Amphiphilic Cell-Penetrating Peptides Containing Arginine and Hydrophobic Residues as Protein Delivery Agents

The entry of proteins through the cell membrane is challenging, thus limiting their use as potential therapeutics. Seven cell-penetrating peptides, designed in our laboratory, were evaluated for the delivery of proteins. Fmoc solid-phase peptide synthesis was utilized for the synthesis of seven cyclic or hybrid cyclic–linear amphiphilic peptides composed of hydrophobic (tryptophan (W) or 3,3-diphenylalanine (Dip) and positively-charged arginine (R) residues, such as [WR]4, [WR]9, [WWRR]4, [WWRR]5, [(RW)5K](RW)5, [R5K]W7, and [DipR]5. Confocal microscopy was used to screen the peptides as a protein delivery system of model cargo proteins, green and red fluorescein proteins (GFP and RFP). Based on the confocal microscopy results, [WR]9 and [DipR]5 were found to be more efficient among all the peptides and were selected for further studies. [WR]9 (1–10 µM) + protein (GFP and RFP) physical mixture did not show high cytotoxicity (>90% viability) in triple-negative breast cancer cells (MDA-MB-231) after 24 h, while [DipR]5 (1–10 µM) physical mixture with GFP exhibited more than 81% cell viability. Confocal microscopy images revealed internalization of GFP and RFP in MDA-MB-231 cells using [WR]9 (2–10 μM) and [DipR]5 (1–10 µM). Fluorescence-activated cell sorting (FACS) analysis indicated that the cellular uptake of GFP was concentration-dependent in the presence of [WR]9 in MDA-MB-231 cells after 3 h of incubation at 37 °C. The concentration-dependent uptake of GFP and RFP was also observed in the presence of [DipR5] in SK-OV-3 and MDA-MB-231 cells after 3 h of incubation at 37 °C. FACS analysis indicated that the cellular uptake of GFP in the presence of [WR]9 was partially decreased by methyl-β-cyclodextrin and nystatin as endocytosis inhibitors after 3 h of incubation in MDA-MB-231 cells, whereas nystatin and chlorpromazine as endocytosis inhibitors slightly reduced the uptake of GFP in the presence of [DipR]5 after 3 h of incubation in MDA-MB-231. [WR]9 was able to deliver therapeutically relevant proteins (Histone H2A) at different concentrations. These results provide insight into the use of amphiphilic cyclic peptides in the delivery of protein-related therapeutics.


Introduction
Proteins are macromolecules with diverse roles as enzymes, receptors, or channels in cell membranes, catalyzing biochemical reactions and therapeutic agents [1][2][3]. There are about 19,000-20,000 different protein-coding genes in the human genome [4,5]. If we consider the post-translational modification of proteins, such as phosphorylation, cleavage, glycosylation, and acylation, a large number of functionally distinct proteins are present in the human proteome. If viewed from a therapeutics perspective, these estimates represent a huge opportunity to develop protein therapeutics to ease many diseases.
We have previously reported that cyclic peptides containing tryptophan (W) and arginine (R) residues, [WR]4 [37], [WR]5 [38] [39], can act as molecular transporters. The physical mixture of [WR]9 ( Figure 1) (10 µM) and F′-GpYEEI at (2 µM) significantly enhanced the cellular uptake by 20-fold when compared to F′-GpYEEI alone at the same concentration after 3 h of incubation in CCRF-CEM cells, while the physical mixture of [WR]5 (10 µM) with F′-GpYEEI (2 µM) enhanced the uptake by only 4-fold [39]. These results indicate the possibility of using large CPPs as molecular transporters for macromolecules. In fact, many studies have reported multiple efficient CPPs for macromolecule delivery, sharing the feature of the presence of arginine in the sequence [40][41][42][43][44]. In addition to [WR]4 [37], [WR]5 [38], and [WR]9 [39], we have also reported a number of cyclic peptides and hybrid cyclic-linear peptides, [WWRR]4 and [WWRR]5 [45], [(RW)5K](RW)5 [46], [R5K]W7 [47,48], and [DipR]5 [49] (Figure 1) as molecular transporters of phosphopeptides, nucleic acids, or small molecules.  Herein, we aimed to investigate the potential of cyclic and cyclic-linear CPPs as delivery agents for proteins. This study is distinct from the previous strategies [34][35][36], focusing on the non-covalent delivery of several proteins using various peptide delivery tools containing R and hydrophobic residues. Figure 1 depicts the chemical structures of selected peptides used herein for evaluating the delivery of proteins. We hypothesized that the positively charged R and hydrophobic residues could have electrostatic and hydrophobic interactions with negatively charged phosphate in the phospholipid bilayer and with charged/hydrophobic residues on the protein leading to more internalization. Different protein models, GFP, red fluorescent protein (RFP), and a clinically relevant protein (Histone H2A) were utilized to assess the ability of peptides to deliver those macromolecules intracellularly. This study expands our knowledge about the ability of cyclic CPPs to efficiently transport functional proteins into cells.

Cytotoxicity of Peptide-Protein Physical Mixture
Cytotoxicity of the peptides has been previously reported by us [38,39,[45][46][47][48][49]. For example, [(RW) 5 K](RW) 5 did not exhibit any significant cytotoxicity in different cell lines, such as human epithelial mammary gland adenocarcinoma cells (MDA-MB-231), human leukemia carcinoma cell line (CCRF-CEM), human ovarian adenocarcinoma cells (SK-OV-3), and human epithelial embryonic kidney healthy (HEK-293) at the concentration of 10 µM after 3 h of incubation [46]. As a representative example, the cytotoxicity of [WR] 9 with proteins was examined at different concentrations. Fluorescently labeled proteins, such as GFP, RFP, and Cyan Fluorescent Protein (CFP), are used as cellular tags to visualize and track intracellular functions. However, numerous reports suggest that those tags exhibit cellular damage by different mechanisms, including apoptosis initiation, direct damage by reactive oxygen species (ROS) generation, and immunogenicity damage [50,51]. Therefore, to exclude the effect of possible toxicity on the interpretation of experimental data, [WR] 9 -protein (GFP/RFP) mixtures were evaluated for their toxic effect on MDA-MB-231 for 24 h. [WR] 9 and GFP or RFP were mixed and incubated together for 30 min before adding the cells at room temperature. The GFP/RFP concentration was (50 nM) during the experiment, and the peptide concentration ranged from (1 to 20 µM). The results showed concentration-dependent cytotoxicity. This cytotoxicity is presumably due to higher peptide concentrations in the mixture (15 and 20 µM). GFP alone did not exhibit any sign of cytotoxicity at 50 nM concentration ( Figure 2, Table S1 (Supporting Information)). Furthermore, RFP exhibited a similar pattern by revealing concentrationdependent cytotoxicity due to higher [WR] 9 concentrations, while RFP alone was not cytotoxic at 50 nM ( Figure 3, Table S2 (Supporting Information)). The cytotoxicity of cyclic peptides [WWRR] 4 and [WWRR] 5 with GFP (50 nM) is shown in Figure S1 and Table S3 (Supporting Information). Additional cytotoxicity studies found no cytotoxicity associated with the increase in GFP concentration alone (100 nM to 500 nM) and in a physical mixture with [WR] 9 (3 µM) ( Figure S2, Table S4 (Supporting Information)). The combination of [DipR] 5 (Figures 4 and S3, Tables S3 and S5, Supporting Information) with GFP (50 nM to 500 nM) after 3 h in MDA-MB-231 cells showed cell viability of more than 81%. Based on these findings and previous data about the cytotoxicity of other peptides [38,39,[45][46][47][48][49] used in this study, cellular uptake experiments were conducted with peptides and protein Pharmaceuticals 2023, 16, 469 5 of 26 models (GFP and RFP) concentrations not higher than 10 µM and 50 nM, respectively, and 3 h of incubation.
Pharmaceuticals 2023, 16, x FOR PEER REVIEW 4 and S3, Tables S3 and S5, Supporting Information) with GFP (50 nM to 500 nM) h in MDA-MB-231 cells showed cell viability of more than 81%. Based on these and previous data about the cytotoxicity of other peptides [38,39,[45][46][47][48][49] used in this cellular uptake experiments were conducted with peptides and protein models (GF RFP) concentrations not higher than 10 μM and 50 nM, respectively, and 3 h of incu

Cellular Internalization Using Confocal Microscopy
Our goal was to explore the ability of selected CPPs containing hydrophobic residues (W or Dip) and positively-charged arginine (R) residues to direct protein cargoes to localize within the cells, which was monitored using confocal microscopy. As a model protein, we chose GFP and RFP, 27 kDa uorescent proteins that are commonly used as uorescence markers in cell microscopy imaging, in a breast cancer cell line (MDA-MB-231). The cells were incubated with the physical mixture of proteins and selected peptides for 3 h in serum-free media, Opti-MEM, to reduce unnecessary interactions between the mixture and transfection media.
Using GFP, we rst studied the optimal time and concentration needed for [WR]9 to form proper interactions with the cargo that could lead to internalization into cells. By using (1:1, 1:8, 1:20, 1:40, and 1:60 of GFP:[WR]9), we showed that the peptide was able to internalize GFP protein in a concentration-dependent manner ( Figure 5). An amount of 50 nM-2 μM of peptide did not show any sign of uptake for GFP inside the cells, and only 3 μM of [WR]9 showed cytosolic delivery. Based on this nding, higher [WR]9

Cellular Internalization Using Confocal Microscopy
Our goal was to explore the ability of selected CPPs containing hydrophobic residues (W or Dip) and positively-charged arginine (R) residues to direct protein cargoes to localize within the cells, which was monitored using confocal microscopy. As a model protein, we chose GFP and RFP, 27 kDa fluorescent proteins that are commonly used as fluorescence markers in cell microscopy imaging, in a breast cancer cell line (MDA-MB-231). The cells were incubated with the physical mixture of proteins and selected peptides for 3 h in serum-free media, Opti-MEM, to reduce unnecessary interactions between the mixture and transfection media.
Using GFP, we first studied the optimal time and concentration needed for [WR] 9 to form proper interactions with the cargo that could lead to internalization into cells. By using (1:1, 1:8, 1:20, 1:40, and 1:60 of GFP:[WR] 9 ), we showed that the peptide was able to internalize GFP protein in a concentration-dependent manner ( Figure 5). An amount of 50 nM-2 µM of peptide did not show any sign of uptake for GFP inside the cells, and only 3 µM of [WR] 9 showed cytosolic delivery. Based on this finding, higher [WR] 9 concentrations (4, 5, 6, 8, and 10 µM) were used with 50 nM GFP in another confocal microscopy experiment to examine the effect of increasing peptide concentrations on GFP Pharmaceuticals 2023, 16, 469 8 of 26 cellular internalization. The results showed that at concentrations of 3-10 µM, the uptake of GFP increased intracellularly. However, we noticed cell morphology changed when 10 µM of [WR] 9 was used, as shown in Figure 6. We have previously shown that [WR] 9 had minimal toxicity at a concentration of 10 µM in SK-OV-3 and MDA-MB-231 after 24 h of incubation [52]. Herein, as shown in Figures 2 and 3 (Tables S1 and S2), the cell viability was ≥90%. Slight morphological changes are presumably due to the minimal toxicity of this compound at this concentration. Since a significant uptake of the proteins was also observed at concentrations less than 10 µM, this slight morphological change did not affect the uptake of the proteins. cellular internalization. The results showed that at concentrations of 3-10 µM, the uptake of GFP increased intracellularly. However, we noticed cell morphology changed when 10 µM of [WR]9 was used, as shown in Figure 6. We have previously shown that [WR]9 had minimal toxicity at a concentration of 10 µM in SK-OV-3 and MDA-MB-231 after 24 h of incubation [52]. Herein, as shown in Figures 2 and 3 (Tables S1 and S2), the cell viability was ≥90%. Slight morphological changes are presumably due to the minimal toxicity of this compound at this concentration. Since a significant uptake of the proteins was also observed at concentrations less than 10 µM, this slight morphological change did not affect the uptake of the proteins.
We also conducted a time-dependent study to evaluate the time needed for [WR]9 to deliver GFP inside cells. The time points selected were (5,10,20,30, and 60 min), and the experiment was performed according to the protocol described above using GFP (50 nM) + [WR]9 (4 µM). Indeed, the uptake appeared to be time-dependent, and we were able to detect a low signal for GFP uptake as early as 5 min of incubation with the cells, but the intensity of the GFP signal was at the maximum at 30 and 60 min in the cytosol of MDA-MB-231 cells (Figure 7).    We also conducted a time-dependent study to evaluate the time needed for [WR] 9 to deliver GFP inside cells. The time points selected were (5, 10, 20, 30, and 60 min), and the experiment was performed according to the protocol described above using GFP (50 nM) + [WR] 9 (4 µM). Indeed, the uptake appeared to be time-dependent, and we were able to detect a low signal for GFP uptake as early as 5 min of incubation with the cells, but the intensity of the GFP signal was at the maximum at 30 Figure S4, Supporting Information), suggesting that the uptake for this peptide was cell-dependent, possibly due to the differential nature of interaction with the cell membrane in cancer cells that have different membrane lipid compositions.
We have previously shown that [WR] 4 enhanced the cellular uptake of cell-impermeable cargo molecules even in the presence of different endocytosis inhibitors [37], indicating that the mechanism of cellular uptake is not dependent exclusively on endocytosis. For [WR] 4 , the direct membrane translocation mainly occurs via an energy-independent pathway and through the electrostatic interactions of positively charged R residues and negatively charged phospholipid bilayer. Bypassing the entrapment in endosomes has significant clinical applications since the cargo molecules can be available immediately for biological activity without the need to be released from endosomes.
All peptides were able to deliver GFP to the cells. Among all the peptides, [DipR] 5 (1-10 µM) (Figures 8 and 9) and [WR] 9 (3-10 µM) (Figures 5 and 6) showed to be the most efficient peptides in GFP delivery to the cytosol based on the intensity of fluorescence signals, as shown by confocal microscopy. These data suggest that the nature of hydrophobic residues and the number of arginine residues in the peptide significantly affect the GFP delivery since the presence of Dip in [DipR] 5  Information) when compared with MDA-MB-231 cells ( Figure S4, Supporting Information), suggesting that the uptake for this peptide was cell-dependent, possibly due to the differential nature of interaction with the cell membrane in cancer cells that have different membrane lipid compositions.
We have previously shown that [WR]4 enhanced the cellular uptake of cell-impermeable cargo molecules even in the presence of different endocytosis inhibitors [37], indicating that the mechanism of cellular uptake is not dependent exclusively on endocytosis. For [WR]4, the direct membrane translocation mainly occurs via an energy-independent pathway and through the electrostatic interactions of positively charged R residues and negatively charged phospholipid bilayer. Bypassing the entrapment in endosomes has significant clinical applications since the cargo molecules can be available immediately for biological activity without the need to be released from endosomes.
All peptides were able to deliver GFP to the cells. Among all the peptides, [DipR]5 (1-10 µM) (Figures 8 and 9) and [WR]9 (3-10 µM) ( Figures 5 and 6) showed to be the most efficient peptides in GFP delivery to the cytosol based on the intensity of fluorescence signals, as shown by confocal microscopy. These data suggest that the nature of hydrophobic residues and the number of arginine residues in the peptide significantly affect the GFP delivery since the presence of Dip in [DipR]5 and a higher number of R residues in [WR]9 versus [WR]4 significantly enhanced the GFP uptake.  Based on the confocal microscopy data for efficient delivery of GFP by [WR]9 and [DipR]5, these two peptides were selected for further studies. Another protein model was also used to confirm the ability of [WR]9 to deliver full-length functional protein. Red fluorescence protein (RFP), a 27 kDa fluorescent protein that is commonly used as a fluorescence marker in cell microscopy imaging, was used with the peptide in a breast cancer cell line (MDA-MB-231). Taking advantage of our findings using GFP, cells were incubated with the peptide at concentrations (3-8 µM) and RFP at (50 nM). The results were consistent with GFP confocal images, and all tested peptide concentrations showed cytosolic delivery for RFP. Higher [WR]9 concentrations demonstrated increased uptake of RFP (as shown by the red signal of RFP (Figure 10)). GFP and RFP uptake findings proved the ability of cyclic [WR]9 as a potential delivery agent for macromolecules with a simple mixing approach and without the need for chemical conjugation or fusion of peptide sequence during protein expression. Based on the confocal microscopy data for efficient delivery of GFP by [WR] 9 and [DipR] 5 , these two peptides were selected for further studies. Another protein model was also used to confirm the ability of [WR] 9 to deliver full-length functional protein.
Red fluorescence protein (RFP), a 27 kDa fluorescent protein that is commonly used as a fluorescence marker in cell microscopy imaging, was used with the peptide in a breast cancer cell line (MDA-MB-231). Taking advantage of our findings using GFP, cells were incubated with the peptide at concentrations (3-8 µM) and RFP at (50 nM). The results were consistent with GFP confocal images, and all tested peptide concentrations showed cytosolic delivery for RFP. Higher [WR] 9 concentrations demonstrated increased uptake of RFP (as shown by the red signal of RFP (Figure 10)). GFP and RFP uptake findings proved the ability of cyclic [WR] 9 as a potential delivery agent for macromolecules with a simple mixing approach and without the need for chemical conjugation or fusion of peptide sequence during protein expression.  As observed for GFP delivery in the presence of [DipR]5, this peptide (1-10 µM) was also found to be very efficient in RFP (50 nM) delivery in both MDA-MB-231 and SK-OV-3 cells (Figures 11 and 12). The localization of RFP was found to be in the cytosol, similar As observed for GFP delivery in the presence of [DipR] 5 , this peptide (1-10 µM) was also found to be very efficient in RFP (50 nM) delivery in both MDA-MB-231 and SK-OV-3 cells (Figures 11 and 12). The localization of RFP was found to be in the cytosol, similar to  To further explore the efficiency of [WR]9 in protein delivery, we also used fluorescent-labeled histone H2A. Histone proteins are involved in packaging DNA into nucleosomes. Histones maintain the shape and structure of a nucleosome [53,54]. Histone H2A is among the five main histone proteins responsible for the structure of chromatin in eukaryotic cells. Histones have been proposed as antimicrobial agents via the mediation of neutrophil extracellular traps (NETs), a first-line defense against many microorganisms [55][56][57][58][59]. Moreover, H2A disrupts DNA organization and stops the transcription of microorganisms upon cellular entry [60]. However, due to the size and hydrophilic/phobic nature of histones, they pose low cellular uptake. Regardless of cellular composition and differences between cancer and bacterial cells, here we aimed to evaluate whether [WR]9 was able to deliver fluorescent-labeled histone H2A. Cells were incubated with the peptide at concentrations of 3-8 µM and fluorescentlabeled histone H2A at (50 nM). Confocal images revealed that most of the uptake occurred with peptide concentrations of 4, 5, and 6 µM, and the delivery was mainly in the cytoplasm ( Figure 13).
These results open new venues for applying cyclic CPPs for the delivery of therapeutic proteins. At the same time, these avenues need to be explored further in protein uptake. Our laboratory is currently exploring other lead peptides for protein delivery applications. To further explore the efficiency of [WR] 9 in protein delivery, we also used fluorescentlabeled histone H2A. Histone proteins are involved in packaging DNA into nucleosomes. Histones maintain the shape and structure of a nucleosome [53,54]. Histone H2A is among the five main histone proteins responsible for the structure of chromatin in eukaryotic cells. Histones have been proposed as antimicrobial agents via the mediation of neutrophil extracellular traps (NETs), a first-line defense against many microorganisms [55][56][57][58][59]. Moreover, H2A disrupts DNA organization and stops the transcription of microorganisms upon cellular entry [60]. However, due to the size and hydrophilic/phobic nature of histones, they pose low cellular uptake.
Regardless of cellular composition and differences between cancer and bacterial cells, here we aimed to evaluate whether [WR] 9 was able to deliver fluorescent-labeled histone H2A. Cells were incubated with the peptide at concentrations of 3-8 µM and fluorescentlabeled histone H2A at (50 nM). Confocal images revealed that most of the uptake occurred with peptide concentrations of 4, 5, and 6 µM, and the delivery was mainly in the cytoplasm ( Figure 13).  We have previously reported the synthesis and localization of fluorescence-labeled [KDipR]5 (F′-[DipR]5, Figure 1) in the cytosol [49]. We explored to determine whether the fluorescence signal for this peptide coalesces with the red fluorescence signal of RFP after incubation with MDA-MB-231 and SK-OV-3 cells. The data showed that both green and red fluorescence signals from fluorescence-labeled [KDipR]5 and RFP merge as yellow, suggesting colocalization of the peptide and protein and that the peptide indeed carries the protein to the cytosol (Figures 14 and 15). These results open new venues for applying cyclic CPPs for the delivery of therapeutic proteins. At the same time, these avenues need to be explored further in protein uptake. Our laboratory is currently exploring other lead peptides for protein delivery applications.
We have previously reported the synthesis and localization of fluorescence-labeled [KDipR] 5 (F -[DipR] 5 , Figure 1) in the cytosol [49]. We explored to determine whether the fluorescence signal for this peptide coalesces with the red fluorescence signal of RFP after incubation with MDA-MB-231 and SK-OV-3 cells. The data showed that both green and red fluorescence signals from fluorescence-labeled [KDipR] 5 and RFP merge as yellow, suggesting colocalization of the peptide and protein and that the peptide indeed carries the protein to the cytosol (Figures 14 and 15).

Fluorescent-Assisted Cell Sorting (FACS)
We used FACS to quantify GFP uptake using MDA-MB-231 cell line. Cells were incubated with [WR]9 at concentrations (3-10 µM) and GFP at (50 nM), as described above for 3 h. The results showed significant uptake (p-value < 0.001) with all peptide

Fluorescent-Assisted Cell Sorting (FACS)
We used FACS to quantify GFP uptake using MDA-MB-231 cell line. Cells were incubated with [WR]9 at concentrations (3-10 µM) and GFP at (50 nM), as described above for 3 h. The results showed significant uptake (p-value < 0.001) with all peptide

Fluorescent-Assisted Cell Sorting (FACS)
We used FACS to quantify GFP uptake using MDA-MB-231 cell line. Cells were incubated with [WR] 9 at concentrations (3-10 µM) and GFP at (50 nM), as described above for 3 h. The results showed significant uptake (p-value < 0.001) with all peptide concentrations used. However, 3 and 4 µM of [WR] 9 showed significantly higher GFP uptake compared to [WR] 9 concentrations (5-10 µM). When compared to GFP alone, 3 and 4 µM of [WR] 9 showed a 11-and 9.5-fold higher uptake, respectively, while [WR] 9 (5-10 µM) showed a range of 6-to 3-fold increase in the uptake of GFP ( Figure 16). These data were consistent with the confocal microscopy data shown in Figures 5 and 6, with a high GFP uptake at concentrations of 3-4 µM. Although the uptake was also observed at higher concentrations, the cytotoxicity of [WR] 9 was a limiting factor, as shown by cell morphological changes at 10 µM. . When compared to GFP alone, 3 and 4 μM of [WR]9 showed a 11-and 9.5-fold higher uptake, respectively, while [WR]9 (5-10 μM) showed a range of 6-to 3-fold increase in the uptake of GFP ( Figure 16). These data were consistent with the confocal microscopy data shown in Figures 5 and 6, with a high GFP uptake at concentrations of 3-4 μM. Although the uptake was also observed at higher concentrations, the cytotoxicity of [WR]9 was a limiting factor, as shown by cell morphological changes at 10 μM.  Figure 17). The data correlate well with confocal microscopy data for [DipR]5 with GFP (Figures 8 and 9) and RFP (Figures 11 and 12). As shown in confocal microscopy (Figure 15), F -[DipR]5 enhanced the uptake of RFP. FACS studies were consistent with confocal microscopy ( Figure S10, Supporting Information).  5 enhanced the uptake of both GFP and RFP, and the uptake was concentration-dependent, with the highest uptake at 10 µM ( Figure 17). The data correlate well with confocal microscopy data for [DipR] 5 with GFP (Figures 8 and 9) and RFP (Figures 11 and 12). As Mechanistic studies were performed to determine whether the cellular uptake of the GFP in the presence of [WR]9 or [DipR]5 is endocytosis-dependent. These studies were conducted by confocal microscopy ( Figure S11 (Figures 18 and 19).
signi cantly by chlorpromazine, which inhibits clathrin-mediated endocytosis. Chloroquine (30 μM) and nystatin (50 μg/mL) decreased the uptake of GFP (50 nM) and [WR]9 (3 μM) by 21 and 32%, respectively. None of the endocytosis inhibitors could completely stop the cellular uptake; however, a signi cant reduction in uptake was in the presence of methyl--cyclodextrin, which disrupts the lipid rafts of the cell membrane. The uptake of GFP (50 nM) in the presence of [WR]9 (3 μM) was slightly inhibited by methyl--cyclodextrin and nystatin endocytosis inhibitors after 3 h of incubation in MDA-MB-231 cells (Figure 18), suggesting the partial uptake through the caveolae/lipid-mediated endocytosis pathway. As a result, the combination of direct penetration and caveolae/lipid-mediated endocytosis may be involved in the uptake of GFP (50 nM) in the presence of [WR]9 (3 μM) across the cell membrane. The cellular uptake of GFP (50 nM) in the presence of [DipR]5 (10 μM) was signicantly reduced by all inhibitors. However, chloroquine, a clathrin-mediated endocytosis inhibitor, reduced it the least. Methyl--cyclodextrin (2.25 nM) and chlorpromazine (30 μM) decreased the uptake of GFP (50 nM) in the presence of [DipR]5 (10 μM) by 17 and 26%, respectively. Thus, the uptake of GFP in the presence of [DipR]5 was di erent from that of [WR]9. Even though none of the endocytosis inhibitors could completely stop the cellular uptake, a major reduction in uptake was observed in the presence of nystatin, which disrupts the caveolae/lipid-mediated endocytosis. The uptake of GFP (50 nM) and [DipR]5 (10 μM) was partially inhibited by nystatin and chlorpromazine after 3 h of incubation in MDA-MB-231 cells (Figure 19), suggesting a combination of multifaceted mechanisms. As a result, a combination of direct penetration and various endocytic processes

Materials
All protected amino acids and resins were purchased from AAPPTEC (Louisville, KY, USA). All the other chemical reagents were purchased from Millipore Sigma (Milwau- The cellular uptake of GFP (50 nM) in the presence of [WR] 9 (3 µM) was not reduced significantly by chlorpromazine, which inhibits clathrin-mediated endocytosis. Chloroquine (30 µM) and nystatin (50 µg/mL) decreased the uptake of GFP (50 nM) and [WR] 9 (3 µM) by 21 and 32%, respectively. None of the endocytosis inhibitors could completely stop the cellular uptake; however, a significant reduction in uptake was in the presence of methyl-β-cyclodextrin, which disrupts the lipid rafts of the cell membrane. The uptake of GFP (50 nM) in the presence of [WR] 9 (3 µM) was slightly inhibited by methyl-βcyclodextrin and nystatin endocytosis inhibitors after 3 h of incubation in MDA-MB-231 cells (Figure 18), suggesting the partial uptake through the caveolae/lipid-mediated endocytosis pathway. As a result, the combination of direct penetration and caveolae/lipidmediated endocytosis may be involved in the uptake of GFP (50 nM) in the presence of [WR] 9 (3 µM) across the cell membrane.
The cellular uptake of GFP (50 nM) in the presence of [DipR] 5 (10 µM) was significantly reduced by all inhibitors. However, chloroquine, a clathrin-mediated endocytosis inhibitor, reduced it the least. Methyl-β-cyclodextrin (2.25 nM) and chlorpromazine (30 µM) decreased the uptake of GFP (50 nM) in the presence of [DipR] 5 (10 µM) by 17 and 26%, respectively. Thus, the uptake of GFP in the presence of [DipR] 5 was different from that of [WR] 9 . Even though none of the endocytosis inhibitors could completely stop the cellular uptake, a major reduction in uptake was observed in the presence of nystatin, which disrupts the caveolae/lipid-mediated endocytosis. The uptake of GFP (50 nM) and [DipR] 5 (10 µM) was partially inhibited by nystatin and chlorpromazine after 3 h of incubation in MDA-MB-231 cells (Figure 19), suggesting a combination of multifaceted mechanisms. As a result, a combination of direct penetration and various endocytic processes presumably contribute to the uptake of GFP (50 nM) in the presence of [DipR] 5 (10 µM) across the cell membrane.

Materials
All protected amino acids and resins were purchased from AAPPTEC (Louisville, KY, USA). All the other chemical reagents were purchased from Millipore Sigma (Milwaukee, WI, USA). Recombinant GFP and RFP were obtained from Novus Biologicals (Centennial, CO, USA).
[ The cytotoxic activity of [WR] 9 + GFP/RFP physical mixtures was evaluated in MDA-MB-231 cells, according to the previously reported procedure [38]. In brief, the cells were seeded at 5000 cells (0.1 mL per well in 96-well plates). An appropriate growth medium was used for each cell line (for MDA-MB-231 DMEM/F12 (1:1) (1x) with L-Glutamine and 15 mM HEPES containing FBS (10%) and penicillin or streptomycin (1%)). The cells were seeded in a complete growth medium 24 h prior to the experiment. The protein concentration was fixed at (50 nM). Different peptide concentrations ranging from 1 to 20 µM were added to each well in triplicate. Then, the cells were incubated for 24 h at 37 • C in a humidified atmosphere of 5% CO 2 . After the incubation period, MTS reagent (20 µL) was added to each well. The incubation was continued for 3 h. The MTS protocol is based on the reduction reaction of MTS tetrazolium by the viable cells. Cell viability was then measured by the determination of the fluorescence intensity at 490 nm using a SpectraMax M2 microplate spectrophotometer. The percentage of cell viability was then calculated using the following equation: [(OD value of cells treated with the compound) − (OD value of culture medium)]/[(OD value of control cells) − (OD value of culture medium)] × 100%.

Confocal Microscopy
MDA-MB-231 and SK-OV-3 cells (7 × 10 4 cells/well) were seeded with a medium on a coverslip 24 h prior to the experiment in 6-well plates. After 24 h, the medium was changed, and cells were treated with protein-peptide physical mixtures using the method explained for flow cytometry experiments (below). Peptide/protein physical mixtures were prepared using serum-free media as a diluent to achieve a final concentration of 50 nM for GFP, RFP, and histone H2A proteins and different peptide concentrations. Cells exposed to the mixtures were incubated at 37 ºC and standard growth conditions for 3 h. After 3 h, the growth media was removed, and cells were washed with clear HBSS. The cells were then fixed using 3.7% formaldehyde solution in HBSS for about 10 min and were then exposed to Texas Red for 60 min and DAPI overnight, away from light, at room temperature. Texas red and DAPI are used to stain cell membranes and nuclei, respectively. Coverslips were examined using a Nikon A1R high-definition resonant scanning confocal microscope and NIS-Elements software (AR 4.30.02, 64 bit).

Cellular Internalization and Mechanistic Studies (Flow Cytometry)
The efficiency of the peptides in internalizing model protein (GFP) into human cells was evaluated by quantifying the uptake of fluorescence by flow cytometry (BD-FACSVerse; BD Biosciences; San Jose, CA, USA). MDA-MB-231 and SK-OV-3 cells were used for this study and were seeded in 24-well plates (~200,000 cells per well). The peptide/protein mixtures were prepared with a final concentration of 50 nM for GFP or RFP protein and different peptide concentrations. Cells exposed to the mixtures were incubated at 37 • C and standard growth conditions for 3 h. Then, the cells were washed with clear HBSS, trypsinized, and fixed using 3.7% formaldehyde solution. Suspended cells were analyzed using the FITC channel to quantify cell-associated fluorescence for GFP or RFP. The uptake analysis was processed according to the calibration of the signal gated from the non-treated cells (as the negative control). The mixture was incubated for 30 min before any assays to obtain a satisfactory binding.
Mechanistic studies were performed in MDA-MB-231 cells to determine the mechanism of the cellular uptake of GFP in the presence of [WR] 9 or [DipR] 5 . The cells were incubated with various endocytosis inhibitors, including chlorpromazine (30 µM), chloroquine (100 µM), methyl-β-cyclodextrin (2.25 mM), and nystatin (50 µg/mL) for 30 min. Simultaneously, the GFP and peptide physical mixtures were left for 30 min. The cells were then incubated with GFP and [WR] 9 or [DipR] 5 for 3 h. The flow cytometry study was performed as mentioned above.

Data Analysis
The data are presented as the mean standard deviation for the stated number of samples. A significant difference test was performed using a Student's t-test between the two groups. For data over 3 groups, one-way ANOVA followed by post hoc Tukey tests were performed. The alpha threshold was set to 0.05 with a 95% confidence interval.

Conclusions
In a continuous effort to explore cyclic CPPs applications, our laboratory investigated the ability of a number of cyclic and hybrid cyclic-linear peptides, [ 9 was also shown to improve the delivery of histone H2A. The mechanistic studies revealed that a combination of mechanisms contributes to the uptake of GFP (50 nM) in the presence of [WR] 9 (3 µM) or [DipR] 5 (10 µM) across the cell membrane. These results open new avenues for the application of CPPs for the delivery of therapeutic proteins. While these avenues need to be explored further to determine the function of delivered proteins, our laboratory is currently exploring other lead peptides for protein delivery applications. We have already conducted CPP-drug conjugates that showed the effectiveness of the covalent strategy for small drug molecules, such as doxorubicin [47,52,61,62], camptothecin, and paclitaxel [63]. Thus, future steps include applying chemical conjugation of lead peptides with proteins of therapeutic relevance. Further studies are needed to assess their potential as a non-toxic protein delivery agent.  5  Author Contributions: Conceptualization, K.P., R.K.T., J.M. and K.Z., investigation, K.P., J.M., K.Z., D.S., L.K. and S.K.S., writing-original draft preparation, K.P., J.M., K.Z., K.P. and R.K.T., writing-review and editing, K.P., R.K.T., J.M. and K.Z. All authors have read and agreed to the published version of the manuscript.

Funding:
The authors acknowledge financial support from Chapman University School of Pharmacy. We acknowledge Albert Siryaporn from the University of California Irvine for providing Histone 2A.

Institutional Review Board Statement: Not applicable.
Informed Consent Statement: Not applicable.

Data Availability Statement:
The data presented in this study are available on request from the corresponding author.