Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2013 Jul 27.
Published in final edited form as: Chem Biol. 2012 Jul 27;19(7):819–830. doi: 10.1016/j.chembiol.2012.05.022

Arginine topology controls escape of miniature proteins from early endosomes to the cytoplasm

Jacob S Appelbaum 1, Jonathan R LaRochelle 2, Betsy A Smith 2,4, Daniel M Balkin 1, Justin M Holub 2, Alanna Schepartz 2,3,*
PMCID: PMC3488872  NIHMSID: NIHMS386997  PMID: 22840770

Introduction

Protein drugs represent a rapidly expanding class of therapeutic molecules (Strohl and Knight, 2009). However, their actions are limited primarily to extracellular targets (Fischer, 2007; Johnson et al., 2011) because the size and composition of most polypeptides and peptide mimetics do not facilitate their uptake into mammalian cells (Luedtke et al., 2003). It has been known for over 40 years that addition of cationic charges to a peptide or protein can aid transport into cells (Ryser and Hancock, 1965), and many reports have demonstrated the utility of appending basic sequences derived from the HIV Trans-Activator of Transcription (Tat) (Zhou et al., 2009), D. melanogaster Antennapedia (Théodore et al., 1995), or simply polyarginine (for example, Arg8) (Futaki et al., 2001) to peptides or small molecules (Wender et al., 2008) to increase their cytoplasmic access. Certain highly positively charged proteins (Cronican et al., 2011) and toxins (Johannes and Popoff, 2008) also possess cell penetrating properties, but it remains unknown how one can leverage these examples to control the precise entry pathway or enhance the uptake of designed peptides.

Two contrasting mechanisms have been proposed for the cytosolic entry of cationic proteins and related molecules. The first (ion pair-guided passive diffusion) posits that guanidinium side chains on the polypeptide form hydrogen bonds with cell surface phospholipids creating neutral ion pairs that passively diffuse across the plasma membrane (Rothbard et al., 2005). The second model (endosomal release), asserts that endocytosis is a major portal through which cationic polypeptides and peptide mimetics enter the cell (Fischer, 2007). Previous investigations have attempted to distinguish between these two models by blocking endocytosis, via thermal (Derossi et al., 1996), pharmacologic (Wadia et al., 2004; Fischer et al., 2004), or genetic means (Ter-Avetisyan et al., 2008). The interpretation of these experiments is complicated, however, by differences in protein/polypeptide concentration and analytical method. For example, incubation of living cells with cationic proteins/polypeptides at concentrations ≥ 10 μM leads to the formation of nucleation zones (Duchardt et al., 2007) that transiently disrupt membranes (Palm-Apergi et al., 2009), causing the spontaneous release of peptide into the cytosol. Incubation of cells at lower concentrations (≤ 5 μM) of peptide, in the presence of drugs that inhibit endocytosis, prevents cytoplasmic access (Wadia et al., 2004), implying that at low concentrations, the molecules studied cannot diffuse through the plasma membrane. Moreover, the many studies using microscopy to examine cells fixed by treatment with formaldehyde or methanol must be reevaluated in light of evidence that the fixation process can release fluorescently labeled peptides from endosomes (Belitsky et al., 2002 and Richard et al., 2003), an artifact not observed during microscopic examination of living cells. Finally, the high intensity light used during microscopy can itself facilitate the redistribution of fluorescently labeled peptides from endosomes to cytoplasm (Maiolo et al., 2004). Thus, whether, when, and how these cationic molecules escape endocytic vesicles to access the cytosol remain unanswered questions.

Attempts to identify structural determinants of cell permeability are complicated by the above experimental details as well as the fact that neither Tat nor Arg8 possesses a defined fold. Miniature proteins are a family of small (36-aa), well-folded polypeptides that adopt a characteristic hairpin fold consisting of axially packed α- and PPII helices (Blundell et al., 1981; Hodges and Schepartz, 2007). Miniature proteins identified through both rational design (Zondlo and Schepartz, 1999; Zellefrow et al., 2006) and molecular evolution (Chin and Schepartz, 2001; Rutledge et al., 2003; Golemi-Kotra et al., 2004; Gemperli et al., 2005) can modulate protein function by inhibiting protein interactions (Rutledge et al., 2003; Gemperli et al., 2005); both loss of function and gain of function activities have been observed (Golemi-Kotra et al., 2004; Gemperli et al., 2005; Zellefrow et al., 2006). We reported previously that minimally cationic miniature proteins containing between 2 and 6 arginine residues embedded within the α- or PPII helix were taken up by mammalian cells in culture more efficiently than Tat or Arg8 (Daniels and Schepartz, 2007; Smith et al., 2008). In this report we investigate whether, when, and how miniature proteins containing arginine access the cytoplasm.

To learn more about the structural determinants of cytoplasmic access, we designed a set of miniature proteins that differed in the number and density of α-helical arginine side chains, and tracked their passage into the cell. Using low concentrations (1 μM) of fluorophore-conjugated variants, we found that a minimum of 4 α-helical arginines was required for uptake, and that cell uptake was enhanced when the arginines were clustered on the same α-helix face. Next, a novel and rapid assay for evaluating cytoplasmic access revealed that of four cationic miniature proteins taken up by cells, only one reaches the cytosol. This miniature protein, which we named 5.3, possesses a distinct array of five dispersed α-helical arginines. Live cell confocal microscopy revealed that fluorophore-labeled 5.3 (5.3R) is taken up by an endocytic pathway that includes Rab5+ and Rab7+ endosomes. This pathway is shared by Tat, Arg8, and cationic miniature proteins that do not reach the cytosol. However, for 5.3, Tat or Arg8 to gain cytosolic access, active endocytosis and endosomal acidification were required. These data indicate that none of the molecules evaluated could directly cross the plasma membrane; rather, they reach the cytoplasm by escaping from intracellular vesicles. We find further that unlike Tat or Arg8, which require vesicle maturation beyond the Rab5+ stage, miniature protein 5.3 accesses the cytosol by crossing specific membrane regions present in early Rab5+ endosomes. Finally, grafting the arginine array present in 5.3 into a well-folded zinc finger domain (Krizek et al., 1991) successfully endowed this domain with the trafficking properties of 5.3. These experiments demonstrate that discrete arginine arrangements embedded within a well-folded miniature protein can direct cytosolic access by facilitating both efficient uptake and early endosomal escape.

Results

Miniature Protein Design

To examine the effect of charge density and orientation on cell uptake, we prepared eight miniature proteins containing between one and six arginine residues at various positions on the solvent-exposed α-helical surface of the hairpin fold. These molecules also contained two arginines near the C-terminus (Figure 1A). Seven of these cationic miniature proteins were characterized by circular dichroism (CD) spectra at 37°C that were virtually indistinguishable from that of the parent molecule lacking additional arginines, aPP (Figure S1A). The CD spectra of six were temperature-dependent with cooperative transitions between 49 and 67°C (Figure S1B,C) suggesting that they each retained a stable and characteristic hairpin fold (Hodges and Schepartz, 2007). Miniature protein 6.3, containing the greatest number of arginine substitutions (6), showed reduced ellipticity at 222 nm and 208 nm (Figure S1A), along with a reduced Tm of 33°C (Figure S1B,C); 6.3 was not studied further. For the remaining molecules, these CD data suggest that arginine substitution does not significantly alter miniature protein secondary structure, and that Figure 1B accurately represents the arginine side chain arrangement in miniature proteins 2.1, 2.2, 3.2, 4.2, 4.3, 5.2, and 5.3.

Figure 1.

Figure 1

Open in a new tab

Miniature protein design. (A) Sequences of cationic miniature proteins evaluated in this work. (B) A plot of the relationship between #Rtotal (the number of (-helical arginine residues) and #Rfaces (the number of α-helical faces on which these arginines are displayed) for each aPP variant. The location of each α–helical arginine residue is represented by a blue circle on the helical wheel. (C) A plot of mean cellular fluorescence at 530 nm of HeLa cells treated with fluorescently tagged miniature protein variants (5 μM, 30 min). See also Figure S1.

Cell Uptake

Initially we used flow cytometry to assess the influence of arginine number and orientation on miniature protein uptake. In preliminary studies, we evaluated molecules labeled with fluorescein on their C-termini (as denoted with superscript F). To ensure that our experiment measured miniature proteins that had entered cells, we included a trypsin wash just before analysis via flow cytometry. Treatment of HeLa cells with 5 μM aPPF, 2.1F, 2.2F, or 3.2F resulted in only small increases (< 3 fold) in cell fluorescence, while treatment with 4.2F, 4.3F, 5.2F, or 5.3F resulted in increases in cell fluorescence between 7 and 40 fold (Figure 1C). We therefore chose to focus on miniature proteins showing significant uptake, and confirmed these results by synthesizing analogs labeled instead with tetraethyl rhodamine sulfate (denoted with superscript R), a dye with several desirable properties including resistance to photobleaching and an emission spectrum unaffected by pH changes and far from the autofluorescence spectrum of cells (Fernández-Suárez and Ting, 2008). As found for miniature proteins labeled with fluorescein, rhodamine labeled miniature proteins containing four or five α-helical arginines were taken up efficiently, in some cases (5.2R and 5.3R) more efficiently than TatR or Arg8R (Figure 2A).

Figure 2.

Figure 2

Open in a new tab

Arginine topology controls cell binding and uptake. (A) Surface binding of rhodamine labeled cationic miniature proteins in the absence of endocytosis and removal by trypsin treatment. HeLa cells were treated with 1 μM rhodamine labeled cationic miniature protein for 30 min. Cell were then treated with trypsin (0.05%, 10 min, 37°C) or PBS before washing and analysis by flow cytometry. (B) Fraction of cell-associated fluorescence remaining after trypsin treatment. These data represent the ratio of black to red bars shown in A. (C) Cell uptake of rhodamine labeled peptides by HeLa cells after 30 min and 90 min.AlexaFluor-488-transferrin (Tf488) when added to HeLa cells (D) colocalizes with (E) Tf546 and (F) rhodamine labeled miniature proteins. Perfect colocalization is characterized by a Pearson's R value (R) equal to 1, while R values near 0 represent little or no colocalization. The correlation value observed when cells were treated with both Tf488 and alexa-fluor-546-labeled transferrin (Tf46) is 0.905. (G) When added with Tf488, aPPR shows little intracellular signal. Rhodamine labeled cationic miniature proteins and Tf546 are shown in red, Tf488 is shown in green, Hoescht (nucleus) is shown in blue. See also Figure S2.

Because previous studies have shown that cell-penetrating peptides, including Tat and Arg8, bind to cell surface proteoglycans (Payne et al., 2007), we verified that trypsin treatment removed miniature proteins that were bound to the cell surface. Peptides and proteins that enter the cell are inaccessible to trypsin added to the culture media (Frankel and Pabo, 1988). Therefore, we arrested membrane traffic by incubating cells at 4°C (Hanover et al., 1984; Vonderheit and Helenius, 2005) for 15 min prior to and during a 30 min treatment with 1 μM 4.2R, 4.3R, 5.2R, 5.3R, TatR, Arg8R or aPPR. After incubation, cells were washed with PBS and incubated with 0.05% trypsin or PBS (as a control) before analysis by flow cytometry. These results were compared to those obtained when cells were incubated at 37°C and treated with trypsin (Figure 2A). Cationic miniature proteins 4.2R, 4.3R, 5.2R, and 5.3R bound to cells between 2.7-fold and 35-fold more than aPPR and in some cases (5.2R and 5.3R) to an extent comparable to TatR and Arg8R. For cells incubated at 4°C, trypsin treatment decreased the fluorescent signal between 77-89% (Figure 2B). These data confirm that at 4°C, incubation of cells with 1 μM cationic miniature protein leads to little if any cell uptake, and confirms that the vast majority of material bound to the cell surface is degraded and/or effectively removed by trypsin treatment. Comparison of the uptake of rhodamine and fluorescein labeled miniature proteins (Figure 2C) shows that cell uptake depends on arginine density: miniature proteins containing four arginine residues clustered on two helical faces were taken up more efficiently than those containing four arginines on three helical faces (Figure 2A,C and Figure 1C). Molecules containing five α-helical arginines were taken up to a similar extent irrespective of density, revealing that among these molecules the impact of arginine arrangement was smaller. Consistent with our previous work (Smith et al., 2008), the cationic miniature proteins 5.2R and 5.3R are taken up with an efficiency ≥ 2 fold better than that of Tat or Arg8, despite the fact that they possess twice the mass and fewer (7 rather than 8) positive charges.

Cationic miniature proteins traffic first into endocytic vesicles

To better understand the uptake pathway, we treated HeLa cells with rhodamine labeled miniature proteins in the presence of transferrin labeled with AlexaFluor-488 (Tf488) and quantified fluorescence overlap using confocal microscopy. Transferrin is rapidly internalized from the plasma membrane into endocytic vesicles (Hanover et al., 1984; Lakadamyali et al., 2006) and observing transferrin within rhodamine+ vesicles would suggest the vesicles were endocytic in nature, originating from the plasma membrane. We incubated HeLa cells for 30 min with Tf488 (25 nM) and 1 μM 4.2R, 4.3R, 5.2R, 5.3R, TatR or Arg8R before washing with media, staining with Hoescht (to visualize DNA), and imaging without fixation by confocal microscopy (Figure 2D-F). Cells treated with Tf488 showed small, discrete areas of intense green fluorescent signal. Treatment with aPPR led to little or no red fluorescent signal (Figure 2G) confirming earlier results that aPPR is not taken up efficiently. By contrast, HeLa cells treated with 1 μM 4.2R, 4.3R, 5.2R, 5.3R, TatR or Arg8R showed red fluorescent puncta throughout the cytosol in a distribution similar to that seen with Tf488 and at levels that qualitatively reproduce the trends detected by flow cytometry (Figure 2F). The fluorescent signals from Tf488 and Tf546 were highly correlated (Figure 2E), as were, with one exception, the fluorescent signals from Tf488 and rhodamine labeled miniature proteins/peptides (R488,rhodamine = 0.619 − 0.779),. Taken together, these data suggest that transferrin and cationic miniature proteins/peptides 4.2R, 5.2R, 5.3R, TatR and Arg8R are taken up into the same endocytic compartments. Miniature protein 4.3R is also taken up into a transferrin+ compartment, but the correlation (R488,rhodamine = 0.493) is lower, possibly because the uptake is low (Figure 2F). Additional experiments to assess whether peptide trafficking was affected by fluorophore identity confirmed that rhodamine and fluorescein labeled variants of cationic miniature proteins as well as Tat and Arg8 traffic together within cells, as indicated by highly correlated rhodamine and fluorescein intensities (R > 0.6, p < 1 × 10−10, Figure S2).

Cationic Miniature Proteins Reach The Cytoplasm

Polypeptides and peptide mimetics showing cellular uptake by flow cytometry may remain trapped in endosomes and fail to reach the cytosol (Yu et al., 2011 and Maiolo et al., 2004). Previous assays for the cytosolic localization of peptide- or peptoid-dexamethasone (Dex) conjugates exploited the interaction of the Dex-labeled molecule with the ligand-binding domain of the cytosolic glucocorticoid receptor (GR), which led eventually to the transcription of luciferase and its detection in cell lysates 40-48 hours later (Yu et al., 2005). Recognizing that treatment of cells with Dex leads to rapid (15 min) nuclear accumulation of a GR-green fluorescent protein fusion (GR-GFP) (Carey et al., 1996), we asked whether microscopy could provide a rapid assay for cytosolic localization by revealing the nuclear accumulation of GR-GFP in the presence of peptide-Dex conjugates.

To test this hypothesis, HeLa cells transiently transfected with GR-GFP were incubated for 30 min with between 0 and 10 μM Dex (Figure 3A and Figure S3A). When Dex was absent from the incubation media, these cells exhibited GFP signal throughout the cytoplasm and the nucleus (Figure 3A). Addition of between 3 nM and 10 μM Dex led to a dose-dependent decrease in the cytosolic GFP signal and a concomitant increase in the nuclear GFP signal (Figure S3A). We quantified these changes using the automated image processing package CellProfiler (Carpenter et al., 2006) to measure the ratio of the median GFP signal in the nucleus to the median signal within a 2 μm region surrounding region of cytosol (the ‘translocation ratio’, TR, Figure 3B). TR values near 1 indicate equivalent intensity between the nucleus and the surrounding region. Treatment of HeLa cells with 1 μM Dex for 30 min resulted in an increase in the TR from 1.07 ± 0.02 to 3.93 ± 0.14 (p = 1.92 × 10−22, vs untreated cells, two-tailed t-test), roughly 90.5% of value achieved with a 10-fold higher concentration (Figure 3C). We therefore chose the lower concentration for subsequent studies. When cells transfected with GR-GFP were treated for 30 min with 1 μM aPPDex, virtually no change in GR-GFP localization or the TR was observed (1.41 ± 0.09, vs 1.25 ± 0.04, p = 0.1473, two-tailed t-test, Figure 3A). These data are consistent with the observation that aPPR fails to enter cells and with previous results (Yu et al., 2005) that simply adding the Dex label does not confer upon an otherwise cell impermeable peptide the ability to reach the cytoplasm.

Figure 3.

Figure 3

Open in a new tab

Translocation of GR-GFP after treatment with dexamethasone and dexamethasone labeled peptides but not aPPDex. (A) HeLa cells transfected with GR-GFP (which appears black in the top row) after no treatment (−), or treatment with 1 μM dexamethasone or 1 μM aPPDex for 30 min at 37°C. The lower panel is an overlay of the GFP signal, shown in green, and the nuclear Hoescht signal shown in blue. (B) Quantification scheme. (C) Change in GR-GFP after treatment with 1 μM Dex-labeled cationic miniature protein for 30 min. (D) Quantification of the changes visible in (c). ns, not significant. * p ≤ 0.05, *** p ≤ 0.001, ANOVA. See also Figure S3.

We next evaluated the extent to which Dex-labeled miniature proteins, as well as Tat and Arg8, induce the nuclear translocation of GR-GFP (Figure 3C,D). Treatment of HeLa cells expressing GR-GFP with 1 μM 5.3Dex for 30 min led to a large increase in TR (3.1 ± 0.1) both compared to an untreated sample (−) (p = 2.68 × 10−48, ANOVA with Bonferroni post-test) or one treated with aPPDex (p = 1.96 × 10−16). Treatment of cells with 1 μM 4.2Dex, 4.3Dex, or 5.2Dex for 30 min led to a small or absent increase in nuclear GFP signal, and TRs (1.18 ± 0.04, 1.92 ± 0.10, 1.77 ± 0.11 respectively) that were not significantly different from cells treated with aPPDex. The TR measured after treatment of cells with Arg8Dex (2.19 ± 0.13) or TatDex (2.85 ± 0.10) were increased over control samples treated with aPPDex (Arg8Dex, p = 0.0053; TatDex, p = 2.54 × 10−9), and comparable to cells treated with 5.3Dex. Control experiments verified that the affinities of Dex labeled cationic miniature proteins and peptides for the human glucocorticoid receptor in vitro were similar to each other (between 1 – 36 nM), though slightly poorer than Dex itself (0.1 nM, see Figure S3B-G).

We considered that degradation of miniature proteins could lead to GR-GFP activation through release of the Dex moiety. While aPPR, 5.2R, and 5.3R are cleaved by cathepsin D (and to a smaller extent cathepsin L) in vitro (Figure S6D-F), HPLC analysis of whole cell lysates revealed minimal to no cleavage of the aPPR and 5.3R backbones in cells under the conditions of the translocation assay (Figure S3H, I). The stability of aPPR and 5.3R under these conditions minimizes the possibility that the increased TR observed in the presence of 5.3Dex results from increased degradation or protease susceptibility of this protein compared to aPPDex or 5.2Dex. To further validate the results of the translocation assay, we generated cytoplasmic extracts from HeLa cells treated with rhodamine labeled miniature proteins using streptolysin O (SLO) (Androlewicz et al, 1993). HPLC analysis of the SLO-extracts from HeLa cells treated with aPPR showed little to no fluorescent material, while analysis of cytoplasmic extracts from HeLa cells treated with 5.3R confirmed the presence of intact 5.3R (Figure S6B-H). Analysis of cytoplasmic extracts from HeLa cells treated with TatR or Arg8R also showed fluorescent material, but with retention times distinct from the starting material (Figure S6G,H), consistent with previous observations (Palm et al, 2007) that unstructured cell penetrating peptides can be rapidly degraded.

Cytoplasmic access requires active endocytosis

Two limiting models have been invoked to explain the trafficking of cationic peptides and proteins across the plasma membrane and into the cytoplasm. One model invokes ion-pair guided passive membrane diffusion (Rothbard et al., 2005); the other invokes endocytosis followed by endosomal release. We sought to distinguish between these models in two ways. First, we asked whether inhibitors of endocytosis block the uptake of cationic miniature proteins, their cytosolic localization, or both (Figure 4). Second, we confirmed that cell membrane integrity is not disrupted by the presence of 1 μM cationic miniature protein (Figure S4).

Figure 4.

Figure 4

Open in a new tab

Endocytosis is required for cytosolic access. (A) Endocytosis inhibitors block the uptake of 5.3R, TatR and Arg8R. Translocation of GR-GFP after treatment with 1 μM Dex, 5.3Dex, TatDex, Arg8Dex, or aPPDex in the presence (gray) or absence (black) of various small molecule inhibitors. Inhibitors of endocytosis included (B) 80 μM dynasore, (C) 5 mM methyl-β-cyclodextrin (MβCD), (D) 50 μM EIPA. (E) Translocation ratio after treatment with cationic miniature proteins and 200 nM bafilomycin . * p ≤ 0.05; *** p ≤ 0.001, ANOVA with Bonferroni post test. See also Figure S4.

To determine whether active endocytosis is required for 5.3Dex to reach the cytoplasm, we treated HeLa cells expressing GR-GFP with inhibitors of endocytosis before and during exposure to 1 μM 5.3Dex. Clathrin mediated endocytosis (CME), pinocytosis, and caveolin mediated endocytosis are dependent on dynamin (Doherty and McMahon, 2009), whose activity is inhibited by the small molecule dynasore (Macia et al., 2006). Depleting cellular cholesterol by treatment with methyl-β-cyclodextrin (MβCD) also inhibits these three processes (Rodal et al., 1999). Actin remodeling, a process inhibited by addition of N-ethyl-isopropyl amiloride (EIPA) (Koivusalo et al., 2010) facilitates clathrin coated pit formation (Doherty and McMahon, 2009) and is required for some dynamin and cholesterol independent endocytic pathways. Notably, addition of EIPA does not block the uptake of some CME ligands such as transferrin (Koivusalo et al., 2010).

To test the involvement of these pathways in the uptake of cationic miniature proteins and peptides, we pretreated HeLa cells for 30 min with 80 μM dynasore, 5 mM MβCD or 50 μM EIPA before adding 1 μM aPPR, 5.3R, TatR, or Arg8R for 30 min at 37°C in the presence of inhibitor. The cells were subsequently washed and visualized by confocal microscopy (Figure 4A). The presence of dynasore completely blocked the uptake of 5.3R, as well as TatR and Arg8R, suggesting that all three molecules are taken up in a dynamin-dependent fashion. The presence of EIPA also dramatically reduced the uptake of all three molecules, suggesting a role for actin metabolism in the internalization process. MβCD completely blocked the uptake of 5.3R, and TatR, but not Arg8R. Surprisingly cellular uptake of Arg8R was increased in the presence of MβCD, leading to diffuse fluorescence throughout the cytosol, a pattern not observed in the absence of the inhibitor.

We next measured the effect of blocking endocytosis on the ability of 5.3Dex, TatDex, and Arg8Dex to reach the cytoplasm. We were mindful that exchanging the rhodamine label for Dex could alter the physical properties of the molecule as well as the manner in which it trafficked in cells. HeLa cells transfected with GR-GFP were pre-treated for 30 min with the same inhibitors prior to the addition of either 1 μM Dex (positive control), aPPDex (negative control), 5.3Dex, TatDex, or Arg8Dex for 30 min at 37°C (Figure 4B-D). The cells were then washed and imaged to measure the TR. None of the endocytosis inhibitors altered the TR calculated for cells treated with aPPDex. However, all three inhibitors reduced, to background levels, the TR calculated after treatment with 5.3Dex, TatDex, and both dynasore and EIPA reduced the TR observed after treatment with Arg8Dex. As anticipated from the microscopy results discussed above (Figure 4), the cytoplasmic access of Arg8Dex was not decreased in the presence of MβCD. Neither EIPA nor dynasore blocked the increase in the TR calculated after treatment with Dex, but MβCD reduced this increase by 53%, presumably the result of direct complexation of free Dex by MβCD (Moya-Ortega et al., 2010). The results with Arg8R notwithstanding, these data support the contribution of endocytic pathways to the uptake of cationic miniature proteins and peptides, and not a model based on passive diffusion.

The acidification of endosomes begins almost immediately upon scission from the plasma membrane, as their lumen no longer communicates with the surrounding media. Bafilomycin is a potent inhibitor of the vesicular ATPase (Yoshimori et al., 1991), and its addition to culture media prevents endosomal acidification (Yoshimori et al., 1991; Fischer et al., 2004). To ask whether low vesicular pH was required for cytoplasmic escape, we pretreated HeLa cells expressing GR-GFP with 200 nM bafilomycin for 1 h before exposure to 1 μM Dex, 5.3Dex, TatDex, Arg8Dex or aPPDex (Figure 4E). Treatment with bafilomycin did not alter the effect of Dex or aPPDex on the TR, but completely blocked the increase in TR seen after cells were exposed to 5.3Dex (p = 1.4 × 10−11), TatDex (p = 5.3 × 10−11) or Arg8Dex (p = 8.2 × 10−6). Thus, while 5.3Dex, TatDex, and Arg8Dex reach the cytoplasm, they fail to do so in the absence of endosome acidification. This finding further supports the model that these molecules do not penetrate the plasma membrane directly, but rather escape to the cytoplasm from acidified endocytic vesicles.

Escape to the Cytoplasm from Early Endosomes

Phospholipids present in newly formed clathrin-coated vesicles and macropinosomes undergo rapid modification resulting the recruitment of Rab5 (Lakadamyali et al., 2006; Zoncu et al., 2009). Rab5 is a master regulator of endosome biogenesis (Zeigerer et al, 2012) and recruits additional cellular factors required for vesicle maintenance, fusion and maturation, including the phosphotidyl inositol (PI) 3-OH kinase (PI3K), Vps34 (Christoforidis et al., 1999b). The resulting early endocytic compartment mixes via homotypic fusion with other Rab5+ vesicles (Stenmark et al., 1994), and delivers cargo to other cellular locales through the budding off of transport vesicles (Puthenveedu et al., 2010) or Rab conversion (Rink et al., 2005). While some cargo (such as transferrin) is recycled to the cell surface, other cargo, such as low density lipoprotein, epidermal growth factor (Cantalupo et al., 2001) and several types of viruses (Vonderheit and Helenius, 2005) are delivered to late endosomes, marked by Rab7, for degradation in lysosomes (Huotari and Helenius, 2011). To characterize the intracellular route taken by 5.3R, TatR and Arg8R, we looked for overlap of these molecules with markers of endocytic uptake and GFP tagged Rab proteins. We also used small molecule inhibitors and dominant negative Rab variants to test the cellular activities required for 5.3Dex, TatDex and Arg8Dex to enter the cytoplasm.

Guided by the observation that 5.3R colocalizes with transferrin (Figure 3), a substrate known to internalize into Rab5+ vesicles (Lakadamyali et al., 2006), we asked whether 5.3R was also present in Rab5+ vesicles. HeLa cells were transfected with GFP-Rab5 and treated for 30 min with 1 μM 5.3R. When these cells were examined by confocal microscopy, 66% of the rhodamine signal overlapped with the signal from GFP-Rab5, confirming that 5.3R is present in Rab5+ vesicles (Figure 5A). Because Rab5 vesicles rapidly deliver their cargo to downstream vesicles (Rink et al., 2005), we also evaluated colocalization of 5.3R with Rab7-GFP (Figure 5C). HeLa cells transfected with Rab7-GFP and treated with 5.3R as above, showed a large fraction (87%) of the rhodamine signal located in the Rab7-GFP compartment, confirming that 5.3R enters both early (Rab5+) and late (Rab7+) endosomes. We found no overlap between 5.3R, TatR, or Arg8R and galT-GFP (Cole et al., 1996), a marker of the golgi (Figure S5). To test whether trafficking of 5.3R could be arrested at the Rab5 stage, we overexpressed a GTPase-inactive Rab5 mutant, Rab5Q79L (Stenmark et al., 1994), that blocks delivery of cargo to late endosomes and arrests vesicle maturation (Rink et al., 2005). Observation of HeLa cells transfected with Rab5Q79L-GFP and treated with 5.3R showed that nearly all (99%) of the miniature protein localized to enlarged GFP+ endosomes (Figure 5D), suggesting that arresting early endosome maturation arrests trafficking of 5.3R at the Rab5+ stage. Similar results were seen with Arg8R and TatR (Figure 5D-I). These data suggest that 5.3R, TatR and Arg8R follow a shared path through Rab5+ and then Rab7+ vesicles and provide a starting point to ask which, if any, of these trafficking events are required to reach the cytoplasm.

Figure 5.

Figure 5

Open in a new tab

Miniature protein 5.3R enters via endocytosis into Rab5+ vesicles before trafficking to Rab7+ vesicles. HeLa cells transfected with the indicated GFP fusion protein (panels A-I) were treated with 1 μM 5.3R (panels A-C), TatR (panels D-F) or Arg8R (panels G-I) before being washed and imaged by confocal microscopy. Colocalization of 5.3R, TatR and Arg8R with Rab5-GFP is moderate (panels A, D, G) but can be increased by arresting Rab5 maturation via overexpression of Rab5Q79L-GFP. (panels B, E, H). 5.3R, TatR and Arg8R are delivered to Rab7+ endosomes (panels C, F, I). (J) Transfection with Rab5Q79L-GFP does not block the increase of translocation ratio seen after treatment with dexamethasone, 5.3Dex, or Arg8Dex, but decreases the translocation ratio measured after treatment with TatDex (p = 0.005). (K) HeLa cells treated with 200 nM wortmannin for 30 min before treatment with 1 μM Dex, 5.3Dex, TatDex, Arg8Dex, or aPPDex for 30 additional min in the continued presence of the drug. Wortmannin decreased the translocation ratio measured after treatment with TatDex (p = 8.1 ×10−15) and Arg8Dex (p = 2.3 ×10−5). ns, not significant; * p ≤ 0.05; ** p ≤ 0.01; *** p ≤ 0.001, ANOVA with Bonferroni post test. See also Figure S5.

To identify the point or points along the endocytic pathway at which these three molecules escape to the cytsosol, we blocked vesicle maturation through overexpression of Rab5Q79L-GFP and assayed for cytosolic localization using dual color microscopy and an orthogonally labeled GR-mCherry. The TR of untreated HeLa cells expressing Rab5Q79L-GFP or Rab5-GFP and GR-mCherry was near unity (TR = 1.32 ± 0.07), as expected, and remained unchanged after treatment for 30 min with 1 μM aPPDex (TR = 1.25 ± 0.04). As expected, the TR values of Rab5Q79L-GFP and Rab5-GFP expressing cells increased after treatment for 30 min with 1 μM of Dex (TR = 2.24 ± 0.15 and 2.45 ± 0.21 for Rab5-GFP and Rab5Q79L-GFP expressing cells, respectively; Figure 5J). Treatment with 1 μM 5.3Dex resulted in TR values that were similar regardless of whether cells were transfected with wild-type Rab5-GFP or Rab5Q79L-GFP (1.80 ± 0.08 vs. 1.90 ± 0.07, respectively); similar findings were observed when cells were treated with Arg8Dex (1.59 ± 0.11 vs. 1.60 ± 0.09). By contrast, treatment with TatDex resulted in TR values that differed depending on whether cells were transfected with wild type Rab5-GFP or Rab5Q79L-GFP (1.57 ± 0.04 vs. 1.26 ± 0.03, p = 0.0057, respectively). Thus, arresting vesicle maturation with the GTPase-inactive Rab5 mutant Rab5Q79L blocked the ability of TatDex to reach the cytoplasm, but had no effect on 5.3Dex or Arg8Dex. These findings suggest that 5.3 and Arg8 escape to the cytoplasm from Rab5+ vesicles, whereas Tat can only escape later along the endocytic pathway.

To arrest endocytic traffic at an earlier stage, we treated cells with 200 nM wortmannin, a pharmacologic inhibitor of PI3K that blocks the maturation of Rab5+ vesicles (Christoforidis et al., 1999a) by decreasing the recruitment of Rab5 effectors that bind simultaneously to Rab5 and PI-3-phosphate containing membranes (Zoncu et al., 2009). Wortmannin treatment blocked the increase in TR seen after treatment with TatDex and Arg8Dex by 72% and 77%, respectively (Figure 5K). In contrast, the TR seen after treatment with 5.3Dex in the presence or absence of wortmannin was similar (2.36 ± 0.08 vs 2.76 ± 0.10), confirming that 5.3Dex escapes at or before the earliest Rab5+ state. Taken together, these data suggest that arresting vesicle maturation inhibits the cytosolic access of both TatDex and Arg8Dex, but that 5.3Dex is capable of leaving these earliest vesicles and reaching the cytoplasm.

Increased cytosolic localization of zinc finger proteins modified with 5.3 Arg motif

The results described above suggest that the arginine motif present in the aPP variant 5.3 facilitates the release of this molecule from Rab5+ endosomes into the cytosol. Notably, the regioisomer 5.2, although also decorated with five arginine side chains on the external α-helix face, is not released (Figure 3). To determine whether the arginine motif present on the 5.3 α-helix is transportable, that is, whether it can promote the cytosolic localization of other α-helix containing proteins, we turned to the zinc finger proteins, as their therapeutic potential is well known (Sera, 2009; Urnov et al., 2010). We began with the sequence of CP1, a single zinc finger possessing high zinc ion affinity (Krizek et al., 1991), and prepared a dexamethasone-labeled variant (ZF5.3Dex) whose α-helix was substituted with the arginine motif in 5.3 (Figure 6A), a change that neither prevented zinc binding nor significantly changed the secondary structure as measured by CD (Figure S6A). We next evaluated the extent to which ZF5.3Dex induced the nuclear translocation of GR-GFP when compared with a variant lacking the arginine motif (ZFDex) (Figure 6B). Treatment of HeLa cells expressing GR-GFP with 1 μM ZF5.3Dex for 30 min led to a large increase in TR (2.4 ± 0.09) compared to both an untreated sample (0.9 ± 0.05) and one treated with ZFDex (1.4 ± 0.05) (p < 0.0001, ANOVA with Bonferroni post-test). Control experiments confirmed that ZF5.3R can be recovered from cytosolic extracts of HeLa cells treated with ZF5.3R (Figure S3L), and that the peptide backbone of ZF5.3 is not degraded under the conditions in this assay (Figure S6I). As was observed with 5.3, inhibition of early endocytic events with dynasore, bafilomycin, EIPA or methyl-β-cyclodextrin reduced to background levels the increase in translocation ratios observed in the presence of ZF5.3Dex (Figure 6B). Furthermore, as observed with 5.3, when cells expressing GFP-Rab5 were treated with ZF5.3R and examined by confocal microscopy, 68% of the rhodamine channel overlapped with the signal from GFP-Rab5, confirming that ZF5.3R is present in Rab5+ vesicles (Figure 6C). Finally, as observed with 5.3, the TR of cells treated with ZF5.3R was unaffected when endocytic traffic was arrested by treatment with wortmannin or overexpression of dominant negative Rab5Q79L-GFP (Figure 6D). Together, these data imply that that the arginine motif in ZF5.3,like the arginine motif in 5.3, functions to promote endocytic uptake and release from early Rab5+ endosomes, and emphasizes the potential of helical-arginine display for modulating the escape of cationic miniature proteins and peptidomimetics from early endosomes into the cytoplasm. Although further work is necessary to evaluate how broadly the motif identified here can be applied, these results provide a structural and mechanistic framework for efficiently increasing the cell permeabilities of therapeutic peptides and proteins.

Figure 6.

Figure 6

Open in a new tab

The 5.3 Arginine Motif is Transportable. (A) Primary sequences of parent zinc finger (ZF), with charged residues mutated to alanine, and zinc finger displaying 5.3 arginine motif (ZF5.3). (B) Translocation of GR-GFP after treatment with 1 μM Dex-labeled zinc fingers in the presence and absence of endocytic inhibitors, as described in Figure 5. (C) Colocalization of Rab5-GFP with rhodamine labeled zinc fingers. Rhodamine labeled ZF5.3 shown in red, Rab5-GFP in green, Hoescht 33342 in blue. (D) Rab5Q79L-GFP overexpression does not block the increase of translocation ratio seen after treatment with ZF 5.3Dex. ns, not significant. *** p ≤ 0.001, ANOVA with Bonferroni post test. See also Figure S6.

Discussion

The interfaces that form between and among proteins and DNA–often large, flat, and polar–do not resemble those that bind small molecule substrates or traditional inhibitors (Rutledge et al., 2003). Targeting these ‘undruggable’ interfaces is a task well suited to protein and peptide ligands, but can only be successful if such molecules reach their cytosolic targets. Unfortunately, the very properties that endow peptide mimetics with their promise–size and polarity–are precisely those properties forbidden by Lipinski's rules (Lipinski et al., 2001). The challenge, therefore, is to identify the determinants that guide the uptake of peptide-like molecules and the mechanisms through which they gain cytosolic access, generating a new set of rules applicable to large peptidic molecules and their mimetics. Advancing this goal has been constrained by the absence of a rapid and robust assay capable of distinguishing between peptide-like molecules that remain trapped within endosomes and those that escape into the cytosol.

Here we took advantage of developments in automated image analysis and the rapid cytoplasmic-to-nuclear translocation of the glucocorticoid receptor (GR) to develop an assay that, by monitoring nuclear accumulation of a GR-GFP fusion, reports on the cytoplasmic entry of traditionally impermeant molecules tagged with Dex. This assay offers advantages of both speed and cost over a first-generation assay developed by Kodadek and colleagues (Yu et al., 2005), as it provides a readout in living cells within 30 min (as opposed to 48 hours) and eliminates the requirement for cell lysis or costly enzymatic substrates. With this assay, we measured clear differences in the cytosolic localization of four actively endocytosed miniature proteins and identified 5.3 as a cationic miniature protein whose rapid cytosolic localization is equivalent to or better than Tat and Arg8. The large differences in cytosolic localization among the set of closely related miniature proteins emphasizes that distinct structural determinants control both endocytic uptake and endosomal release. Uptake is favored by clustered α-helical arginine side chains, whereas release requires a more dispersed arginine array.

Coupling this new assay for cytosolic localization with live cell confocal microscopy allowed us to further clarify contrasting models for the intracellular pathway taken by cationic miniature proteins and peptides en route to the cytoplasm. At low concentrations, none of the molecules studied here crossed the plasma membrane directly. Miniature protein 5.3, with the lowest charge density, enters into and efficiently escapes from early (Rab5+) endosomes. In contrast, Tat and Arg8, which are also present in Rab5+ vesicles, require delivery to downstream Rab7+ vesicles or the recruitment of Rab5 effectors in order to reach the cytoplasm (Figure 7). The differences in arginine/lysine number and orientation in 5.3, Tat, and Arg8 will likely affect their side chain pKa values and thus the overall charge of each molecule at any given pH. Thus it is possible that the distinct arginine array in 5.3 favors formation of a critical protonated species within the early endosome (pH ≈ 6.5), whereas the lower pH present in lysosomes may be required to generate an equivalent state for Tat and Arg8. Alternatively, the distinct arginine array in 5.3 could also represent an export signal for cellular machinery that has yet to be identified.

Figure 7.

Figure 7

Open in a new tab

Scheme illustrating the stepwise pathway of traffic taken by 5.3, ZF5.3, Tat and Arg8 from the cell exterior into the cytosol. Endocytosis from the plasma membrane is required for all three molecules to reach the cytoplasm. Endosomes rapidly acquire Rab5, followed by Vps34, which phosphorylates phosphoinositide lipids forming PI3P. Rab5 and PI3P recruit Rab5 effectors beginning endosome maturation. Recruitment of Rab7 leads to the formation of late endosomes (LE), but the Rab5+ to Rab7+ transition is not required for 5.3 or ZF5.3 to access the cytoplasm. Wortmannin treatment blocks Vps34 and recruitment of Rab5 effectors as well as Tat and Arg8 escape. Rab5Q79L acts later, allowing escape of Arg8 but not Tat. Endosomes are progressively acidified, a process blocked by Bafilomycin (structure shown) and required for 5.3, ZF5.3, Tat, and Arg8 to reach the cytoplasm.

Significance

Proteins capable of crossing biological membranes show great promise as therapeutics as well as agents for delivery of macromolecules, such as siRNA, to the cytoplasm of target cells. In a broader sense, proteins that effectively traffic across membranes offer the potential to illuminate fundamental principles of cell biology. In this work we identified two small folded proteins, the cationic miniature protein 5.3 and the zinc finger module ZF5.3, that achieve cytosolic access through rapid internalization and efficient escape from Rab5+ endosomes. The trafficking pathway that we mapped for these molecules is similar to that taken by botulinum toxin and anthrax toxins, which also escape from early endosomes (Simpson, 2004). The pathways followed by Tat and Arg8, however, resemble those of SV40 (Vonderheit and Helenius, 2005), and HIV-1 (Vidricaire and Tremblay, 2005), and require transport beyond early Rab5+ endosomes to gain cytosolic access. This difference reveals that the ability of 5.3 and ZF5.3 to rapidly escape from early endosomes is a unique feature, not shared by canonical cell penetrating peptides, and implies the existence of distinct signals, encodable within short peptide sequences, that favor early versus late endosomal release. Identifying these signals and understanding their mechanistic basis will illustrate how cells control the movement of endocytic cargo and may allow researchers to engineer molecules to follow a desired delivery pathway for rapid cytosolic access. Collectively, our investigations represent a starting point for the optimization of well-folded functional cell penetrating proteins useful as pharmacologic tools and capable of modulating cytoplasmic protein function.

Experimental Procedures

Cell Culture and Transfections

HeLa cells (ATCC, Manassas, VA) were grown in T-75 culture flasks containing Dulbecco's Modified Essential Medium (DMEM, Gibco Cat. #11995-065) supplemented with 10% FBS, and 100 U/mL each Penicillin and Streptomycin. Transient transfections were performed using Fugene 6 or XtremeGene HP (Roche) and protocols recommended by the manufacturer. Plasmids encoding Rab5-GFP and Rab5Q79L-GFP were gifts from Pietro DeCamilli. A plasmid containing Rab7-GFP was a gift of Qing Zhong (Addgene plasmid # 28047). A plasmid containing GalT-EGFP was a gift of Jeniffer Lippincott-Schwartz (Addgene plasmid # 11937). A plasmid containing the glucocortcoid receptor fused to EGFP (pK7-GR-GFP) was a gift of Ian Macara (Addgene plasmid # 15534).

Analysis of surface binding and cell uptake

HeLa cells grown to ∼90% confluency were dissociated from flasks by incubation for 15 min at 37oC with 2 mL PBS, 1 mM EDTA, 1 mM EGTA. Cells were collected in warmed media, and aliquots (150 μL, 200,000 cells) distributed to 96 well plates. The cells were incubated at 4oC or 37oC for 10 min, 1 μM fluorescently labeled peptide was added, and incubation continued for an additional 30 min. Cells were then washed twice with DMEM + 10% FBS and treated with trypsin (30 μL per well, 0.05%, 37°C × 10 min) before resuspending in 300 μL PBS. Cells were analyzed by flow cytometry using an Accuri C6 flow cytometer. To confirm that trypsin treatment removed fluorescently labeled peptide remaining on the cell surface, HeLa cells were treated with 1 μM miniature protein or peptide at 4°C to inhibit endocytosis. After washing twice, the cells were then treated with trypsin (30 μL, 0.05%, 37°C × 10 min) or PBS as a control. Cells were then resuspended in 300 μL PBS before analysis by flow cytometry. Data presented are the mean ± SEM for 4 biological replicates measuring the mean fluorescence intensity of 30,000 cells. Dead cells defined by forward and side scatter were excluded.

Colocalization of miniature proteins with Alexa-488-transferrin and Rab-GFP fusions

To examine the colocalization of rhodamine labeled miniature proteins or peptides with Alexa-488-transferrin (Tf488), HeLa cells were plated (200 μL, 104 cells/well, 96 well glass bottom plates, Matrical) the day prior to experiments. The media was replaced with 150 μL Hepes-Krebs-Ringer's (HKR) buffer (140 mM NaCl, 2 mM KCl, 1 mM CaCl2, 1 mM MgCl2, 10 mM Hepes pH 7.4) containing 1 μM miniature protein or 25 nM Alexa-488-transferrin (Molecular Probes) and the cells incubated at 37°C for 30 min. The cells were then rinsed twice with 200 μL HKR buffer and nuclei were labeled by overlaying 200 μL HKR containing 300 nM Hoechst 33342 (Molecular Probes Cat. # H3570) for 5 minutes. Images of cells were acquired using a PerkinElmer LiveView spinning disk confocal microscope fitted with a 60 × 1.2 NA objective. Colocalization with Rab-GFP fusions was examined in an analogous way using HeLa cells transfected with the appropriate expression plasmid. See Supplementary Experimental Procedures online for details.

Effects of inhibitors on cell uptake

HeLa cells grown for 24 h in glass bottom plates were incubated with HKR buffer, or HKR buffer containing 80 μM dynasore, 50 μM N-ethyl-isopropyl-amiloride, or 5 mM methyl-β-cyclodextrin for 30 min at 37°C prior to the addition of 1 μM rhodamine-labeled miniature protein or peptide (Tat and Arg8). The cells were washed twice with DMEM, the nuclei labeled with Hoescht, and images acquired as described above for colocalization experiments.

GR-GFP translocation assay

HeLa cells transfected with pK7-GR-GFP were plated in Matrical plates as described above. To label nuclei, the media was replaced with HKR buffer containing 300 nM Hoescht and the cells incubated for 30 min at 37°C. Cells were then treated with 150 μL HKR buffer or HKR buffer containing 1 μM dexamethasone, dexamethasone labeled miniature protein, or dexamethasone labeled peptides for 30 min at 37°C before epifluorescence imaging. The translocation ratio (the ratio of the median intensities of GFP in the nuclear and surrounding region) for each cell imaged was measured using CellProfiler(Carpenter et al., 2006). For further details see Supplementary Experimental Procedures online.

To examine the effects of various inhibitors, HeLa cells transfected and plated as above were pretreated with HKR buffer containing 300 nM Hoescht and 80 μM dynasore, 50 μM N-ethyl-isopropyl-amiloride, 200 nM wortmannin, 200 nM bafilomycin or 5 mM methyl-β-cyclodextrin for 30 min at 37°C, after which was added 1 μM dexamethasone, dexamethasone labeled miniature protein or peptide for an additional 30 min at 37°C. Cells were then analyzed as described above.

To evaluate the requirement of Rab5 activity for the ability of peptides to reach the cytoplasm, HeLa cells were transfected with pGR-mCherry and either Rab5-EGFP or Rab5Q79L-EGFP for 24 hrs before treatment with dexamethasone or dexamethasone labeled miniature proteins or peptides and imaging as described above. Cotransfected cells expressing Rab-GFP fusions and pGR-mCherry were identified via their characteristic pattern of green and red fluorescence. The translocation ratio was determined using median values of mCherry fluorescence within the nucleus and surrounding region using CellProfiler.

Statistical Analysis

Comparisons within groups were made using ANOVA. Pairwise comparisons within groups were made using Bonferroni's post-test after finding a significant difference using ANOVA. P-values are corrected using Bonferroni's method(Shaffer, 1995) so that the family-wise error rate = 0.05. Otherwise, comparisons were made using a two tailed f-test.

Supplementary Material

01

Highlights.

  • A transportable helical arginine motif promotes cell uptake and endosomal release

  • Cytosolic access is monitored with a rapid, low cost, live cell assay

  • Clustered arginines promote endocytosis, while dispersed arginines promote release

  • Cationic proteins escape from Rab5+ endosomes but do not cross the plasma membrane

Acknowledgments

We are grateful to the NIH for support of this work (GM 74756). Support for JSA and DMB was provided in part by NIH MSTP TG T32GM07205. JSA was also supported by NIH F30 HL 094078-03. Support for JRL was provided by T32GM067543.

Footnotes

Supplementary Procedures: containing additional descriptions of materials, polypeptide synthesis and characterisation, pGR-mCherry vector construction, circular dichroism analysis, and image acquisition and analysis may be found online.

Competing financial interests: The authors declare no competing financial interests.

Publisher's Disclaimer: 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 citable 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.

References

  1. Androlewicz MJ, Anderson KS, Cresswell P. Evidence that transporters associated with antigen processing translocate a major histocompatibility complex class I-binding peptide into the endoplasmic reticulum in an ATP-dependent manner. Proc Natl Acad Sci USA. 1993;90:9130–9134. doi: 10.1073/pnas.90.19.9130. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Belitsky JM, Leslie SJ, Arora PS, Beerman TA, Dervan PB. Cellular uptake of N-methylpyrrole/N-methylimidazole polyamide-dye conjugates. Bioorg Med Chem. 2002;10:3313–3318. doi: 10.1016/s0968-0896(02)00204-3. [DOI] [PubMed] [Google Scholar]
  3. Blundell TL, Pitts JE, Tickle IJ, Wood SP, Wu CW. X-ray analysis (1. 4-A resolution) of avian pancreatic polypeptide: Small globular protein hormone. Proc Natl Acad Sci USA. 1981;78:4175–4179. doi: 10.1073/pnas.78.7.4175. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Cantalupo G, Alifano P, Roberti V, Bruni CB, Bucci C. Rab-interacting lysosomal protein (RILP): the Rab7 effector required for transport to lysosomes. EMBO J. 2001;20:683–693. doi: 10.1093/emboj/20.4.683. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Carey KL, Richards SA, Lounsbury KM, Macara IG. Evidence using a green fluorescent protein-glucocorticoid receptor chimera that the Ran/TC4 GTPase mediates an essential function independent of nuclear protein import. J Cell Biol. 1996;133:985–996. doi: 10.1083/jcb.133.5.985. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Carpenter AE, Jones TR, Lamprecht MR, Clarke C, Kang IH, Friman O, Guertin DA, Chang JH, Lindquist RA, Moffat J, Golland P, Sabatini DM. CellProfiler: image analysis software for identifying and quantifying cell phenotypes. Genome Biol. 2006;7:R100. doi: 10.1186/gb-2006-7-10-r100. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Chin JW, Schepartz A. Design and Evolution of a Miniature Bcl-2 Binding Protein. Angew Chem Int Ed Engl. 2001;40:3806–3809. doi: 10.1002/1521-3773(20011015)40:20<3806::AID-ANIE3806>3.0.CO;2-B. [DOI] [PubMed] [Google Scholar]
  8. Christoforidis S, McBride HM, Burgoyne RD, Zerial M. The Rab5 effector EEA1 is a core component of endosome docking. Nature. 1999a;397:621–625. doi: 10.1038/17618. [DOI] [PubMed] [Google Scholar]
  9. Christoforidis S, Miaczynska M, Ashman K, Wilm M, Zhao L, Yip SC, Waterfield MD, Backer JM, Zerial M. Phosphatidylinositol-3-OH kinases are Rab5 effectors. Nat Cell Biol. 1999b;1:249–252. doi: 10.1038/12075. [DOI] [PubMed] [Google Scholar]
  10. Cole NB, Smith CL, Sciaky N, Terasaki M, Edidin M, Lippincott-Schwartz J. Diffusional mobility of Golgi proteins in membranes of living cells. Science. 1996;273:797–801. doi: 10.1126/science.273.5276.797. [DOI] [PubMed] [Google Scholar]
  11. Cronican JJ, Beier KT, Davis TN, Tseng JC, Li W, Thompson DB, Shih AF, May EM, Cepko CL, Kung AL, Zhou Q, Liu DR. A Class of Human Proteins that Deliver Functional Proteins into Mammalian Cells In Vitro and In Vivo. Chem Biol. 2011;18:833–838. doi: 10.1016/j.chembiol.2011.07.003. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Daniels DS, Schepartz A. Intrinsically cell-permeable miniature proteins based on a minimal cationic PPII motif. J Am Chem Soc. 2007;129:14578–14579. doi: 10.1021/ja0772445. [DOI] [PubMed] [Google Scholar]
  13. Derossi D, Calvet S, Trembleau A, Brunissen A, Chassaing G, Prochiantz A. Cell internalization of the third helix of the Antennapedia homeodomain is receptor-independent. J Biochem. 1996;271:18188–18193. doi: 10.1074/jbc.271.30.18188. [DOI] [PubMed] [Google Scholar]
  14. Doherty GJ, McMahon HT. Mechanisms of Endocytosis. Annu Rev Biochem. 2009;78:857–902. doi: 10.1146/annurev.biochem.78.081307.110540. [DOI] [PubMed] [Google Scholar]
  15. Duchardt F, Fotin-Mleczek M, Schwarz H, Fischer R, Brock R. A comprehensive model for the cellular uptake of cationic cell-penetrating peptides. Traffic. 2007;8:848–866. doi: 10.1111/j.1600-0854.2007.00572.x. [DOI] [PubMed] [Google Scholar]
  16. Fernández-Suárez M, Ting AY. Fluorescent probes for super-resolution imaging in living cells. Nat Rev Mol Cell Biol. 2008;9:929–943. doi: 10.1038/nrm2531. [DOI] [PubMed] [Google Scholar]
  17. Fischer PM. Cellular uptake mechanisms and potential therapeutic utility of peptidic cell delivery vectors: Progress 2001-2006. Med Res Rev. 2007;27:755–795. doi: 10.1002/med.20093. [DOI] [PubMed] [Google Scholar]
  18. Fischer R, Köhler K, Fotin-Mleczek M, Brock R. A stepwise dissection of the intracellular fate of cationic cell-penetrating peptides. J Biol Chem. 2004;279:12625–12635. doi: 10.1074/jbc.M311461200. [DOI] [PubMed] [Google Scholar]
  19. Frankel AD, Pabo CO. Cellular uptake of the tat protein from human immunodeficiency virus. Cell. 1988;55:1189–1193. doi: 10.1016/0092-8674(88)90263-2. [DOI] [PubMed] [Google Scholar]
  20. Futaki S, Suzuki T, Ohashi W, Yagami T, Tanaka S, Ueda K, Sugiura Y. Arginine-rich peptides. An abundant source of membrane-permeable peptides having potential as carriers for intracellular protein delivery. J Biol Chem. 2001;276:5836–5840. doi: 10.1074/jbc.M007540200. [DOI] [PubMed] [Google Scholar]
  21. Gemperli AC, Rutledge SE, Maranda A, Schepartz A. Paralog-selective ligands for bcl-2 proteins. J Am Chem Soc. 2005;127:1596–1597. doi: 10.1021/ja0441211. [DOI] [PubMed] [Google Scholar]
  22. Golemi-Kotra D, Mahaffy R, Footer MJ, Holtzman JH, Pollard TD, Theriot JA, Schepartz A. High affinity, paralog-specific recognition of the Mena EVH1 domain by a miniature protein. J Am Chem Soc. 2004;126:4–5. doi: 10.1021/ja037954k. [DOI] [PubMed] [Google Scholar]
  23. Hanover JA, Willingham MC, Pastan I. Kinetics of transit of transferrin and epidermal growth factor through clathrin-coated membranes. Cell. 1984;39:283–293. doi: 10.1016/0092-8674(84)90006-0. [DOI] [PubMed] [Google Scholar]
  24. Hodges AM, Schepartz A. Engineering a Monomeric Miniature Protein. J Am Chem Soc. 2007;129:11024–11025. doi: 10.1021/ja074859t. [DOI] [PubMed] [Google Scholar]
  25. Huotari J, Helenius A. Endosome maturation. EMBO J. 2011;30:3481–3500. doi: 10.1038/emboj.2011.286. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Johannes L, Popoff V. Tracing the Retrograde Route in Protein Trafficking. Cell. 2008;135:1175–1187. doi: 10.1016/j.cell.2008.12.009. [DOI] [PubMed] [Google Scholar]
  27. Johnson RM, Harrison SD, Maclean D. Therapeutic applications of cell-penetrating peptides. Methods Mol Biol. 2011;683:535–551. doi: 10.1007/978-1-60761-919-2_38. [DOI] [PubMed] [Google Scholar]
  28. Krizek BA, Amann BT, Kilfoil VJ, Merkle DL, Berg JM. A consensus zinc finger peptide: design, high-affinity metal binding, a pH-dependent structure, and a His to Cys sequence variant. J Am Chem Soc. 1991;113:4518–4523. [Google Scholar]
  29. Koivusalo M, Welch C, Hayashi H, Scott CC, Kim M, Alexander T, Touret N, Hahn KM, Grinstein S. Amiloride inhibits macropinocytosis by lowering submembranous pH and preventing Rac1 and Cdc42 signaling. J Cell Biol. 2010;188:547–563. doi: 10.1083/jcb.200908086. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Lakadamyali M, Rust M, Zhuang X. Ligands for Clathrin-Mediated Endocytosis Are Differentially Sorted into Distinct Populations of Early Endosomes. Cell. 2006;124:997–1009. doi: 10.1016/j.cell.2005.12.038. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Lipinski CA, Lombardo F, Dominy BW, Feeney PJ. Experimental and computational approaches to estimate solubility and permeability in drug discovery and development settings. Adv Drug Deliv Rev. 2001;46:3–26. doi: 10.1016/s0169-409x(00)00129-0. [DOI] [PubMed] [Google Scholar]
  32. Luedtke NW, Carmichael P, Tor Y. Cellular uptake of aminoglycosides, guanidinoglycosides, and poly-arginine. J Am Chem Soc. 2003;125:12374–12375. doi: 10.1021/ja0360135. [DOI] [PubMed] [Google Scholar]
  33. Macia E, Ehrlich M, Massol R, Boucrot E, Brunner C, Kirchhausen T. Dynasore, a Cell-Permeable Inhibitor of Dynamin. Developmental Cell. 2006;10:839–850. doi: 10.1016/j.devcel.2006.04.002. [DOI] [PubMed] [Google Scholar]
  34. Maiolo JR, Ottinger EA, Ferrer M. Specific redistribution of cell-penetrating peptides from endosomes to the cytoplasm and nucleus upon laser illumination. J Am Chem Soc. 2004;126:15376–15377. doi: 10.1021/ja044867z. [DOI] [PubMed] [Google Scholar]
  35. Moya-Ortega MD, Alvarez-Lorenzo C, Sigurdsson HH, Concheiro A, Loftsson T. Gamma-Cyclodextrin hydrogels and semi-interpenetrating networks for sustained delivery of dexamethasone. Carbohydrate Polymers. 2010;80:900–907. [Google Scholar]
  36. Palm C, Jayamanne M, Kjellander M, Hällbrink M. Peptide degradation is a critical determinant for cell-penetrating peptide uptake. Biochim Biophys Acta. 2007;1768:1769–1776. doi: 10.1016/j.bbamem.2007.03.029. [DOI] [PubMed] [Google Scholar]
  37. Palm-Apergi C, Lorents A, Padari K, Pooga M, Hallbrink M. The membrane repair response masks membrane disturbances caused by cell-penetrating peptide uptake. The FASEB Journal. 2009;23:214–223. doi: 10.1096/fj.08-110254. [DOI] [PubMed] [Google Scholar]
  38. Payne CK, Jones SA, Chen C, Zhuang X. Internalization and Trafficking of Cell Surface Proteoglycans and Proteoglycan-Binding Ligands. Traffic. 2007;8:389–401. doi: 10.1111/j.1600-0854.2007.00540.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Puthenveedu MA, Lauffer B, Temkin P, Vistein R, Carlton P, Thorn K, Taunton J, Weiner OD, Parton RG, von Zastrow M. Sequence-Dependent Sorting of Recycling Proteins by Actin-Stabilized Endosomal Microdomains. Cell. 2010;143:761–773. doi: 10.1016/j.cell.2010.10.003. [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Richard JP, Melikov K, Vives E, Ramos C, Verbeure B, Gait MJ, Chernomordik LV, Lebleu B. Cell-penetrating peptides. A reevaluation of the mechanism of cellular uptake. J Biol Chem. 2003;278:585–590. doi: 10.1074/jbc.M209548200. [DOI] [PubMed] [Google Scholar]
  41. Rink J, Ghigo E, Kalaidzidis Y, Zerial M. Rab conversion as a mechanism of progression from early to late endosomes. Cell. 2005;122:735–749. doi: 10.1016/j.cell.2005.06.043. [DOI] [PubMed] [Google Scholar]
  42. Rodal SK, Skretting G, Garred O, Vilhardt F, van Deurs B, Sandvig K. Extraction of cholesterol with methyl-beta-cyclodextrin perturbs formation of clathrin-coated endocytic vesicles. Molecular Biology of the Cell. 1999;10:961–974. doi: 10.1091/mbc.10.4.961. [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Rothbard JB, Jessop TC, Wender PA. Adaptive translocation: the role of hydrogen bonding and membrane potential in the uptake of guanidinium-rich transporters into cells. Adv Drug Deliv Rev. 2005;57:495–504. doi: 10.1016/j.addr.2004.10.003. [DOI] [PubMed] [Google Scholar]
  44. Rutledge SE, Volkman HM, Schepartz A. Molecular Recognition of Protein Surfaces: High Affinity Ligands for the CBP KIX Domain. J Am Chem Soc. 2003;125:14336–14347. doi: 10.1021/ja034508o. [DOI] [PMC free article] [PubMed] [Google Scholar]
  45. Ryser HJ, Hancock R. Histones and basic polyamino acids stimulate the uptake of albumin by tumor cells in culture. Science. 1965;150:501–503. doi: 10.1126/science.150.3695.501. [DOI] [PubMed] [Google Scholar]
  46. Sera T. Zinc-finger-based artificial transcription factors and their applications. Adv Drug Deliv Rev. 2009;61:513–526. doi: 10.1016/j.addr.2009.03.012. [DOI] [PubMed] [Google Scholar]
  47. Shaffer J. Multiple hypothesis testing. Annual Review of Psychology. 1995;46:561–584. [Google Scholar]
  48. Simpson LL. Identification of the major steps in botulinum toxin action. Annu Rev Pharmacol Toxicol. 2004;44:167–193. doi: 10.1146/annurev.pharmtox.44.101802.121554. [DOI] [PubMed] [Google Scholar]
  49. Smith BA, Daniels DS, Coplin AE, Jordan GE, McGregor LM, Schepartz A. Minimally cationic cell-permeable miniature proteins via alpha-helical arginine display. J Am Chem Soc. 2008;130:2948–2949. doi: 10.1021/ja800074v. [DOI] [PubMed] [Google Scholar]
  50. Stenmark H, Parton RG, Steele-Mortimer O, Lütcke A, Gruenberg J, Zerial M. Inhibition of rab5 GTPase activity stimulates membrane fusion in endocytosis. EMBO Journal. 1994;13:1287–1296. doi: 10.1002/j.1460-2075.1994.tb06381.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  51. Strohl WR, Knight DM. Discovery and development of biopharmaceuticals: current issues. Curr Opin Biotechnol. 2009;20:668–672. doi: 10.1016/j.copbio.2009.10.012. [DOI] [PubMed] [Google Scholar]
  52. Ter-Avetisyan G, Tunnemann G, Nowak D, Nitschke M, Herrmann A, Drab M, Cardoso MC. Cell Entry of Arginine-rich Peptides Is Independent of Endocytosis. Journal of Biological Chemistry. 2008;284:3370–3378. doi: 10.1074/jbc.M805550200. [DOI] [PMC free article] [PubMed] [Google Scholar]
  53. Théodore L, Derossi D, Chassaing G, Llirbat B, Kubes M, Jordan P, Chneiweiss H, Godement P, Prochiantz A. Intraneuronal delivery of protein kinase C pseudosubstrate leads to growth cone collapse. J Neurosci. 1995;15:7158–7167. doi: 10.1523/JNEUROSCI.15-11-07158.1995. [DOI] [PMC free article] [PubMed] [Google Scholar]
  54. Urnov FD, Rebar EJ, Holmes MC, Zhang HS, Gregory PD. Genome editing with engineered zinc finger nucleases. Nat Rev Genet. 2010;11:636–46. doi: 10.1038/nrg2842. [DOI] [PubMed] [Google Scholar]
  55. Vidricaire G, Tremblay MJ. Rab5 and Rab7, but not ARF6, govern the early events of HIV-1 infection in polarized human placental cells. J Immunol. 2005;175:6517–6530. doi: 10.4049/jimmunol.175.10.6517. [DOI] [PubMed] [Google Scholar]
  56. Vonderheit A, Helenius A. Rab7 associates with early endosomes to mediate sorting and transport of Semliki forest virus to late endosomes. Plos Biol. 2005;3:e233. doi: 10.1371/journal.pbio.0030233. [DOI] [PMC free article] [PubMed] [Google Scholar]
  57. Wadia JS, Stan RV, Dowdy SF. Transducible TAT-HA fusogenic peptide enhances escape of TAT-fusion proteins after lipid raft macropinocytosis. Nat Med. 2004;10:310–315. doi: 10.1038/nm996. [DOI] [PubMed] [Google Scholar]
  58. Wender PA, Galliher WC, Goun EA, Jones LR, Pillow TH. The design of guanidinium-rich transporters and their internalization mechanisms. Adv Drug Deliv Rev. 2008;60:452–472. doi: 10.1016/j.addr.2007.10.016. [DOI] [PMC free article] [PubMed] [Google Scholar]
  59. Yoshimori T, Yamamoto A, Moriyama Y, Futai M, Tashiro Y. Bafilomycin A1, a specific inhibitor of vacuolar-type H(+)-ATPase, inhibits acidification and protein degradation in lysosomes of cultured cells. J Biochem. 1991;266:17707–17712. [PubMed] [Google Scholar]
  60. Yu P, Liu B, Kodadek T. A high-throughput assay for assessing the cell permeability of combinatorial libraries. Nat Biotechnol. 2005;23:746–751. doi: 10.1038/nbt1099. [DOI] [PubMed] [Google Scholar]
  61. Yu H, Zou Y, Wang Y, Huang X, Huang G, Sumer BD, Boothman DA, Gao J. Overcoming Endosomal Barrier by Amphotericin B-Loaded Dual pH-Responsive PDMA- b-PDPA Micelleplexes for siRNA Delivery. Acs Nano. 2011;5:9246–9255. doi: 10.1021/nn203503h. [DOI] [PMC free article] [PubMed] [Google Scholar]
  62. Zeigerer A, Gilleron J, Bogorad RL, Marsico G, Nonaka H, Seifert S, Epstein-Barash H, Kuchimanchi S, Peng CG, Ruda VM, Del Conte-Zerial P, Hengstler JG, Kalaidzidis Y, Koteliansky V, Zerial M. Rab5 is necessary for the biogenesis of the endolysosomal system in vivo. Nature. 2012;485:465–470. doi: 10.1038/nature11133. [DOI] [PubMed] [Google Scholar]
  63. Zellefrow CD, Griffiths JS, Saha S, Hodges AM, Goodman JL, Paulk J, Kritzer JA, Schepartz A. Encodable activators of SRC family kinases. J Am Chem Soc. 2006;128:16506–16507. doi: 10.1021/ja0672977. [DOI] [PubMed] [Google Scholar]
  64. Zhou H, Wu S, Joo JY, Zhu S, Han DW, Lin T, Trauger S, Bien G, Yao S, Zhu Y, Siuzdak G, Schöler HR, Duan L, Ding S. Generation of induced pluripotent stem cells using recombinant proteins. Cell Stem Cell. 2009;4:381–384. doi: 10.1016/j.stem.2009.04.005. [DOI] [PMC free article] [PubMed] [Google Scholar]
  65. Zoncu R, Perera RM, Balkin DM, Pirruccello M, Toomre D, De Camilli P. A phosphoinositide switch controls the maturation and signaling properties of APPL endosomes. Cell. 2009;136:1110–1121. doi: 10.1016/j.cell.2009.01.032. [DOI] [PMC free article] [PubMed] [Google Scholar]
  66. Zondlo NJ, Schepartz A. Highly Specific DNA Recognition by a Designed Miniature Protein. J Am Chem Soc. 1999;121:6938–6939. [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

01
NIHMS386997-supplement-01.doc (1.4MB, doc)

RESOURCES