Internalization of novel non-viral vector TAT-streptavidin into human cells

This is an electronic reprint of the original article. This reprint may differ from the original in pagination and typographic detail. Author(s): Title: Year: Version: Please cite the original version: All material supplied via JYX is protected by copyright and other intellectual property rights, and duplication or sale of all or part of any of the repository collections is not permitted, except that material may be duplicated by you for your research use or educational purposes in electronic or print form. You must obtain permission for any other use. Electronic or print copies may not be offered, whether for sale or otherwise to anyone who is not an authorised user. Internalization of novel non-viral vector TAT-streptavidin into human cells Rinne, Johanna; Albarran, Brian; Jylhävä, Juulia; Ihalainen, Teemu O; Kankaanpää, Pasi; Hytönen, Vesa P; Stayton, Patrick S; Kulomaa, Markku S; Vihinen-Ranta, Maija Rinne, J., Albarran, B., Jylhävä, J., Ihalainen, T., Kankaanpää, P., Hytönen, V., Stayton, P., Kulomaa, M. & Vihinen-Ranta, M. (2006). Internalization of novel non-viral vector TAT-streptavidin into human cells. BMC Biotechnology, 7 (1). doi:10.1186/1472- 6750-7-1 2007 BioMed Central Page 1 of 14 (page number not for citation purposes) BMC Biotechnology Open Access Research article Internalization of novel non-viral vector TAT-streptavidin into human cells Johanna Rinne1, Brian Albarran2, Juulia Jylhävä1, Teemu O Ihalainen1, Pasi Kankaanpää1, Vesa P Hytönen1,3, Patrick S Stayton2, Markku S Kulomaa1,4 and Maija Vihinen-Ranta*1 Address: 1NanoScience Center; Department of Biological and Environmental Science, University of Jyväskylä, Jyväskylä, Finland, 2Department of Bioengineering, University of Washington, Seattle, WA, USA, 3Department of Materials, ETH Zürich, Zürich, Switzerland and 4Institute of Medical Technology, University of Tampere, Tampere, Finland Email: Johanna Rinne - [email protected]; Brian Albarran - alba[email protected]; Juulia Jylhävä - [email protected]; Teemu O Ihalainen - [email protected]; Pasi Kankaanpää - [email protected]; Vesa P Hytönen - [email protected]; Patrick S Stayton - [email protected]u; Markku S Kulomaa - m[email protected]; Maija Vihinen-Ranta* - [email protected] * Corresponding author Abstract Background: The cell-penetrating peptide derived from the Human immunodeficiency virus-1 transactivator protein Tat possesses the capacity to promote the effective uptake of various cargo molecules across the plasma membrane in vitro and in vivo. The objective of this study was to characterize the uptake and delivery mechanisms of a novel streptavidin fusion construct, TAT47– 57-streptavidin (TAT-SA, 60 kD). SA represents a potentially useful TAT-fusion partner due to its ability to perform as a versatile intracellular delivery vector for a wide array of biotinylated molecules or cargoes. Results: By confocal and immunoelectron microscopy the majority of internalized TAT-SA was shown to accumulate in perinuclear vesicles in both cancer and non-cancer cell lines. The uptake studies in living cells with various fluorescent endocytic markers and inhibiting agents suggested that TAT-SA is internalized into cells efficiently, using both clathrin-mediated endocytosis and lipidraft-mediated macropinocytosis. When endosomal release of TAT-SA was enhanced through the incorporation of a biotinylated, pH-responsive polymer poly(propylacrylic acid) (PPAA), nuclear localization of TAT-SA and TAT-SA bound to biotin was markedly improved. Additionally, no significant cytotoxicity was detected in the TAT-SA constructs. Conclusion: This study demonstrates that TAT-SA-PPAA is a potential non-viral vector to be utilized in protein therapeutics to deliver biotinylated molecules both into cytoplasm and nucleus of human cells. Background Due to the limitations of current drug delivery systems, which have been hampered by their inefficiencies in traversing the cell membrane, there is a pressing need to develop methods for increasing intracellular delivery of protein-based cargoes. Over the past decade, numerous Published: 02 January 2007 BMC Biotechnology 2007, 7:1 doi:10.1186/1472-6750-7-1 Received: 15 September 2006 Accepted: 02 January 2007 This article is available from: http://www.biomedcentral.com/1472-6750/7/1 © 2007 Rinne et al; licensee BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. BMC Biotechnology 2007, 7:1 http://www.biomedcentral.com/1472-6750/7/1 Page 2 of 14 (page number not for citation purposes) strategies to overcome the cell membrane barrier have been proposed, including electroporation, microinjection, viral vectors, liposome encapsulation and receptormediated endocytosis. These methods have, however, been plagued by low delivery efficiencies and, to some extent, increased cellular toxicity. Naturally occurring short cell-penetrating peptides (CPPs) derived from viral, insect or mammalian proteins, have attracted considerable interest in the field of drug delivery for their ability to direct cellular uptake through active transport mechanisms. CPPs are oligopeptides, 11–30 amino acid residues in length, that are capable of conferring their apparent translocation activity to proteins and other macromolecular cargo to which they are linked [1]. In recent years, CPPs have been studied extensively, both in vitro and in vivo, for their ability to delivery an array of pharmalogically relevant cargoes, such as antisense oligonucleotides, peptides, proteins, plasmids, liposomes and nanometer-sized particles, with encouraging results [2]. Lately, CPPs have also been used to treat preclinical models of human disease [3,4]. One of the most well-studied and efficient cell penetrating peptides is the 11-amino-acid peptide of the Human immunodeficiency virus type 1 (HIV-1) Tat protein. This basic region of Tat, containing amino acid residues 47–57 (YGRKKRRQRRR; TAT47–57) [1,5], is crucial for many key functions of the protein, including interaction with the transactivation-responsive region in viral mRNA [6], nuclear localization [7]and most importantly, cellular uptake [8,9]. TAT47–57 has been shown to direct the internalization of an extensive list of cargoes ranging from small peptides [4] to proteins and polymers [3,10-12], liposomes [13,14], phage vectors [15], plasmid DNAs [16,17] and even nanoparticles [18]. Moreover, TAT47–57 has also been used in vivo to deliver biologically active βgalactosidase into all tissues of the mouse, even the brain [12]. A variety of internalization routes for the TAT47–57 sequence and TAT-mediated cargoes have been suggested. In the study of Fittipaldi et al. (2003) and Ferrari et al. (2003) Tat11EGFP and GST-Tat-eGFP proteins were reported to internalize into cells via caveolae-mediated endocytosis and transported further to the perinuclear area via an actin cytoskeleton-mediated mechanism [19,20]. Wadia et al. (2004) in turn showed the internalization of the TAT-Cre protein into cells by lipid raftdependent macropinocytosis [21], and Richard et al. (2005) suggested the uptake of the TAT peptide via clathrin-mediated endocytosis [22]. Recently, also Säälik et al. (2004) demonstrated the uptake of biotinylated TAT, detected with FITC-labeled avidin, via both clathrindependent and clathrin-independent endocytosis [23]. Central to the use of TAT, however, is not only its ability to deliver cargo to cells but, importantly, its non-cytotox- icity and stable biological activity over long time periods [5]. Core streptavidin (SA; 125–127 aa) from Streptomyces avidinii has been used in many pharmalogical applications. The exact mechanism of its cellular uptake and intracellular delivery is, however, not well known. The internalization of SA has been suggested to occur via receptor-mediated endocytosis, involving lysine residues [24] and the RYD sequence (Arg-Tyr-Asp) [25]. In vivo, the biodistribution of SA has been shown to exhibit slow clearance from the bloodstream due to accumulation in the kidney [26,27]. The present study was designed to gain insight into the internalization of a novel TAT- streptavidin (TAT-SA) construct [10] in human cells. Additionally, the ability of TAT-SA as a transporter of biotin and biotinylated molecules was examined. The subcellular distribution of TAT-SA was altered by biotinylated, pH-responsive polymer poly(propylacrylic acid) (PPAA), which further promoted the endosomal release. These studies provide insights into the mechanism of TATSA uptake in cells and may have implications for the optimal use of TAT-SA and PPAA for the intracellular delivery of numerous biotinylated macromolecules. Results Characterization of TAT-SA constructs The structure of TAT-streptavidin (TAT-SA) has been previously described [10] as a tetrameric fusion protein, in which the TAT47–57 peptide has been attached to the N-ter- minus of each streptavidin monomer. Biotins or biotinylated molecules are located into binding pockets of SA. In this study, the stability and biotin-binding ability of TAT-SA were analyzed by sodium dodecyl sulphate-poly- acrylamide gel electrophoresis (SDS-PAGE) and immunoblot analysis. Both constructs, TAT-SA and Alexa488- labeled TAT-SA (TAT-SA-A488), retained their tetrameric (60 kD) conformations in reducing conditions as either apoform or bound to biotin. When samples were preheated to 68°C, a minor proportion of TAT-SA or TAT-SA- A488 were detected as monomeric (15 kD) or dimeric (30 kD) forms. Binding of biotin to TAT-SA, however, stabilized the constructs at 68°C, since only tetrameric proteins were detected (unpublished data). Importantly, reducing conditions and preheatmeant to 37°C did not change the tetrameric conformations of either TAT-SA or TAT-SA bound to biotin (Additional file 1). Microscopical analysis of TAT-SA and SA uptake TAT-SA (2 μM) was shown to internalize into human epithelial carcinoma (HeLa), human lung carcinoma (A549) and human lung fibroblast (MRC-5) cell lines at 4 h post transduction (Fig. 1A–C). In living HeLa cells TAT-SA- A488 was observed to cross the cell membrane rapidly, BMC Biotechnology 2007, 7:1 http://www.biomedcentral.com/1472-6750/7/1 Page 3 of 14 (page number not for citation purposes) starting at 5 min post transduction, by confocal microscopy. At later timepoints the majority of TAT-SA-A488 was localized in vesicular compartments within the cytoplasm, and only few cells displayed detectable nuclear accumulation at 4 h post transduction (Fig. 1D–E). Control studies in which cells were first transduced either with TAT-SA-A488 or TAT-SA, fixed at 4 post transduction and immunostained for SA confirmed that both studies in living and fixed cells resulted in a similar subcellular localization. Additionally, in cells transduced with SA-A488 (2 μM) alone, the cellular uptake was reduced (Fig. 1G–H). The role of TAT during the nuclear import of SA was confirmed by injecting TAT-SA-A488 or SA-A488 (~1.5 × 106 molecules/cell) directly into the cytosol of HeLa cells. At 4 h post injection the majority of TAT-SA-A488 was found to accumulate in the nucleoplasm with only minor localization in the cytoplasm. In the nucleus TAT-SA-A488 was distributed evenly throughout the nucleoplasm without accumulation into distinct subnuclear structures (Fig. 1F). In contrast, SA-A488 was distributed randomly throughout the cytoplasm but not in the nucleus (Fig. 1I). To monitor further the internalization and intracellular transport of TAT-SA and SA, transduced HeLa cells were labeled with an antibody against SA and nanogold-immu- nolabeling electron microscopy with a silver enhancement technique was used. Intracellular TAT-SA, visualized as small, intensely labeled, grainy spots was localized in large intracellular vesicles near the cell surface or in endocytic-like vesicles adjacent to the nuclear membrane at 4 h p.t (Fig. 2A). At the same time a small proportion of TATSA was observed in the cytosol and the nucleus of cells (Fig. 2B). Interestingly, in the control experiments SA was observed in smaller cytoplasmic vesicles without any nuclear localization than in those displaying TAT-SA at 30 min to 4 h post transduction. Endocytic entry pathways of TAT-SA To study the potential role of caveolar endocytosis in the uptake of TAT-SA, HeLa cells were transduced and immunolabeled with caveolin-1 antibody. No colocalization of TAT-SA-A488 and caveolin was, however, apparent at 15– 60 min post transduction. In addition, the internalization of TAT-SA-A488 into living HeLa cells was not blocked in the presence of the cholesterol-depleting agent β-methyl- cyclodextrin, a known inhibitor of caveolae-mediated endocytosis. Importantly, control studies with human hepatoma (HepG2) cells, which do not express caveolins endogenously, demonstrated the efficient internalization of TAT-SA (unpublished data). In order to characterize clathrin-dependent and clathrinindependent endocytosis in the uptake of TAT-SA, AP180- C plasmid encoding a dominant-negative form of AP180, an inhibitor of clathrin-mediated endocytosis, was used [28]. In HeLa cells overexpressing AP180-C, the internalization of TAT-SA was reduced, but not completely inhibited at 4 h post transduction (Fig. 3A), whereas the uptake of a clathrin-mediated endocytotic marker, Transferrin (Tf; 200 μg/ml), was markedly decreased. In the absence of AP180-C plasmid, immunofluorescence studies revealed slight colocalization of TAT-SA-A488 and an early endosomal marker (rab5) at 15 min post transduction as well as of TAT-SA-A488 and a recycling endosomal marker (rab11) at 30 min post transduction (unpublished data). In addition, the lysosomal marker LAMP-2 showed clear colocalization with TAT-SA-A488 at 2–4 h post transduction (Fig. 3B). The kinetics of TAT-SA translocation from the cell periphery towards the nuclear periphery at different times was monitored in living HeLa cells in the presence of the fluorescent endocytic markers transferrin and dextran. Partial colocalization of TAT-SA-A488 and TRITC-labeled Transferrin (TRITC-Tf) was observed first at 5 min post transduction in close proximity to the cell membrane and later at 15–30 min post transduction in perinuclear vesicles (Fig. 4A). In order to examine the role of non-clathrin- mediated endocytosis, a fluid-phase endosomal marker TRITC-labeled Dextran (TRITC-Dextran; 10 MW, 250 μg/ ml), was used to monitor macropinocytic entry pathway. As shown in Figures 4B, 4C and in Additional File 2, extensive colocalization of TAT-SA-A488 and TRITC-Dex- tran was observed at both 15 min and 4 h post transduction. Moreover, quantitative analysis of the confocal images (n = 25–30) revealed an approximately 10-fold higher colocalization between TAT-SA-A488 and TRITC- Dextran (39.2% ± 3.8%) than TAT-SA-A488 and TRITC-Tf (3.1% ± 0.4%) at 15 min post transduction (Fig 5A). The uptake of TAT-SA was further examined in living HeLa cells by quantitative image analysis (n = 25–30) in the presence of different endocytic inhibitors. In comparison to relative fluorescence intensity of untreated, transduced control cells (1.0 ± 0.3; Fig. 5B) a microtubule-disrupting drug, nocodazole, caused only a slight decrease (0.78 ± 0.3) of intracellular TAT-SA-A488. In cells treated with the filamentous F-actin elongation inhibitor cytochalasin D, affecting both macropinocytosis and clathrin-dependent endocytosis, a clear decrease (0.22 ± 0.1) in the amount of cytoplasmic TAT-SA was, however, measured. Furthermore, amiloride, an inhibitor of Na+/H+ exchange required for macropinocytosis, displayed significantly reduced (0.14 ± 0.1) amounts of TAT-SA uptake. Importantly, in control experiments extensive disruption of microtubules or actin filaments was observed in cells treated either with nocodazole or cytochalasin D and amiloride. Additionally, cytochalasin D and amiloride markedly decreased the internalization of TRITC-Dextran in living cells (unpublished data). BMC Biotechnology 2007, 7:1 http://www.biomedcentral.com/1472-6750/7/1 Page 4 of 14 (page number not for citation purposes) The endosomal release of TAT-SA and TAT-SA bound to biotin via the PPAA polymer The streptavidin-biotin association is one of the strongest known non-covalent interactions in nature. In order to alter the subcellular distribution of TAT-SA, a biotinylated, pH-responsive polymer poly(propylacrylic acid) (PPAA; 4 μM) was complexed to TAT-SA (2 μM) to promote endosomal release. The 3D illustrations of live HeLa Confocal microscopy analysis of intracellular distribution of TAT-SA or SA at 4 h post transduction or injectionFigure 1 Confocal microscopy analysis of intracellular distribution of TAT-SA or SA at 4 h post transduction or injection. (A-C) Human cancer cell lines (HeLa, A549) and a non-cancer cell line (MRC-5) were transduced with TAT-SA prior to PFA fixation, permeabilization with Triton-X and immunolabeling with rabbit SA Ab followed by Alexa-488-conjugated goat anti-rabbit IgG (green). (D-E) Living HeLa cells were transduced or (F) injected to the cytoplasm with TAT-SA-A488 (green), or (G-H) transduced or (I) injected with SA-A488 (green). Scale bars, 10 μm. BMC Biotechnology 2007, 7:1 http://www.biomedcentral.com/1472-6750/7/1 Page 5 of 14 (page number not for citation purposes) cells showed that TAT-SA complexed with biotinylated PPAA was localized both in cytoplasmic vesicles and in the nucleus at 4 h post transduction (Fig. 6A, Additional file 3). Interestingly, TAT-SA was shown to accumulate into distinct nuclear foci, suggesting interactions with subnuclear components (Fig. 6A). Moreover, quantitative analysis (n = 25–30) of the relative nuclear fluorescence intensity demonstrated that PPAA induced over 2-fold increase in the intranuclear TAT-SA-488 (1.0 ± 0.3) when compared to TAT-SA-A488 transduced control cells (0.4 ± 0.1; Fig. 6B). The ability of TAT-SA to act as an intracellular delivery vector by transporting biotin and biotinylated molecules into cells was also examined by confocal microscopy. As shown in the 3D illustration, TAT-SA-A488 (2 μM) and Biotin-DY-633 (1–2 μM) accumulated in large intracellular vesicles near the nucleus (Fig. 7A). No nuclear localization of TAT-SA was observed, however. Further experiments demonstrated strong colocalization of the TAT-SA-biotin complex and TRITC-Dextran in vesicles at the nuclear periphery at 2 h and 4 h post transduction, implying that most of the TAT-SA-biotin complexes were unable to escape from the intracellular vesicles. When endosomal releasing polymer PPAA was complexed with TAT-SA bound biotin, however, a major subnuclear accumulation was observed (Fig. 7B, Additional file 4). Finally, cell viability experiments with MTT cell proliferation assay showed that TAT-SA (2 μM) was not cytotoxic to HeLa cells at 4–72 h post transduction (Table 1). Also no indication of a cytotoxic effect was observed in the presence of TAT-SA-complexed, biotinylated PPAA (4 μM) at 4–24 h post transduction and only a slight decrease in cell viability was detected at 48 and 72 h post transduction (94.0% ± 3.8% survival) compared to untransduced control cells (100.0% ± 3.3% survival). Discussion The cell membrane is a barrier to the intracellular delivery of many pharmacologically important biological macromolecules. Several cell penetrating peptides have been shown to possess the ability to direct cellular uptake, including the HIV-1 TAT peptide. So far TAT has been used for directing the intracellular uptake of various biomolecular cargoes. In this study, however, a novel streptavidin fusion protein TAT-SA [10] was examined, which enables the transport of biotinylated cargoes into cells. The internalization of TAT-SA was characterized in both human cancer (HeLa and A549) and non-cancer cell lines (MRC-5) (Fig. 1A–E). The internalization of SA was a relatively slow process in which detectable amounts of SA- A488 were present in the cytoplasm but not the nucleus of HeLa cells at 4 h post transduction (Fig. 1G–H). On the contrary, TAT-SA displayed rapid and efficient internalization in living HeLa cells starting at 5 min post transduc- Localization of internalized TAT-SA in HeLa cells by immunoelectron microscopy at 4 h post transductionFigure 2 Localization of internalized TAT-SA in HeLa cells by immunoelectron microscopy at 4 h post transduction. Intracellular TAT-SA was detected with pre-embedding immunolabeling technique, in which anti-SA staining was followed by silver-enhanced nanogold and gold-toning treatments. Localization of SA was shown as dark spots (arrow heads) within (A) cytoplasmic vesicular structures and further (B) in the nucleus of cells (nu). Scale bars, 1 μm. BMC Biotechnology 2007, 7:1 http://www.biomedcentral.com/1472-6750/7/1 Page 6 of 14 (page number not for citation purposes) tion with distribution into vesicular structures in the cytoplasm after 15 min (Fig. 1A, 1D–E). At later time points, confocal and EM imaging demonstrated that the majority of TAT-SA was localized in the cytoplasmic vesicles with trace amounts in the nucleus (Fig, 1D–E, Fig. 2A and 2B). Previously, numerous possible internalization routes for TAT have been proposed, such as lipid-raft- mediated macropinocytosis [21], caveolae-mediated endocytosis [19,20] and clathrin-independent and dependent endocytosis [22,23]. However, the uptake characteristics of the TAT peptide alone and of TAT-conju- gated cargoes have been demonstrated to differ significantly [1,4,22]. Furthermore, TAT-mediated internalization process has proposed to be dependent on the properties of the cargo molecule, TAT concentration and cell line [1,29,30]. In our study, streptavidin (60 kD) as a larger partner of the TAT-SA fusion construct (TAT47– 57-peptide, 11aa) is likely to affect the uptake and intracellular trafficking of the vector. Notably, we show here that direct microinjection of high concentrations of TATSA or SA into the cytoplasm resulted in efficient nuclear uptake of TAT-SA but not SA (Fig. 1F and 1I). This verifies previous findings that upon introduction into the cytosol, the TAT peptide is capable of mediating the nuclear import of its streptavidin fusion partner. Moreover, it is known that positively charged molecules internalize the cells efficiently. Consequently, taken to account that the plain SA is negatively charged (theoretical pI 6.04) and TAT-SA is positively charged (theoretical pI 9.92), TAT-SA internalizes the cells more efficiently than SA. Taken together, these data demonstrate that TAT-SA is efficiently internalized into various human cells but that only a relatively small proportion is further released into the cytoplasm and transported into the nucleus, most likely reflecting the inability of TAT-SA to escape from endocytic vesicles. In order to use TAT-SA as a vector to deliver biotinylated molecules into cells, the internalization and delivery mechanisms of the construct have to be characterized in vitro and in vivo. In recent years a number of studies have suggested a variety of internalization routes for TAT pep- Immunofluorescence microscopy studies of the cellular distribution of TAT-SAFigure 3 Immunofluorescence microscopy studies of the cellular distribution of TAT-SA. (A) HeLa cells were transfected with mutant AP180-C, a specific dominant-negative inhibitor of clathrin-mediated endocytosis, 24 h prior to transduction with TAT-SA-A488 (green) and clathrin-endocytosis marker TRITC-Transferrin (TRITC-Tf, red). Cells were fixed with PFA at 4 h post transduction, permeabilized with Triton-X and finally stained with antibody against myc-tagged AP180-C followed by an Alexa-633-conjugated anti-mouse antibody (purple). (B) The distribution of TAT-SA-A488 and lysosomal marker (red) was monitored in HeLa cells fixed (PFA) at 4 h post transduction and stained with an antibody against lysosomal marker LAMP-2 followed by an Alexa-546-conjugated anti-mouse antibody (red). Scale bars, 10 μm. BMC Biotechnology 2007, 7:1 http://www.biomedcentral.com/1472-6750/7/1 Page 7 of 14 (page number not for citation purposes) tide and TAT-mediated cargoes. To study this, three major endocytic pathways involving caveolae, lipid rafts and clathrin-coated pits were analyzed using specific endocytic markers, inhibition agents and immunofluorescence labelings of each pathway. Previous studies have suggested the TAT peptide enters cells by temperaturedependent, caveolae-mediated endocytosis [19,20]. However, no evidence for the use of the caveolae route in the internalization of TAT-SA was observed in the present confocal microscopial colocalization studies with a caveolar marker protein. Moreover, treatment with methyl-β- cyclodextrin, a cholesterol depletion agent known to inhibit the caveolae route [31] or transduction of the caveolin-deficient HepG2 (unpublished data) and Jurkat T cells [10,21,32,33] did not prevent the entry of TAT-SA- A488 into living cells. Altogether, these data imply that endosomal routes other than caveolae-mediated entry are required for the uptake of TAT-SA. It has been suggested that the internalization of SA occurs via clathrin-mediated endocytosis [24,25]. Here, the role of clathrin-dependent uptake of TAT-SA was monitored in cells overexpressing the mutant AP180-C protein [34]. AP- 180C is required for the efficient assembly of clathrincoated pits by interacting with the clathrin heavy chain through its C-terminal clathrin-binding motifs. Our data indicated that in AP180 overexpressing cells the internalization of TAT-SA was only slightly affected, whereas the uptake of Tf, a marker of clathrin-dependent endocytosis [35], markedly decreased (Fig. 3A). The double-immu- Colocalization of TAT-SA with endocytic markers in living cellsFigure 4 Colocalization of TAT-SA with endocytic markers in living cells. (A) HeLa cells were transduced first with TAT-SA- A488 (green) and then with TRITC-labeled transferrin (TRITC-Tf, red), a marker clathrin-mediated endocytosis, prior to analysis with confocal microscope at 15 min post transduction. (B) Cells transduced with TAT-SA-A488 and fluid-phase endosomal marker TRITC-labeled dextran (10 kD, red) were monitored at 15 min and (C) at 4 h post transduction. The white rectangles show the close-ups of representative structures. Scale bars, 10 μm. BMC Biotechnology 2007, 7:1 http://www.biomedcentral.com/1472-6750/7/1 Page 8 of 14 (page number not for citation purposes) nolabeling studies of clathrin-mediated endocytosis showed only minor colocalization of TAT-SA with early and recycling endosomal markers at early time points (unpublished data). However, at later stages of endocytosis TAT-SA accumulated in lysosomes, the final destination of multiple endocytic pathways (Fig. 3B). Clathrinmediated endocytosis seemed therefore to be only partially involved in the uptake of TAT-SA, the major internalization being mediated by another, more efficient pathway. Many oligoarginine peptides have been proposed as candidates for cellular internalization via macropinocytosis [30]. Recently, uptake via clathrin-independent endocytosis was also suggested in the internalization of TAT-Cre and biotinylated TAT [21,23,36]. In our study, the intracellular localization of TAT-SA was monitored in living HeLa cells with various fluorescent endocytic markers. Shortly after internalization, only slight colocalization was detected with TAT-SA and Tf, a clathrin-dependent endocytic marker (Fig. 4A), whereas extensive colocalization between TAT-SA and the fluid-phase endosomal and macropinocytic marker dextran was observed (4B, 4C and Additional file 2). Quantitative measurements of fluorescent intensity at 4 h post transduction displayed an approximately 10-fold increase in the colocalization of TAT-SA and dextran as compared to TAT-SA and Tf (Fig. 5A), thus supporting previous observations of the role of macropinocytosis as an entry mechanism for TAT. Additionally, when cells were treated with amiloride, a known inhibitor of macropinocytosis [37], the uptake of TAT-SA- A488 was almost entirely inhibited. In the presence of cytochalasin D, an F-actin-disrupting agent affecting both macropinocytosis and clathrin-mediated endocytosis [34,38], only a minor internalization of TAT-SA-A488 occurred (Fig. 5B). To conclude, we were able to establish macropinocytosis as a major internalization pathway of TAT-SA, as illustrated (Fig. 8). Quantification of internalized TAT-SA in living HeLa cellsFigure 5 Quantification of internalized TAT-SA in living HeLa cells. (A) Quantitative analysis of the colocalization between both TAT-SA-A488 and TRITC-labeled transferrin and TRITC-labeled dextran (10 kD) at 15 min post transduction. (B) The relative fluorescence intensity of internalized TAT-SA-A488 was measured in cells treated with cytochalasin D (cytD), amiloride (ami) or nocodazole (noco) at 4 h post transduction. Control cells (C) were untreated. The fluorescence intensity data was collected from multiple series of cells by confocal microcopy and processed with the 3D LSM program.