino acids. DAAs have the unique features of both stabilizing the helical secondary structure of the peptides that contain them and providing resistance to degradation by hydrolytic enzymes.16,17) It has also been reported that they improve the function of cationic CPPs.18–21) In our previous studies, we investigated the effects of replacing various hydrophobic amino acids in the pep-1 sequence with dAAs and confirmed that replacing the three glutamic acid (Glu) residues in the hydrophobic domain with a dAA (1-aminocyclopentanecarboxylic acid (Ac5c)) results in high cell membrane permeability.22) For the present study, we synthesized and evaluated four peptide analogs to investigate in more detail the effects of replacing Glu with a dAA on cell membrane permeability.
To evaluate membrane permeability, peptides were labeled with 5(6)-carboxyfluorescein (5(6)-CF) through a glycine (Gly) linker, serving as a spacer, on the N-terminal side. The peptide based on the pep-1 sequence was designated peptide 1, and the peptide in which the Glu residues in the hydrophobic domain were replaced with Ac5c was designated peptide 2. A peptide in which the Glu residues were replaced with 1-aminoisobutylic acid (Aib), which is the simplest dAA and often used as a helical promoter for peptides, was designated peptide 3. Finally, in peptide 4 Glu was replaced with alanine (Ala) to make the overall charge of the peptide the same as that of peptide 2. These peptides were synthesized using standard 9-fluorenylmethyl-oxycarbonyl (Fmoc) solid-phase synthesis and purified with preparative HPLC. Their synthesis was confirmed with matrix assisted laser desorption/ionization-time of flight (MALDI-TOF) MS and analytical HPLC (Table 1).
| Peptide | Sequencea) | Retention time (min)b) | Net chargec) |
|---|---|---|---|
| 1 | 5(6)-CF-GKETWWETWWTEWSQPKKKRKV-NH2 | 16.0 | +2 |
| 2 | 5(6)-CF-GKXTWWXTWWTXWSQPKKKRKV-NH2 | 17.2 | +5 |
| 3 | 5(6)-CF-GKUTWWUTWWTUWSQPKKKRKV-NH2 | 16.7 | +5 |
| 4 | 5(6)-CF-GKATWWATWWTAWSQPKKKRKV-NH2 | 16.2 | +5 |
The letters X and U indicate Ac5c and Aib, respectively. a) The letters in bold are the substituted amino acids. The C-terminus was amidated. b) Retention times analyzed with RP-HPLC on a COSMOSIL 5C18-AR-II column (4.6 mm I.D. × 250 mm) using a linear gradient from 0% to 100% CH3CN in H2O (each containing 0.1% TFA) over 30 min (detection at 220 nm). c) Net charge of the peptides at pH 7.0, not including the charge of the N-terminus.
The CD spectra for peptides 1–4 were measured to obtain information about their secondary structures in 2,2,2-trifluoroethanol (TFE) and 10 mM phosphate buffer (PB, pH 7.4) (Fig. 1). A positive maximum at 192 nm and negative maxima at 208 and 222 nm were diagnostic of a right-handed (P) helical structure, while a negative maximum at 195–200 nm and a positive maximum at 217 nm was diagnostic of a random coil structure.23) Measurements in TFE, which induces formation of a stable secondary structure in peptides,24) showed a negative maximum at 208 nm and a negative maximum around 230 nm for peptide 2 and 3. The maximum at 208 nm may be due to the P-helical structure, while the maximum at 230 nm may be due to the interaction of the aromatic tryptophan side chain with the peptide backbone and the stacking of the indole rings.25) On the other hand, peptide 1 had a negative maximum at 201 nm in 10 mM PB, indicating that it has a random coil structure. In addition, peptide 2 had a negative maximum at 210 nm, and peptide 3 had a negative maximum at 206 nm, suggesting that their structures are more helical than a random coil. Peptide 4 had large negative maxima at 208 nm and 222 nm, suggesting that the P-helical structure is maintained in water.
Peptide concentration: 46.7 µM.
The synthesized peptides were incubated with HeLa human cervical cancer cells and HEK293 human embryonic kidney cells for various times to evaluate cellular uptake. With both cell types, peptides 2–4 exhibited much greater membrane permeability than peptide 1 at all times (Fig. 2). The intracellular levels of peptide 4 increased for up to 6 h but then declined such that the level after 24 h of incubation was about the same as after 0.5 h. On the other hand, intracellular levels of peptides 2 and 3, which contained dAAs, were higher after 24 h of incubation than after 0.5 h, which indicated these peptides tended to maintain membrane permeability at a high level, even after prolonged incubation of the cells.
Mean fluorescence intensities were evaluated using flow cytometry. Values are means ± standard deviation of three independent experiments. Asterisks indicate significant differences vs. peptide 1 (*: p < 0.05, **: p < 0.01, ***: p < 0.001).
To investigate the reason for the more prolonged membrane permeability of peptides, their resistance to proteolytic enzymes was evaluated using trypsin, an endopeptidase that cleaves peptides chains mainly at their carboxyl side. The time-dependent degradation of the peptides by trypsin was assessed by using HPLC to measure the relative levels of peptide remaining at selected times after exposure to trypsin (Fig. 3). Peptide 1 was degraded by 50% after incubation for 6 h with trypsin and was degraded by more than 90% after 24 h. By contrast 60% peptide 2, which contains Ac5c, remained after 24 h of incubation with trypsin. Interestingly, nearly 90% of peptides 3 and 4, in which Glu residues were replaced with Aib or Ala residues, respectively, were degraded after only 1 h of incubation with trypsin, and they were nearly completely degraded after 6 h. Although introduction of dAAs into peptides reportedly increases their resistance of peptidases, these results show that the type of dAA introduced can have a profound effect on their enzymatic stability.
Shown are relative levels of the peptide after the indicated incubation times.
Cytotoxicity is often a concern with the use of amphipathic peptides as CPPs. The cytotoxicity of peptides 1–4 was evaluated based on two indicators: cell proliferation assessed with a cell counting kit-8 (CCK-8) assay and cell membrane integrity assessed with a lactate dehydrogenase (LDH) assay (Fig. 4). HeLa and HEK293 cells were incubated for 24 h with peptides 1–4 at a concentration of 2 µM, after which the aforementioned assays were carried out. At a concentration of 2 µM, both cytotoxicity and inhibition of cell proliferation were negligible for all four peptides tested.
(a) Cell viability was assessed using CCK-8 assays, and the data are presented as percentages relative to untreated control cells. (b) LDH activity was assessed using a Cytotoxicity LDH Assay Kit-WST, and the data are presented as percentages relative to control cells lysed in lysis buffer. Asterisks indicate significant differences vs. control (***: p < 0.001).
After incubating HeLa and HEK293 cells with the four peptides, the intracellular distribution of the peptides was evaluated using CLSM (Fig. 5, Supplementary Figs. S5, S6). In HeLa cells, all of the peptides were observed as intracellular dots that co-localized with lysosomes and endosomes identified using LysoTrackerRed. By contrast, in HEK293 cells, almost no intracellular peptide 1 was detected, whereas peptides 2–4 diffused throughout the cytoplasm, in addition to a slight observation of co-localized with lysosomes and endosomes.
Acidic late endosomes/lysosomes were stained with LysoTracker Red (red), and nuclei were stained with Hoechst 33342 (blue).
For this study, we designed and synthesized peptides in which the three Glu residues in the pep-1 sequence were replaced with hydrophobic amino acids. We then evaluated the cell membrane permeability of the modified peptides. Our measurement of the peptides’ CD spectra revealed that all the peptides synthesized for this study adopt a helical structure in TFE, whereas their secondary structures differ in PB. This result is important, as all the cell culture experiments were performed in an aqueous environment. Peptide 1 mainly adopted a random coil structure in PB and showed little cellular uptake. By contrast, peptides 2–4, which appeared to maintain their helical conformation in water, showed relatively high cell membrane permeability.
If these peptides adopt a helical structure, even in an aqueous environment, their amphipathic helical structure likely enables them to interact with the cell membrane surface. Although it is the “hydrophobic domain” of the pep-1 sequence that contains Glu residues, this domain, in fact, contains a balance of hydrophilic and hydrophobic amino acids (7 hydrophilic and 5 hydrophobic residues) and can also be considered an “amphipathic domain.” This is noteworthy because if this domain takes a helical structure, the peptides as a whole would have an amphipathic helical structure. Considering the helical wheel of this amphipathic domain, the angle of the hydrophobic surface is about 160° in peptide 1, while the angle of the hydrophobic surface is about 240° in peptide 2–4 (Fig. 6). Given that the actions of amphiphilic helical peptides at lipid membranes differ depending on the angle of the hydrophobic surfaces,26,27) this difference may have contributed to the observed differences in membrane permeability.
(a) Peptide 1 (b) peptides 2–4.
When cells were incubated with the peptides, the time-dependent changes in the intracellular levels of peptide 4 were characterized by a sharp rise during the first 6 h followed by a steady decline over the next 18 h. That decline in intracellular peptide levels was slower for peptides 2 and 3 than for peptide 4. We also observed that peptide 2 is highly stable against the hydrolytic enzyme trypsin, which may explain why its levels remained comparatively stable in medium and intracellularly, even after prolonged contact with cells. By contrast, Aib-containing peptide 3 exhibited overall lower cellular uptake, despite the presence of di-substituted amino acids. Ac5c is a more sterically hindered dAAs compared to Aib, which may have increased the stability of the peptide against enzymatic degradation. It was unexpected that peptide 4, in which the Glu residues were substituted with Ala residues, would show higher membrane permeability than peptide 3, though after incubation for 24 h, intracellular levels of peptide 4 had decreased to a level not significantly different from that of peptide 3.
Observations made using CLSM revealed differences in the uptake of these peptides into HeLa and HEK293 cells. In HeLa cells, all of the peptides were internalized through endocytosis and localized to endosomes. In HEK293 cells, by contrast, peptides 2–4 seemed to be diffusely distributed in the cytoplasm, in addition to a slight observation of localized to endosomes. Although no clear differences between the peptides 2–4 were detected from the CLSM observations, it was clear that they are capable of distribution into the cytoplasm in some cell types. The CLSM image difference between peptide 1 and peptides 2–4 may be due to the replacement of Glu residues with hydrophobic amino acids, which changed the mechanism of cell penetrating, but this also requires further investigation. It is anticipated that future studies will reveal further details of the pathway by which these peptides are taken up by cells and the subsequent actions of the intracellular peptides.
Our findings indicate that replacing the hydrophilic Glu residues within the hydrophobic domain of pep-1 with hydrophobic amino acids, especially with dAAs, enhances the membrane permeability of the peptides. This is likely because the overall helical structure of the peptides was stabilized, and the hydrophobic domain acted as an “amphipathic peptide.” In addition, the introduction of Ac5c increased the stability of the peptide against enzymatic degradation, leading to stable cellular uptake over prolonged incubation times. That result along with the observation that the peptides are capable of distribution into the cytoplasm, at least in HEK293 cells, suggests there may be a correlation between the helical structure of a peptides and its intracellular dynamics. We therefore believe that these results may shed light on ways to enhance the utility of existing amphipathic peptides and facilitate future peptide drug discovery.
All solvents and chemicals were purchased from FUJIFILM Wako Pure Chemical Corporation (Osaka, Japan), Nacalai Tesque, Inc. (Kyoto, Japan), Tokyo Chemical Industry Co., Ltd. (Tokyo, Japan), Watanabe Chemical Industries, Ltd. (Hiroshima, Japan), and Sigma-Aldrich Co. LLC. (Saint Louis, MO, U.S.A.) and were used without further purification. Dulbecco’s modified Eagle’s medium (DMEM) and Eagle’s minimum essential medium (EMEM) were purchased from Nacalai Tesque, Inc. and Wako Pure Chemical Corporation, respectively. Fetal bovine serum (FBS) was purchased from Sigma-Aldrich Co. LLC. The FBS was heated to 56 °C for 30 min in a water bath before use. CCK-8, LDH assay kit-WST, and Hoechst 33342 were purchased from Dojindo Laboratories (Kumamoto, Japan). LysoTracker Red was purchased from Thermo Fisher Scientific (Waltham, MA, U.S.A.).
Synthesis of PeptidesPeptides were synthesized on a solid support using Fmoc solid-phase methods with commercially available Rink amide resin and Fmoc-amino acids. The following describes a representative coupling and deprotection cycle on a 25-µmol scale. Fmoc-NH-SAL resin (55.6 mg; loading: 0.45 mmol/g) was soaked in N,N-dimethylformamide (DMF) for 30 min. After removing the DMF, 20% piperidine in DMF was added to the resin for deprotection. After washing out the piperidine, Fmoc-amino acid or 5(6)-CF (4 equivalent (equiv.), (1-cyano-2-ethoxy-2-oxoethylideneaminooxy)dimethylamino(morpholino)carbenium hexafluorophosphate (COMU) (4 equiv.), and DIPEA (8 equiv.) dissolved in DMF were added to the coupling reaction. The resin was then suspended in a cleavage cocktail (TFA, 1.9 mL; H2O, 50 µL; TIS, 50 µL) at room temperature for 90 min. The TFA solution was evaporated to a small volume, after which cold diethyl ether was added to the solution to precipitate the peptides. The dried crude peptides were dissolved in H2O and purified with reverse-phase (RP)-HPLC using a COSMOSIL Packed Column 5C18-AR (20 ID × 250 nm) (Nacalai). Lyophilization resulted in the formation of yellow crystals, which were characterized using analytical RP-HPLC (COSMOSIL Packed Column 5C18-MS-II, 4.6 ID × 250 nm) and MALDI-TOF-MS (Bruker Daltonics Ultraflex, Fremont, CA, U.S.A.). RP-HPLC was performed utilizing a JASCO-PU-2089 Plus with a JASCO-2077 Plus as a UV-detector (JASCO, Tokyo, Japan). Solvent A was 0.1% TFA in H2O; solvent B was 0.1% TFA in CH3CN. The purification procedure required gradient conditions (from 100 to 0% solvent A over 30 min) with a flow rate of 7 mL/min and detection at 220 nm. The purities of the final compounds were further confirmed using similar RP-HPLC conditions (from 0 to 100% solvent B over 30 min) with a flow rate of 1 mL/min.
CD SpectraCD spectra were recorded with a JASCO J-820 spectropolarimeter (JASCO) using a quartz cell with a 10-mm path length. Data are expressed in terms of [θ]R; i.e., residue molar ellipticity (deg·cm2·dmol−1).
Peptide Stability against Protease DigestionAliquots of peptide solution (600 µL; 20 µM) in 0.0005% (w/v) trypsin/phosphate-buffered saline (PBS) were incubated at 37 °C. After each incubation period, 100 µL of peptide solution was taken and diluted with 250 µL of 1% TFA/PBS to inactivate the protease. A saturated α-cyano-4-hydroxycinnamic acid aqueous solution (50 µL) was added as an internal standard, followed by RP-HPLC analysis.
Cell CultureHeLa and HEK293 cells were respectively cultured in DMEM and EMEM supplemented with 10% FBS. Penicillin/streptomycin were used as antibiotics. All cells were incubated at 37 °C under a humidified atmosphere of 5% CO2 in air.
Cellular Uptake of PeptidesHeLa or HEK293 cells were seeded onto 6-well plates (2.0 × 105 cells/well) and cultured in DMEM or EMEM supplemented with 10% FBS. After incubation for 24 h, the medium was refreshed, and peptide solution was added to each well to a final concentration of 2 µM. After each incubation period, the cells were washed with PBS and trypsinized. The collected cells were washed twice with PBS, after which they were suspended in 1 mL of PBS, and the mean intensity of the fluorescence from the cells was measured using flow cytometry (BD FACSAriaTM III, Becton Dickinson, Franklin Lakes, NJ, U.S.A.). The results are presented as the mean and standard deviation from three samples. Values of p were calculated using Welch’s t-test (two-tailed), comparing the results to those obtained with peptide 1.
Cell ViabilityHeLa or HEK293 cells were seeded onto a 96-well culture plate (5 × 103 cells/well) and incubated in DMEM or EMEM containing 10% FBS. After incubation for 24 h, peptide solution diluted with the medium was added to each well to a final concentration of 2 µM, and the cells were incubated for an additional 24 h. Cell counting kit-8 solution was then added to each well, and the cells were incubated for a further 2 h, after which the absorbance at 450 nm in each well was measured to assess cell viability. The results are presented as the mean and standard deviation from six samples. Values of p were calculated using Welch’s t-test (two tailed), comparing the results to controls.
CytotoxicityHeLa or HEK293 cells were seeded onto a 96-well culture plate (5 × 103 cells/well) and incubated in DMEM or EMEM containing 10% FBS. After incubation for 24 h, peptide solution diluted with the medium was added to each well to a final concentration of 2 µM, and the cells were incubated for an additional 24 h. Lysis buffer was then added to some wells as a high control, and the cells were incubated for a further 30 min. Alternatively, a working solution was added to each well, after which the cells were allowed to stand still for 30 min at room temperature protected from light. The absorbance at 490 nm in each well was then measured to assess cell cytotoxicity. The results are presented as the mean and standard deviation from six samples. Values of p were calculated using Welch’s t-test (two-tailed), comparing the results to the control. Symbols *** mean p < 0.001.
CLSMCells were seeded onto 8-well chambered cover glasses (Iwaki, Tokyo, Japan) (2 × 104 cells/well) and incubated overnight in 200 µL of DMEM or EMEM containing 10% FBS. The medium was then refreshed, and peptide solution was added to each well to a final concentration of 2 µM. After incubation for 24 h, the medium was removed, and the cells were washed 3 times with PBS. The intracellular distribution of peptides was observed using CLSM after staining late endosomes/lysosomes with LysoTracker Red and nuclei with Hoechst 33342. The observations were made using an LSM 700 (Carl Zeiss, Oberlochen, Germany) equipped with a 63× objective lens (Plan-Apochromat, Carl Zeiss) at an excitation wavelength of 405 nm (UV laser) for Hoechst 33342, 488 nm (Ar laser) for the peptides, and 543 nm (He–Ne laser) for LysoTracker Red.
The authors declare no conflict of interest.
This article contains supplementary materials.
