The role of endocytosis in the uptake and intracellular trafficking of PepFect14–nucleic acid nanocomplexes via class A scavenger receptors

Biochimica et Biophysica Acta 1848 (2015) 3205–3216

Contents lists available at ScienceDirect

Biochimica et Biophysica Acta journal homepage: www.elsevier.com/locate/bbamem

The role of endocytosis in the uptake and intracellular trafficking of PepFect14–nucleic acid nanocomplexes via class A scavenger receptors Carmen Juks a, Kärt Padari a, Helerin Margus a, Asko Kriiska a, Indrek Etverk a, Piret Arukuusk b, Kaida Koppel a, Kariem Ezzat c,d, Ülo Langel b,d, Margus Pooga a,⁎ a

Institute of Molecular and Cell Biology, University of Tartu, Tartu, Estonia Laboratory of Molecular Biotechnology, Institute of Technology, University of Tartu, Tartu, Estonia Department of Physiology, Anatomy and Genetics, University of Oxford, Oxford, UK d Department of Neurochemistry, Stockholm University, Stockholm, Sweden b c

a r t i c l e

i n f o

Article history: Received 15 April 2015 Received in revised form 2 September 2015 Accepted 22 September 2015 Available online 25 September 2015 Keywords: Cell penetrating peptide Nucleic acid delivery Endocytosis Intracellular trafficking Scavenger receptors

a b s t r a c t Cell penetrating peptides are efficient tools to deliver various bioactive cargos into cells, but their exact functioning mechanism is still debated. Recently, we showed that a delivery peptide PepFect14 condenses oligonucleotides (ON) into negatively charged nanocomplexes that are taken up by cells via class A scavenger receptors (SR-As). Here we unraveled the uptake mechanism and intracellular trafficking of PF14–ON nanocomplexes in HeLa cells. Macropinocytosis and caveolae-mediated endocytosis are responsible for the intracellular functionality of nucleic acids packed into nanocomplexes. However, only a negligible fraction of the complexes were trafficked to endoplasmic reticulum or Golgi apparatus — the common destinations of caveolar endocytosis. Neither were the PF14–SCO nanocomplexes routed to endo-lysosomal pathway, and they stayed in vesicles with slightly acidic pH, which were not marked with LysoSensor. “Naked” ON, in contrary, was rapidly targeted to acidic vesicles and lysosomes. The transmission electron microscopy analysis of interactions between SR-As and PF14–ON nanocomplexes on ultrastructural level revealed that nanocomplexes localized on the plasma membrane in close proximity to SR-As and their colocalization is retained in cells, suggesting that PF14–ON complexes associate with targeted receptors. © 2015 Elsevier B.V. All rights reserved.

1. Introduction Cell penetrating peptides (CPPs) are gaining popularity as non-viral delivery vectors for transfecting nucleic acids into cells, including oligonucleotide- based therapeutics [1–3]. To achieve sufficient transfection efficiency and functionality, several biological hurdles have to be overcome. The plasma membrane is the first barrier, regulating the trafficking between the interior and exterior of the cell. Uptake of CPPs is considered to start upon the interaction with negatively charged plasma membrane components, followed by internalization via different endocytosis mechanisms. The internalization mostly takes place by clathrindependent pathway [4,5], caveolin-dependent endocytosis [6,7] and via macropinocytosis [8]. However, the exact mechanism may depend on

Abbreviations: CPP, cell penetrating peptide; TP10, transportan 10; PF, PepFect; ON, oligonucleotide; SCO, splice-correcting oligonucleotide; pDNA, plasmid DNA; SR-A, class A scavenger receptor. ⁎ Corresponding author at: Institute of Molecular and Cell Biology, 23 Riia Str, Tartu 51010, Estonia. E-mail address: [email protected] (M. Pooga).

http://dx.doi.org/10.1016/j.bbamem.2015.09.019 0005-2736/© 2015 Elsevier B.V. All rights reserved.

CPP, cargo, used cell line or particular experimental conditions and frequently different pathways may be utilized in parallel [4,9,10]. The above mentioned features significantly complicate the design and development of new CPP-based pharmaceuticals. Although, some CPPs can penetrate into cells independently from endocytosis and this pathway has been suggested as the main cell entry mechanism for example for CADY peptide [11] and for some arginine-rich peptides [9,12]; still if CPPs are coupled to big macromolecules, endocytosis is considered as the main uptake mechanism. Once inside the cells, the intracellular fate and trafficking of CPPs or CPP-cargo complexes is even less understood. Entrapment of CPPs in endosomes is the next obstacle that limits their access to intracellular targets. Recently it was demonstrated that, in contrary to expected, the second generation CPP, PepFect14 (PF14) forms negatively charged nanocomplexes with splice correcting oligonucleotides (SCOs) [13], which cannot associate with negatively charged proteoglycans for induction of endocytosis. This finding implies that the cellular uptake of PF14–SCO nanocomplexes has to be mediated by other types of receptors. The plausible candidates, scavenger receptors, form a large superfamily, whose first members were initially characterized by their ability to bind and internalize modified low density lipoproteins. However, they also associate with various other polyanionic ligands

3206

C. Juks et al. / Biochimica et Biophysica Acta 1848 (2015) 3205–3216

[14], which made them good candidates for PF14–SCO nanocomplexes. Moreover, scavenger receptors are known to facilitate the cellular uptake of nucleic acids, like polyribonucleotides [15,16], dsRNA [17–19] and polyvalent oligonucleotide-functionalized gold nanoparticles [20]. Further experiments with SR-A specific inhibitory ligands revealed that the uptake of PF14–SCO complexes was indeed mediated by these receptors, whose two subtypes SR-A3 and SR-A5 were present in HeLa cells. The presence of inhibitory ligands abolished both the binding of nanocomplexes at the plasma membrane and internalization into cells [21]. Involvement of SR-As was further confirmed by fluorescence microscopy, which demonstrated the colocalization of nanocomplexes with both receptors on the plasma membrane as well as after internalization inside the cells. Type A scavenger receptors were recently shown to mediate the cellular uptake of PF14– siRNA nanocomplexes [22], PepFect15 nanoparticles with SCO [23], PF14–pDNA and NickFect nanoparticles with pDNA [24–26]. Therefore it is reasonable to assume that other gene delivery vectors of PepFect and NickFect family also utilize SR-As to gain access into cells. Internalization of scavenger receptors may also occur via different endocytosis pathways, such as clathrin-mediated endocytosis [14], caveolin-dependent endocytosis [27,28] and macropinocytosis [29]. In order to gain more insight into mechanism used by PF14–SCO nanocomplexes, we first aimed to pinpoint the particular endocytic pathways responsible for the uptake and functionality of the nanocomplexes as well as their intracellular trafficking in reporter HeLa pLuc705 cell line. Secondly, we aimed to obtain more detailed information about the interactions between the nanocomplexes and SRAs by using transmission electron microscopy (TEM) to examine how nanocomplexes interact with receptors and whether they can bind directly to receptors. We demonstrate that PF14–SCO nanocomplexes localize in very close proximity to both receptors, indicating that their binding to receptors might be direct. Surprisingly, in cells that were not treated with PF14–SCO nanocomplexes, SR-A3 and SR-A5 were not exposed on the plasma membrane, whereas their intracellular concentration was high. 2. Materials and methods 2.1. Cells and materials HeLa pLuc705 cells [30], kindly provided by Prof. R. Kole, were grown in humidified atmosphere containing 5% CO2 at 37 °C in Iscove's Modified Dulbecco's Medium (IMDM). Culture medium was supplemented with 10% fetal bovine serum (FBS), 100 IU/ml penicillin and 100 μg/ml streptomycin. PF14 was synthesized and purified as described before [13]. Phosphorothioate 2′OMe splice correction oligonucleotides with Cy5 labeling at the 5′ end were purchased from Microsynth AG, Switzerland. 2.2. Formation of PF14–SCO complexes To form complexes, Cy5 labeled SCO was mixed with PF14 (molar ratio MR 1:5) in MQ-water in 1/10 of the final treatment volume to reach 0.2 μM and 1 μM concentration, respectively. Complexes were formed for 30 min at room temperature. 2.3. Cellular uptake experiments 5 × 104 HeLa pLuc705 cells were seeded 2 days before the experiment into 24-well plate. Cells were washed and pre-treated for 30 min at 37 °C in serum containing IMDM with 10 μM chlorpromazine, 50 μM nystatin or 1 mM amiloride (all from Sigma-Aldrich). Lipofectamine 2000 (Invitrogen) was used as a control according to the manufacturer's protocol. After that PF14–SCO complexes were added and incubated for additional 4 h in serum containing media.

Transfection medium was replaced with fresh media and incubated for 20 h. Before the measurement, cells were washed twice with PBS and lysed with 0.1% Triton X-100 in PBS buffer for 30 min at 4 °C. Luciferase activity was measured using Promega's luciferase assay system on GLOMAX 96 microplate luminometer (Promega, Sweden). For control experiments, AF-594-conjugated transferrin, AF-555-labeled cholera toxin B subunit (CtxB) and 70 kDa TAMRA-dextran (all from Molecular Probes) were applied to cells at 20 μg/ml, 3 μg/ml, 0.25 mg/ml concentration to test efficacy of the inhibitors to block the uptake of respective endocytic markers. 2.4. RNAi experiments 4 × 104 HeLa pLuc705 cells were seeded onto 12-well culture plate 24 h before siRNA transfection. siRNAs against SR-A3 and SR-A5 (Ambion, USA) were used as a mixture at final concentration of 25 nM of each and cav-1 siRNA (Santa Cruz Biotechnology, Heidelberg, Germany) was used at final concentration of 100 nM. Negative control siRNA (Ambion, USA) was used at 50 nM or 100 nM concentration and siRNAs were transfected into HeLa cells using Lipofectamine RNAiMax (Life Technologies) according to the manufacturer's protocol. 48 h after the transfection cells were treated with pre-formed PF14–SCO nanocomplexes for 1 h, cells were then washed, lysed and analyzed with Tecan Infinite 200 M plate reader. 2.5. Localization of PF14–SCO complexes in relation to Golgi apparatus, caveolin-1 and LAMP2 5 × 104 HeLa pLuc705 cells were seeded onto round glass coverslips in 24-well plate 2 days before the experiment. Cells were washed and incubated with complexes of 0.2 μM SCO and 1 μM PF14 for 30 min or 1 h at 37 °C in serum-free IMDM. Treated cells were then fixed with 4% paraformaldehyde in PBS for 30 min, washed, permeabilized and blocked with 5% non-fat dry milk solution in PBS for 30 min. Cells were then treated with anti-TGN46 rabbit polyclonal antibody (1:100, Abcam, UK) for 1 h at room temperature followed by incubation with Alexa Fluor 488-conjugated anti-rabbit secondary antibody (1:500, Invitrogen) for 30 min at RT. Colocalization of nanocomplexes with caveolin-1 was analyzed by using anti-caveolin-1 rabbit polyclonal antibody (1:100, BD Transduction Laboratories, Belgium) and Alexa Fluor 488-conjugated goat-anti-rabbit secondary antibody (1:500, Invitrogen) and LAMP2 was visualized with mouse polyclonal antibody LAMP2 (1:100, H4B4, DSHB) and with Alexa Fluor 488-conjugated goatanti-mouse secondary antibody (1:500, Invitrogen). After washing, the coverslips were mounted with Fluoromount G (Electron Microscopy Sciences, PA). Images were captured by Olympus FluoView FV1000 confocal laser scanning microscope using excitation at 488 nm (for TGN46, cav-1 and LAMP2) and 633 nm (for SCO-Cy5) and processed with Adobe Photoshop CS6. 2.6. Co-localization of PF14–SCO complexes with Golgi apparatus, endoplasmic reticulum, dextran and CellMask in live cells 7 × 104 HeLa pLuc705 cells were seeded onto 8-well chambered coverglasses (Lab-Tek, Nalge Nunc International, NY) 2 days prior to experiment. At 80% of confluence, cells were washed and incubated in serum-free media for 2 h. After that cells were pre-incubated with Golgi marker 0.5 μM BODIPY-TR-C5 ceramide (Molecular Probes, UK) in serum-free media for 30 min on ice. Cells were washed again and incubated with PF14–SCO complexes in serum-free media for an additional 30 min or 1 h at 37 °C. To visualize the localization of complexes in relation to endoplasmic reticulum, cells were incubated with PF14–SCO complexes and 0.2 μM Blue–White DPX (Molecular Probes, UK) in serum-free media for 30 min or 1 h at 37 °C. Images were captured by Olympus FluoView FV1000 confocal laser scanning microscope using excitation at 590 nm (for Golgi marker), 380 nm

C. Juks et al. / Biochimica et Biophysica Acta 1848 (2015) 3205–3216

(for ER marker) and 633 nm (for nanocomplexes), and processed with Adobe Photoshop CS6. Colocalization with the respective endocytosis markers was calculated by counting all internalized PF14–SCO nanocomplexes and the number of complexes that colocalized with a particular marker in 105 cells. Colocalization was calculated as the percentage of all internalized nanocomplexes. To examine the colocalization between 70 kDa TAMRA-dextran (Invitrogen) and nanocomplexes, cells were incubated simultaneously with pre-formed PF14–SCO nanocomplexes and 0.25 mg/ml dextran for 30 min in serum-free IMDM medium at 37 °C. Cells were then washed and examined with confocal laser scanning microscope using excitation at 559 nm (for TAMRA-dextran) and 633 nm (for nanocomplexes). To assess the localization of PF14–SCO nanocomplexes relative to the plasma membrane, cells were first incubated with PF14–SCO complexes in serum-free IMDM for 30 min or 1 h. Cells were then washed and incubated with CellMask Green plasma membrane stain (1:1000, Invitrogen) for 10 min at 37 °C, washed and

3207

images were captured with confocal laser scanning microscope using excitation at 488 nm (for CellMask) and 633 nm (for nanocomplexes) and processed with Adobe Photoshop CS6.

2.7. Visualization of PF14–SCO nanocomplexes in relation to vesicles of endo-lysosomal pathway in live cells HeLa pLuc705 were grown on chambered coverglasses as described above. At 80% confluence, cells were washed and incubated with PF14– SCO complexes or “naked” 1 μM SCO-Cy5 for 1 h, 2 h and 4 h in serumfree medium at 37 °C. 2 h prior to desired time point, cells were washed twice with PBS and incubated with 2 μM LysoSensor-DND-189 (Molecular Probes, UK) in serum-free media to distribute LysoSensor along endo-lysosomal pathway. Images were captured by Olympus FluoView FV1000 confocal laser scanning microscope using excitation at 458 nm (for LysoSensor) processed with Adobe Photoshop CS6.

Fig. 1. Cellular uptake of PF14–SCO complexes into HeLa cells. (A) HeLa pLuc705 cells were pretreated with chlorpromazine (Cpr, 10 μM), nystatin (Ny, 50 μM) or amiloride (Am, 1 mM) for 30 min before the addition of complexes of 0.2 μM SCO with 1 μM PF14. Luciferase activity was measured 24 h after the transfection. Results are presented as percentage of activity of PF14– SCO complexes without the presence of inhibitors (PF14). Error bars represent mean ± SEM. (B) HeLa pLuc705 cells were treated with scrambled siRNA (0.1 μM) or cav-1 siRNA (0.1 μM). 48 h after siRNA transfection, cells were incubated with PF14–SCO-Cy5 complexes for 1 h and Cy5 signal was measured with Tecan Infinite 200 M plate reader. Values represent means ± SEM of ≥3 experiments done in duplicate. **P b 0.01, Student's t-test. (C) HeLa pLuc705 cells were treated with nanocomplexes of 1 μM PF14 and 0.2 μM SCO-Cy5 for 30 min, fixed and caveolin was visualized using caveolin-1 primary antibody (1:100) and AF488-conjugated secondary antibody (1:500). To examine the colocalization of nanocomplexes with 70 kDa TAMRA-dextran, HeLa pLuc705 cells were co-incubated with PF14–SCO-Cy5 complexes and 0.25 mg/ml dextran for 30 min. Colocalization is visible as yellow color and is indicated by arrowheads. Scale bar 10 μm.

3208

C. Juks et al. / Biochimica et Biophysica Acta 1848 (2015) 3205–3216

Fig. 2. Localization of PF14–SCO complexes in HeLa cells in relation to Golgi apparatus and endoplasmic reticulum. (A, B) HeLa pLuc705 cells were incubated with complexes of 1 μM PF14 with 0.2 μM SCO-Cy5 (red) for 30 min (A) or 60 min (B). Cells were then fixed and the Golgi region was visualized by antibody staining against the trans-Golgi network (TGN) (green). (C, D) HeLa pLuc705 cells were pre-treated with BODIPY TR-ceramide (red) for 30 min on the ice before the addition of Cy5 labeled SCO complexes (green pseudo color) with PF14 and incubated for an additional 30 min (C) or 60 min (D). (E–F) HeLa pLuc705 cells were pre-treated with ER marker Blue–White DPX (red pseudo color) for 30 min on the ice before the addition of Cy5-labeled SCO complexes (green pseudo color) with PF14 and incubated for an additional 30 min (E) or 60 min (F). Negligible fraction of the complexes was trafficked to Golgi apparatus after 30 min (A, C, insets) and 60 min (B, D, insets). No colocalization was found with ER and complexes neither after 30 min (E) nor after 60 min (F) incubation time. Scale bar 10 μm.

2.8. Formation of PF14–SCO complexes for electron microscopy SCOs were labeled with Nanogold™ (Nanoprobes, Yaphank, NY) and specimens were prepared for TEM as described previously [31]. To form complexes, nanogold-labeled SCO was mixed with PF14 (MR 1:5) in MQ-water in 1/10 of the final treatment volume to reach 0.2 μM and 1 μM concentration, respectively. Complexes were formed for 30 min at room temperature. Biotinylated SCO (0.2 μM) (Microsynth AG, Switzerland) was first associated with 10 nm NA–Au conjugate [6] (in 3:1 molecular ratio) for 30 min at 37 °C, followed by complexation with PF14 (1 μM) in Milli-Q water in 1/10 of the final treatment volume.

permeabilization cells were incubated with goat anti-SCARA3 (G-16, 1:50; Santa Cruz Biotechnology, Heidelberg, Germany) or rabbit antiSCARA5 (HPA024661, 1:50; Sigma-Aldrich, Taufkirchen, Germany) polyclonal antibodies followed by the treatment with 6 nm or 10 nm gold tag labeled Protein-G (1:40) (Aurion Immuno Gold Reagents & Accessories, Netherlands) for an additional hour. After postfixation with 1% osmium tetroxide in 0.1 M sodium cacodylate buffer, samples were dehydrated and embedded in TAAB 812 epoxy resin (TAAB Laboratories Equipment Ltd., UK). Ultra-thin sections were cut in parallel with coverslips, poststained with uranyl acetate and lead citrate and examined with FEI Tecnai G2 Spirit transmission electron microscope at 120 kV. Electron microphotos were analyzed and processed with Adobe Photoshop CS6. Quantifications were made by counting 50 PF14– pDNA particles or receptor complexes from 3 independent experiments.

2.9. Electron microscopy and pre-embedding immunolabeling HeLa pLuc705 cells were seeded onto round glass coverslips in 35 mm cell culture dishes and grown to 80–90% confluence. Cells were incubated with PF14–SCO nanocomplexes for 1 h at 37 °C. After that, cells were washed and fixed for immunoelectron microscopy with PLP fixative [32] at room temperature for 2 h. Cells were permeabilized with 0.01% saponin in 0.1 M sodium phosphate buffer (pH 7.4) supplemented with 0.01% BSA for 8 min. After

3. Results 3.1. PF14–SCO complexes induce splice correction after internalization via macropinocytosis and caveolin-mediated endocytosis Previously, we have shown that PF14–SCO complexes are recognized and endocytosed by class A scavenger receptors [21] but the

C. Juks et al. / Biochimica et Biophysica Acta 1848 (2015) 3205–3216

3209

Fig. 3. Localization of PF14–SCO nanocomplexes in cells in relation to acidic endosomes and lysosomes. HeLa pLuc705 cells were incubated with 1 μM PF14 complexed with 0.2 μM Cy5labeled SCO (A–C) or with “naked” 1 μM Cy5-labeled SCO (red color) (D–F) for 1 h (A, D), 2 h (B, E) and 4 h (C, F) and 2 μM LysoSensor DND-198 (green) to visualize the trafficking of complexes or “naked” SCO in endo-lysosomal pathway. Colocalization between PF14–SCO complexes or “naked SCO” (red) and LysoSensor-positive vesicles (green) is visible as yellow signal and indicated by arrowheads. Scale bar 10 μm.

exact uptake route was not determined. To examine, which internalization modes used by SR-As are responsible for the biological activity in PF14-mediated SCO delivery, we used pharmacological inhibitors to block the relevant endocytosis pathways. HeLa pLuc705 cells were treated with a battery of commonly used inhibitors for 30 min prior addition of PF14–SCO nanocomplexes. Cells were treated with chlorpromazine to suppress clathrin-dependent endocytosis [33], with nystatin to inhibit caveolin-dependent route [34] and amiloride to interfere with macropinocytosis [35]. The efficacy of the inhibitors to inhibit the uptake of the respective endocytic markers, transferrin,CtxB and dextran, was assessed in parallel, and all used inhibitors suppressed the internalization of markers of endocytosis (Fig. S1). In the absence of inhibitors PF14–SCO complexes induced more than a 100-fold increase in luciferase expression due to the splice-switching activity. The most potent inhibitor was amiloride that reduced SCO mediated splice correction up to 80% (Fig. 1A), suggesting that PF14–SCO complexes are mainly internalized via macropinocytosis. To better visualize the importance of macropinocytosis, we co-incubated cells with nanocomplexes and fluid-phase marker 70 kDa dextran and analyzed their location in HeLa cells. Colocalization studies showed that majority of cell surface bound PF14–SCO complexes colocalized with dextran (Fig. 1C, lower panels and insets 1–2). However, after 30 min of incubation some of the complexes had also entered the cells, where they were located in punctuate structures together with dextran. These results corroborate the involvement of macropinocytosis in endosomal uptake of PF14– SCO nanocomplexes. Nystatin reduced the splice correction about 30%, but chlorpromazine, an inhibitor of clathrin-mediated endocytosis had no effect on the uptake of the nanocomplexes (Fig. 1A). Immunofluorescence analysis revealed that a fraction of PF14–SCO complexes on the plasma membrane colocalized with caveolin-1 within 30 min (Fig. 1C upper panels and insets 1–2 arrowheads point to colocalization). This is in good concordance with results with pharmacological inhibitors of

endocytosis that pointed to a minor role of caveolin-dependent endocytosis in the uptake of PF14–SCO complexes. To further confirm the role of caveolin, we depleted caveolin-1 by using siRNA (Fig. 1B). Knockdown of caveolin-1 in HeLa cells reduced the uptake of PF14–SCO nanocomplexes by about 40% compared to control cells not treated with siRNA or cells treated with scrambled siRNA (Fig. 1B). These results suggest that SR-A3 and SR-A5 receptors induce the cellular uptake of PF14–SCO nanocomplexes mainly by triggering macropinocytosis and caveolin-dependent endocytosis, whereas clathrin-dependent internalization route does not seem to contribute to the uptake in these reporter cells.

3.2. PF14–SCO nanocomplexes are not targeted to endoplasmic reticulum and Golgi apparatus SR-As are known to translocate from the plasma membrane into different organelles, and in order to acquire information about the fate of PF14–SCO nanocomplexes after internalization, we examined their destination in cells. Material that enters cells via caveolin-dependent route, including SR-As, can be targeted to Golgi apparatus and endoplasmic reticulum (ER) [36]. After 30 min of incubation time, nanocomplexes were mainly found in vesicles near the plasma membrane (Fig. S2), and a smaller fraction of these were routed into vesicles in the perinuclear region of the cell (Fig. S2). However, only about 1.9% of the internalized complexes showed colocalization with Golgi network, which was visualized either by antibody staining against the trans-Golgi network (TGN) (Fig. 2A and insets 1–2 arrowheads point to colocalization) or with fluorescently tagged ceramide (Fig. 2C and insets 1–2) after 30 min incubation. Later, after 1 h, more complexes had shifted away from the plasma membrane towards the cell nucleus (Fig. S2), but their colocalization with Golgi was even reduced to 1.2% (Fig. 2B, D and insets 1–2).

3210

C. Juks et al. / Biochimica et Biophysica Acta 1848 (2015) 3205–3216

C. Juks et al. / Biochimica et Biophysica Acta 1848 (2015) 3205–3216

Mapping of the PF14–SCO intracellular localization in relation to endoplasmic reticulum revealed the complete lack of colocalization both after 30 min and 1 h (Fig. 2E, F and insets). The absence of colocalization demonstrates that despite being endocytosed by caveolin-dependent route and macropinocytosis, PF14–SCO nanocomplexes are not trafficked to ER or Golgi apparatus by SR-As. 3.3. PF14–SCO complexes avoid the endo-lysosomal pathway inside the cells Since the colocalization with ER and Golgi apparatus was negligible, we next examined the targeting of PF14–SCO complexes to endolysosomal pathway, using LysoSensor reporter to estimate the pH of vesicles that contained nanocomplexes. LysoSensor is internalized into cells via endocytosis and its fluorescence intensity inversely correlates with the pH value: more acidic vesicles emit higher fluorescence signal than vesicles with slight acidity. After incubation for 1 h, the majority of the PF14–SCO complexes were localized at the cell periphery in nonacidic vesicles that resembled early endosomes (Figs. 3A, S3). Later, the complexes reached deeper inside of the cells and also the number of vesicles that contained complexes increased, but still no significant colocalization with LysoSensor was observed (Figs. 3B, S3), suggesting that even within 2 h complexes reside mainly in neutral or slightly acidic vesicles. After 4 h the majority of complexes were still found in nonacidic vesicles (Fig. 3C) and only few of them had reached low-pH vesicles (Figs. 3C and insets 1–2; S3). Experiments with LysoSensor demonstrate that a very small fraction of PF14–SCO complexes was targeted to degradation within 4 h, which is in a good agreement with the data from experiment with pharmacological inhibitors of endocytosis. PF14 as a derivative of TP10 might interfere with the acidification of endocytic vesicles [37,38]; therefore we also assessed targeting of nanocomplexes to LAMP2-positive late endosomes/lysosomes as a function of incubation time (Fig. S4). However, we did not observe colocalization of nanocomplexes with LAMP2 after 1 h of incubation and only very rare colocalization with lysosomes was found after 2– 4 h of incubation (Fig. S4). Since the intracellular trafficking of nanocomplexes could be guided by both the nanocomplexes and receptors, we also analyzed the intracellular localization of “naked” SCO. After 1 h of incubation the “naked” SCO had distributed in HeLa cells in rather analogous manner with its nanocomplexes with PF14, and it was mainly found in vesicles close to the plasma membrane. These vesicles were mostly non-acidic and did not reveal LysoSensor signal (Figs. 3D, S3). However, at later time points the character of vesicles entrapping the complexes and free nucleic acid started to differ. After 2 h, the major fraction of “naked” SCO had reached the LysoSensor-positive vesicles with the dropped pH (Figs. 3E and insets 1–2; S3). In time, the intensity of the LysoSensor signal in vesicles with SCO increased even more, and in parallel, the number of SCO-containing acidic vesicles increased. After 4 h, almost all the SCOs in cells were sequestered into lysosomes (Figs. 3F and insets 1–2; S3). This was further confirmed by analyzing colocalization with LAMP2, which showed a major colocalization between the LAMP2-positive vesicles and SCO after 4 h of incubation (Fig. S4). These results suggest that although “naked” oligonucleotide at high concentration could associate with SR-A3 [21] and enter cells, after internalization they are targeted to lysosomes for degradation. It seems reasonable to speculate that PF14 packs SCO into particles for better recognition by SR-As, helps to interfere with the endosomal acidification and promotes endosomal escape to ensure the functionality of the cargo.

3211

3.4. Escape of PF14–SCO nanocomplexes from endosomes Since PF14–SCO nanocomplexes were not targeted to acidic vesicles and potently corrected splicing of luciferase mRNA, we next examined their intracellular trafficking by TEM. We analyzed when the endosomal escape starts, which type of vesicles preferentially leak, and whether the SCO could reach the nucleus. Although PF14–SCO nanocomplexes have internalized into cells already within 30 min after the transfection, the number of complexes in cells was substantially higher after 1 h and it continued increasing. The majority of complexes localized in cells in large vesicles (about 300 nm in diameter) that resembled multivesicular bodies, but were still rather “electron-light”. Inside vesicles, the nanocomplexes typically localized close to the vesicle's membranes on the electron dense background. In some loci the nanocomplexes were in contact with the endosomal membrane and induced breaches in membranes (Fig. 4A inset, membrane breaches are indicated with arrowheads). Starting from 1 h incubation, we detected endosomal vesicles with fragmentary membrane and at the same time nanocomplexes were found escaping from endosomes and distributing to the cytosol (Fig. 4A). Although the escape of nanocomplexes from endosomes started from 1 h after uptake, in rare cases we detected SCO in cytosol already in 30 min. Later, 4 h after transfection, the nanocomplexes had shifted to the perinuclear region and higher amount of complexes was found in the cytosol (Fig. 4B and insets 1–2 arrowheads indicate to complexes; C). To estimate the number of nanocomplexes that are entrapped into endosomes and the number of which have been escaped to the cytosol, over 2000 particles from 70 TEM images were counted. The fraction of nanocomplexes that had escaped form endosomes to the cytosol over 4 h of incubation time was about 8% (Fig. 4E). In cytosol, the PF14– SCO complexes were detectable as small clusters, containing a few SCO molecules, whereas the nanocomplexes on the plasma membrane or in the vesicles were substantially bigger containing tens of SCO molecules. By this time point a few SCO molecules had also translocated into nucleus (Fig. 4D, inset). Although the number of nanocomplexes detected in the nucleus after 4 h-incubation was very low, this result is in good concordance with earlier results. Ezzat et al. examined the kinetics of PF14–SCO mediated splice-correction and found that about 60% of maximal splice correction was obtained already 8 h after the transfection [13]. In contrary to nanocomplexes “naked” SCO without the peptide did not form such clusters and they are rather associated at the plasma membrane as individual particles (Fig. S5). Despite of having no transport molecule, a small amount of “naked” SCO is still internalized with low efficiency into cells via endocytosis and these molecules could be found in endosomal vesicles (Fig. S5). However, the amount of “naked” SCO detected by TEM on the plasma membrane and in the vesicles in cytoplasm very small compared to its complexes with the peptide and they were much rarer than we observed by confocal microscopy (Figs. 3 and S4). This implicates that the confocal microscopy could overestimate the association of “naked” SCO with cells, or the less tightly bound oligonucleotides are more easily removed upon specimen preparation by TEM. However, after 4 h of incubation still “naked” SCO was detected in vesicles, which had increased size and contained numerous intraluminal vesicles (Fig. S5). Moreover, we never detected “naked” SCO outside endosomal vesicles in cell cytoplasm even after 4 h. 3.5. PF14–SCO complexes can associate with SR-A receptors Using fluorescence microscopy we have shown that PF14–SCO complexes associate with scavenger receptors on the plasma membrane and

Fig. 4. Escape of PF14–SCO nanocomplexes from endosomes and translocation into nucleus. HeLa pLuc705 cells were incubated with nanocomplexes of 1 μM PF14 with 0.2 μM SCO (labeled with 1.4 nm gold tag) at 37 °C for 1 h (A) and 4 h (B, C, D) and their localization was analyzed by TEM. Localization of PF14–SCO nanocomplexes inside the vesicles after 1 h of incubation and endosomal escape (pointed with arrows) (A). Localization of nanocomplexes inside the cells in multivesicular bodies (B) and in the perinuclear region (C) after 4 h of uptake. A significant fraction of the complexes were detected in the cytosol (B and C insets, indicated with arrowheads) and some of them had reached into the nucleus upon 4 h-incubation (D, inset). Scale bar 500 nm, n-nucleus. (E) Shows the number of PF14–SCO nanocomplexes which are entrapped into the endosomes and the amount of that have been escaped to the cytosol.

3212

C. Juks et al. / Biochimica et Biophysica Acta 1848 (2015) 3205–3216

their colocalization is retained after internalization in vesicles [21]. In order to gain a more detailed insight into how these nanoparticles interact with receptors and whether these complexes can directly associate with SR-As we used TEM. To better distinguish and visualize the interactions between SR-A receptors and nanoparticles by TEM, we labeled the PF14–SCO nanoparticles with 10 nm colloidal gold and SR-A3 and SR-A5 receptors were visualized by using respective primary antibodies and secondary antibodies that carried a 6 nm gold tag. Binding of PF14–SCO complexes to the plasma membrane induced their association into small clusters, which typically contained 3–10 gold particles (Fig. 5A). These clusters were often found on the plasma membrane ruffles (Fig. 5A), which further support the involvement of macropinocytosis in the uptake of PF14-nanocomplexes. After 1 h-incubation, the complexes had entered the cells and localized mainly in multivesicular bodies, where they still retained association into smaller cluster like on the plasma membrane (Fig. 5B). However, at this time point we could detect only negligible amount of such complexes in the cytosol out of vesicles. Next we analyzed the localization of nanocomplexes in relation to SR-A receptors. After 1 h of incubation time we could observe both type 3 and 5 of SR-A receptors in very close proximity to complexes on the plasma membrane (Fig. 5C, D, arrowheads point to receptors), and also induction of endosomal invaginations. Nanocomplexes and receptors were located very closely in induced invaginations (Fig. 5C inset, arrowhead indicate to receptor), indicating that PF14–SCO complexes associate with scavenger receptors which assist their internalization into cells. Only a few receptors are required for efficient internalization and already within 1 h of incubation time, nanocomplexes and SR-A receptors were found in endocytic vesicles (Fig. 5E, F). PF14–SCO complexes are preferentially concentrated at the limiting membrane of endocytic vesicle membrane, where we typically also found a few receptors (Fig. 5E and F insets, arrowheads indicate to receptors). To further support the involvement of scavenger receptors in the uptake of PF14-nanocomplexes we next knockeddown both receptors in HeLa cells. Treatment of cells with siRNAs against SR-A3 and SR-A5 (25 nM of each) reduced the internalization of PF14–SCO complexes about 50%, whereas treatment of cells with scrambled siRNA had no effect on the uptake process of nanocomplexes (Fig. 5G). These results further corroborate that scavenger receptors are needed for efficient uptake of PF14-nanocomplexes. 3.6. SR-As are not exposed on the plasma membrane of control cells The colocalization of PF14–nucleic acid complexes with SR-A3/5 on the plasma membrane and ruffles/filopodia impelled us to map the localization of these receptors in HeLa pLuc705 cells in detail. Surprisingly, we could not detect receptors on the surface of control cells that were not treated with PF14–nucleic acid nanocomplexes (Fig. 6A, B). This finding was also supported by confocal microscopy, where in control cells, SR-A3 and SR-A5 receptors were not on the plasma membrane; however, treatment with PF14–SCO complexes induced translocation of both receptors to the cell surface (Fig. S6). Still, receptors were easily detectable inside the cells in substantial amounts. Both, SR-A3 and SRA5 receptors localized in small vesicles very close to the plasma membrane and within more central membranous structures as revealed by TEM analysis (Fig. 6A, B). Remarkably, immunological detection typically revealed the presence of clusters that contain 2–5 gold particles consistent with the trimeric architecture of these receptors. However, when cells were incubated with complexes of PF14 and SCO or pDNA, SR-As were translocated from the cytoplasm to the plasma membrane

(Fig. 6C, D) as evidenced by both confocal microscopy and TEM analysis. TEM images revealed that after induced translocation to the plasma membrane, SR-As can form clusters of 5–10 receptor monomers (Fig. 6C, D). After exposure of HeLa cells to PF14–nucleic acid nanocomplexes, SR-A3 and SR-A5 were often detected on the plasma membrane on filopodia/ruffles, the structures that were absent in control cells. It is likely that the first interaction of PF14–nucleic acid nanocomplexes with scavenger receptors takes place (at least partially) on the membrane protrusions and later complexes are transported towards the cell body for endocytosis. 4. Discussion CPPs have turned into powerful vehicles for the delivery of various cargos into cells and they have an immense potential to deliver different bioactive molecules both in vitro and in vivo. Despite being under investigation for more than 20 years, their mechanisms are not fully understood and several studies have shown controversial results. Recently, we have shown that some CPP-based delivery systems of nucleic acid can act using specific receptors on the plasma membrane to induce endocytosis [21]. Namely, class A scavenger receptor subtypes SR-A3 and SR-A5 were necessary for PepFects and NickFects mediated oligonucleotide delivery [21–26]. Scavenger receptors are known to bind a wide variety of negatively charged macromolecules [39] and to induce their internalization using all endocytic pathways characterized so far [14]. To gain better insight into SR-As as targets for the CPP based nucleic acid delivery, we aimed to study further the uptake mechanisms and intracellular trafficking of PF14–SCO complexes in HeLa pLuc705 cells. We also sought for more detailed information how PF14–SCO complexes associate with SR-As and whether the interaction is direct. To elucidate which endocytic routes are responsible for the activity of PF14–SCO complexes, we used pharmacological inhibitors. Amiloride, an inhibitor of macropinocytosis, reduced the splicecorrecting effect of PF14–SCO nanocomplexes at the greatest extent. Macropinocytosis has previously been demonstrated as the main entry route for arginine-rich CPPs such as nona-arginine and Tatpeptide [8,40], and also for NickFect mediated cargo delivery [24]. Inhibition of caveolin-mediated endocytosis with nystatin or with caveolin1 specific siRNA reduced the activity of nanocomplexes about 30% and 40% respectively, suggesting that at least two different endocytic pathways are active in parallel. On the other hand, inhibition of clathrinmediated endocytosis, that has been considered to be the main uptake route for receptor dependent endocytosis, had no influence on the biological activity of PF14–SCO complexes. Our results seem to contradict recently published data, where Bitsikas et al. have proposed that 95% of the plasma membrane proteins are internalized into cells via rapid clathrin-dependent endocytosis [41]. Biotinylated cell surface proteins, examined in this paper, are probably taken up via constitutive way within very few seconds, as they just follow the uptake without any stimulation. However, in a case of CPP-nanoparticle uptake, endocytosis is rather a stimulated process occurring after binding of complexes to the plasma membrane and translocation of receptors to cell surface. The involvement of caveolae-mediated endocytosis has been reported in the cellular uptake of TP10-cargo delivery, the parent peptide of PepFects, as well as in the uptake of PF14–pDNA and NickFect-cargo complexes into HeLa cells via SR-As [24,26]. Caveolae-dependent scavenger receptor class A mediated endocytosis was recently found to be the major uptake route for spherical nucleic acids [28]. The caveolar pathway is also responsible for the cellular uptake and biological

Fig. 5. Colocalization of PF14–SCO nanocomplexes with SR-A receptors in cells. HeLa pLuc705 cells were incubated with preformed complexes of PF14 and biotinylated-SCO tagged with 10 nm colloidal gold particle for 1 h at 37 °C. SR-A3 (C, E) and SR-A5 (D, F) receptors were visualized with anti-Scara3 and anti-Scara5 primary antibodies and Protein-G coupled to 6 nm gold particles. Interactions of PF14–SCO complexes with the plasma membrane (A) and localization inside the cells (B). Localization of PF14–SCO complexes in relation to SR-A3 (C, E, insets) and SR-A5 (D, F, insets) on the plasma membrane and in the cells. Arrowheads point to receptors that localize in close proximity to nanocomplexes. Scale bar 500 nm. (G) The uptake of complexes by siRNA-treated cells. HeLa cells treated with scrambled siRNA (50 nM) or specific siRNAs against SR-A3 and SR-A5 (25 nM of each) were incubated with complexes of PF14 (1 μM) and SCO-Cy5 (0.2 μM) for 1 h and the fluorescence of cells was quantified by Tecan Infinite 200 M plate reader. Values represent means ± SEM of ≥3 experiments done in duplicate. ***P b 0.001, Student's t-test.

C. Juks et al. / Biochimica et Biophysica Acta 1848 (2015) 3205–3216

3213

3214

C. Juks et al. / Biochimica et Biophysica Acta 1848 (2015) 3205–3216

Fig. 6. SR-A receptors are targeted to the cell surface upon treatment with PF14–SCO nanocomplexes. HeLa pLuc705 cells were incubated with PF14 nanocomplexes with SCO for 1 h at 37 °C. SR-A receptors were visualized with anti-Scara3 and anti-Scara5 primary antibodies and 10 nm gold tag labeled Protein-G. In control cells receptors localized only inside cells (A, B). Incubation of HeLa cells with PF14–SCO nanocomplexes translocated both receptors to the plasma membrane (C, D).

activity of conjugates of splice switching oligonucleotides with cell penetrating peptide in mdx skeletal muscle cells and murine model of Duchenne Muscular Dystrophy [42]. After budding from the plasma membrane, CPP-cargo complexes remain entrapped in endocytic vesicles. Since the exact uptake mechanism used by CPPs may depend on the cargo [9,43,44] and the uptake route determines further intracellular localization, and the fate of endocytosed material, we next followed the intracellular trafficking of the complexes after the internalization. Papers, published in the field of scavenger receptors have mainly focused on their role in different cellular processes, among which some papers have also established the uptake route, but exact cellular trafficking is not well studied so far. In general, binding of ligand to scavenger receptors induces internalization via endocytic pathway(s) and accumulation into endosomes, where the ligand is dissociated from receptor in acidic conditions of the endosomes and is subsequently transported to lysosomes for degradation. Once inside the cells, receptor–ligand complexes are also shown to trigger several intracellular signaling pathways [17,19], however, less discussed is the fate of internalized receptors. One possibility is that receptors are recycled back to the cell surface, where they can bind and internalize next ligands, or they can undergo lysosomal degradation, and thereby control their activity in the cells [14,45]. Several studies have demonstrated that the material endocytosed via caveolin-mediated pathway could be directed to the endoplasmic reticulum and Golgi apparatus [36,46]. To examine, whether PF14– SCO nanocomplexes are also targeted to these organelles, we used confocal microscopy and analyzed the colocalization with ER and Golgi in HeLa cells. However, we did not detect any colocalization of PF14–SCO nanocomplexes and ER, and observed only a negligible colocalization with Golgi apparatus shortly after the internalization. At later time points, the number of the nanocomplexes localizing to Golgi decreased and after 1 h of incubation only about 1% of the complexes were found in Golgi complex. Also our TEM data did not show the accumulation of PF14–SCO nanocomplexes to the Golgi apparatus. Our results contradict

the fact that the contents of the caveosomes are targeted to the ER or Golgi apparatus and allowed us to suggest that some other trafficking routes were utilized. Analogous localization pattern was described earlier in the case of TP10-mediated protein delivery via caveolindependent endocytosis, where only a negligible fraction of the cargo was found to colocalize with Golgi apparatus [32]. Another major intracellular trafficking route is the endo-lysosomal pathway. During the endosome maturation the internalized material is transported to gradually acidifying vesicles that finally mature to lysosomes, where the degradation of endocytosed material occurs. Endosomal entrapment is believed to be the main obstacle that limits the application of bioactive nanocomplexes, whose target is in the cytoplasm or the nucleus. To clarify whether PF14–SCO complexes are targeted along the endo-lysosomal pathway, we used LysoSensor to distinguish the acidic vesicles among endocytic organelles. After transfection, we observed punctuate distribution of the complexes in the cytosol, but no significant colocalization with LysoSensor was detected. In time the number of the complexes that had entered the cells increased, but the majority of these avoided transfer into endolysosomal pathway, even after 4 h. Thus, it can be suggested that the nanocomplexes are located in rather non-acidic or moderately acidic vesicles, which is in good agreement with the results of uptake experiments, which demonstrated the prevalence of caveolin-dependent endocytosis and macropinocytosis in PF14 mediated SCO delivery. Caveosomes, unlike other endocytic organelles, are considered to possess a non-acidic pH [46]. We speculate that the nature of caveosomes and slower maturation rate than observed for clathrin-mediated endocytosis grant the complexes an opportunity to escape into cytosol before the endosomal acidification and fusion with degradative lysosomes occur. Our results also correlate with earlier studies, where transportan and TP10 were found to be targeted to vesicles, whose pH had only slightly decreased [37]. Very recent experiments with lipid-siRNA nanoparticles showed that endosomal escape only occurred at the specific stage of endosomal trafficking [47]. siRNA release was found to occur from

C. Juks et al. / Biochimica et Biophysica Acta 1848 (2015) 3205–3216

moderately acidic vesicles, that had the features of both the early and late endosomal vesicles. Some studies also demonstrated that efficient endosomal escape may occur from macropinosomes as these vesicles are considered to be leakier than other endosomal compartments [48]. Second generation CPPs, like PepFects, are able to promote efficient endosomal leakage due to their modifications, that leads to the liberation of the nucleic acid cargo from entrapment and to high transfection efficiencies [13,49]. Spherical nucleic acids (SNAs), which are also taken up by cells via SR-As and caveolae-mediated pathway, are shown to localize into early endosomes within 1–2 h of incubation time, in analogy with our results [28]. However, in time these vesicles mature into late endosomes, where SNAs stay for at least 24 h, as demonstrated by the authors [50]. Still, SNAs did not follow the classical endo-lysosomal pathway, and showed no colocalization with lysosomal marker LAMP1 or maturation into lysosomes [50]. These findings also support the idea, that CPPs are required for promoting endosomal escape and biological activity of cargo molecules. Furthermore, SNAs did not reveal colocalization or targeting to trans-Golgi network [50], exactly as we showed for PF14–SCO nanoparticles. It seems that SNAs and CPP-SCO nanocomplexes share similar uptake routes and cellular trafficking, at least at early time points. Surprisingly we found that “naked” oligonucleotides at higher concentration (1 μM) were also able to internalize into cells, but already after 2 h the majority of SCO in cells was found in LysoSensor positive vesicles. Later, the amount of SCO entrapped into endosomes increased and these vesicles became more acidic, suggesting that the oligonucleotides were mainly targeted to lysosomes. Although several reports have demonstrated the ability of “naked” oligonucleotides to promote the biological activity [51,52], it perhaps occurs at lower levels, at high concentration and preferentially with extensively modified oligonucleotides. Next we aimed to study how PF14–SCO nanocomplexes interact with receptors and whether these complexes can bind directly to SR-As. Previously, we demonstrated by fluorescence microscopy that the nanocomplexes colocalize with SR-As on the cell surface, and also inside the cells. In order to gain a more detailed insight, we used TEM utilizing differential labeling strategy for distinguishing the SR-As and the nanocomplexes. We found that PF14–SCO nanocomplexes reside in the close proximity to receptors on the plasma membrane, suggesting that the nanocomplexes could directly associate with receptors and promote endocytosis. After internalization, SR-As and nanocomplexes were mainly entrapped in the endocytic vesicles, retaining colocalization. Both SR-A3 and SR-A5 receptors were localized at the boundaries of the complexes close to the membrane of endosomal vesicles. Interestingly, treatment of cells with PF14–SCO nanocomplexes induced the formation of plasma membrane protrusions; however, these protrusions did not form in the untreated control cells. Induction of filopodia and ruffles by the nanocomplexes suggests the triggering of macropinocytosis; and indeed it was found to play a significant role in the biological activity of PF14– SCO complexes. Our results are supported by recently published data, where over-expression of both SR-A3 and SR-A5 in HeLa cells resulted in increased uptake of CPP-plasmid nanocomplexes [22]. Surprisingly, SR-As were not exposed on the plasma membrane of untreated cells, as observed with both TEM and confocal microscopy. On the other hand, the receptors were nicely detectable inside the cells in smaller vesicles or bound to membranous structures. SR-As are trimeric transmembrane receptors, and we often found these labeled with 2–3 nanogold tags in cells by immuno-TEM. However, upon treatment of cells with PF14–SCO nanocomplexes the receptors translocated to the plasma membrane, where these clustered into bigger assemblies that could contain up to 5–10 nanogold particles, implying on the necessity for SR-A clustering for induction of endocytosis. SR-As are reported to be involved in the uptake of various bioactive molecules, but less is known about their signaling and the mechanisms that could be responsible for their targeting to the plasma membrane. We speculate that calcium ions can be involved in this signaling, because calcium have been described to activate some endocytic receptors

3215

[53]. In addition, it has been recently demonstrated that CPPs can modulate intracellular calcium levels or even cause an efflux of calcium into the cytosol upon binding to the plasma membrane, after which membrane repair response is triggered [54]. However, the exact role of calcium, if any, in targeting of SR-As to the plasma membrane needs to be further examined. Alternatively, other pattern recognition receptors that are expressed on the cell membrane, such as TLR2 and TLR4, might be responsible for initial recognition and signaling to mobilize SR-As to the cell surface. Amiel et al. have demonstrated that TLR2 and TLR4, are expressed on the cell surface and and regulate SR-A-mediated phagocytic uptake by MyD88 signaling [55]. Additionally, it has been shown that lipopolysaccharide (LPS) stimulation promotes SR-A association with TLR4 [56]. Thus, it is feasible to speculate that, at least partially, TLR signaling might contribute to the mobilization of SR-As in the presence of certain ligands. Several scavenger receptors including SRAs were shown to function as heteromultimeric signaling complexes, called signalosomes, with a wide range of transmembrane proteins that include TLRs, integrins and tetraspanins [39]. It is not clear whether the association of scavenger receptors with other receptors is constitutive or occurs only in response to exogenous ligands. However, the fact that such association occurs only in the presence of exogenous ligands makes the system more flexible allowing a finite number of scavenger receptors to contribute to different types of interactions only when needed [39]; and this can explain our observation that their mobilization to the cell surface occurs only in the presence of the nanocomplexes. In conclusion, we have characterized the cellular trafficking of PF14– SCO nanocomplexes in HeLa pLuc705 cells. First the association of the nanocomplexes to SR-A3 and SR-A5 induces their internalization via macropinocytosis and caveolin-mediated endocytosis. After internalization, the complexes mainly localize in endocytic vesicles with nearneutral pH that can be early endosomes and/or multivesicular bodies, where from the nanocomplexes could escape to the cell cytosol. The “naked” oligonucleotides, in contrary, were found to be trafficked differently and these remained entrapped in endocytic vesicles. Finally, we demonstrated the interaction of PF14–nucleic acid nanocomplexes with SR-A receptors and that their colocalization was retained even inside the cells at least for 1 h; and that for the internalization only a few trimeric receptors were needed. Surprisingly, SR-A3 and SR-A5 were not exposed on the plasma membrane of untreated HeLa cells, but the treatment with nanocomplexes induced their translocation to the cell surface. Whether other pattern recognition receptors might also be involved in this process is not clear and warrants further examination. Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.bbamem.2015.09.019. Transparency Document The Transparency document associated with this article can be found, in the online version. Acknowledgments The study was supported by grants from the Estonian Science Foundation (ESF 8705), the Estonian Ministry of Education and Research (0180019s11), and European Union Regional Development Fund (grant EU30020) through the Competence Centre on Reproductive Medicine and Biology. References [1] F. Said Hassane, A.F. Saleh, R. Abes, M.J. Gait, B. Lebleu, Cell penetrating peptides: overview and applications to the delivery of oligonucleotides, Cell. Mol. Life Sci. 67 (2010) 715–726. [2] M. Zorko, Ü. Langel, Cell-penetrating peptides: mechanism and kinetics of cargo delivery, Adv. Drug Deliv. Rev. 57 (2005) 529–545.

3216

C. Juks et al. / Biochimica et Biophysica Acta 1848 (2015) 3205–3216

[3] M. Mäe, Ü. Langel, Cell-penetrating peptides as vectors for peptide, protein and oligonucleotide delivery, Curr. Opin. Pharmacol. 6 (2006) 509–514. [4] P. Lundin, H. Johansson, P. Guterstam, T. Holm, M. Hansen, Ü. Langel, S. El Andaloussi, Distinct uptake routes of cell-penetrating peptide conjugates, Bioconjug. Chem. 19 (2008) 2535–2542. [5] J.P. Richard, K. Melikov, H. Brooks, P. Prevot, B. Lebleu, L.V. Chernomordik, Cellular uptake of unconjugated TAT peptide involves clathrin-dependent endocytosis and heparan sulfate receptors, J. Biol. Chem. 280 (2005) 15,300–15,306. [6] K. Padari, P. Säälik, M. Hansen, K. Koppel, R. Raid, Ü. Langel, M. Pooga, Cell transduction pathways of transportans, Bioconjug. Chem. 16 (2005) 1399–1410. [7] A. Fittipaldi, A. Ferrari, M. Zoppé, C. Arcangeli, V. Pellegrini, F. Beltram, M. Giacca, Cell membrane lipid rafts mediate caveolar endocytosis of HIV-1 Tat fusion proteins, J. Biol. Chem. 278 (2003) 34,141–34,149. [8] G. Tanaka, I. Nakase, Y. Fukuda, R. Masuda, S. Oishi, K. Shimura, Y. Kawaguchi, T. Takatani-Nakase, Ü. Langel, A. Gräslund, K. Okawa, M. Matsuoka, N. Fujii, Y. Hatanaka, S. Futaki, CXCR4 stimulates macropinocytosis: implications for cellular uptake of arginine-rich cell-penetrating peptides and HIV, Chem. Biol. 19 (2012) 1437–1446. [9] F. Duchardt, M. Fotin-Mleczek, H. Schwarz, R. Fischer, R. Brock, A comprehensive model for the cellular uptake of cationic cell-penetrating peptides, Traffic 8 (2007) 848–866. [10] A. Walrant, C. Bechara, I.D. Alves, S. Sagan, Molecular partners for interaction and cell internalization of cell-penetrating peptides: how identical are they? Nanomedicine (Lond.) 7 (2012) 133–143. [11] A. Rydström, S. Deshayes, K. Konate, L. Crombez, K. Padari, H. Boukhaddaoui, G. Aldrian, M. Pooga, G. Divita, Direct translocation as major cellular uptake for CADY self-assembling peptide-based nanoparticles, PLoS One 6 (2011), e25924. [12] H. Hirose, T. Takeuchi, H. Osakada, S. Pujals, S. Katayama, I. Nakase, S. Kobayashi, T. Haraguchi, S. Futaki, Transient focal membrane deformation induced by arginine-rich peptides leads to their direct penetration into cells, Mol. Ther. 20 (2012) 984–993. [13] K. Ezzat, S. El Andaloussi, E.M. Zaghloul, T. Lehto, S. Lindberg, P.M.D. Moreno, J.R. Viola, T. Magdy, R. Abdo, P. Guterstam, R. Sillard, S.M. Hammond, M.J.A. Wood, A.A. Arzumanov, M.J. Gait, C.I.E. Smith, M. Hällbrink, Ü. Langel, PepFect14, a novel cell-penetrating peptide for oligonucleotide delivery in solution and as solid formulation, Nucleic Acids Res. 39 (2011) 5284–5298. [14] J.E. Murphy, P.R. Tedbury, S. Homer-Vanniasinkam, J.H. Walker, S. Ponnambalam, Biochemistry and cell biology of mammalian scavenger receptors, Atherosclerosis 182 (2005) 1–15. [15] A.M. Pearson, A. Rich, M. Krieger, Polynucleotide binding to macrophage scavenger receptors depends on the formation of base-quartet-stabilized four-stranded helices, J. Biol. Chem. 268 (1993) 3546–3554. [16] N. Platt, S. Gordon, Scavenger receptors: diverse activities and promiscuous binding of polyanionic ligands, Chem. Biol. 5 (1998) R193–R203. [17] S.J. DeWitte-Orr, S.E. Collins, C.M.T. Bauer, D.M. Bowdish, K.L. Mossman, An accessory to the ‘Trinity’: SR-As are essential pathogen sensors of extracellular dsRNA, mediating entry and leading to subsequent type I IFN responses, PLoS Pathog. 6 (2010), e1000829. [18] M.-C. Saleh, R.P. van Rij, A. Hekele, A. Gillis, E. Foley, P.H. O'Farrell, R. Andino, The endocytic pathway mediates cell entry of dsRNA to induce RNAi silencing, Nat. Cell Biol. 8 (2006) 793–802. [19] G.V. Limmon, M. Arredouani, K.L. McCann, R.A.C. Minor, L. Kobzik, F. Imani, Scavenger receptor class-A is a novel cell surface receptor for double-stranded RNA, FASEB J. 22 (2008) 159–167. [20] P.C. Patel, D.A. Giljohann, W.L. Daniel, D. Zheng, A.E. Prigodich, C.A. Mirkin, Scavenger receptors mediate cellular uptake of polyvalent oligonucleotide-functionalized gold nanoparticles, Bioconjug. Chem. 21 (2010) 2250–2256. [21] K. Ezzat, H. Helmfors, O. Tudoran, C. Juks, S. Lindberg, K. Padari, S. El-Andaloussi, M. Pooga, Ü. Langel, Scavenger receptor-mediated uptake of cell-penetrating peptide nanocomplexes with oligonucleotides, FASEB J. 26 (2012) 1172–1180. [22] S. Lindberg, J. Regberg, J. Eriksson, H. Helmfors, A. Muñoz-Alarcón, A. Srimanee, R.A. Figueroa, E. Hallberg, K. Ezzat, Ü. Langel, A convergent uptake route for peptide- and polymer-based nucleotide delivery systems, J. Control. Release 206 (2015) 58–66. [23] S. Lindberg, A. Munoz-Alarcon, H. Helmfors, D. Mosqueira, D. Gyllborg, O. Tudoran, Ü. Langel, PepFect15, a novel endosomolytic cell-penetrating peptide for oligonucleotide delivery via scavenger receptors, Int. J. Pharm. 441 (2013) 242–247. [24] P. Arukuusk, L. Pärnaste, H. Margus, N.K.J. Eriksson, L. Vasconcelos, K. Padari, M. Pooga, Ü. Langel, Differential endosomal pathways for radically modified peptide vectors, Bioconjug. Chem. 24 (2013) 1721–1732. [25] P. Arukuusk, L. Pärnaste, N. Oskolkov, D.-M. Copolovici, H. Margus, K. Padari, K. Möll, J. Maslovskaja, R. Tegova, G. Kivi, A. Tover, M. Pooga, M. Ustav, Ü. Langel, New generation of efficient peptide-based vectors, NickFects, for the delivery of nucleic acids, Biochim. Biophys. Acta Biomembr. 1828 (2013) 1365–1373. [26] K.L. Veiman, I. Mäger, K. Ezzat, H. Margus, T. Lehto, K. Langel, K. Kurrikoff, P. Arukuusk, J. Suhorutsenko, K. Padari, M. Pooga, Ü. Langel, PepFect14 peptide vector for efficient gene delivery in cell cultures, Mol. Pharmacol. 10 (2013) 199–210. [27] X.-D. Zhu, Y. Zhuang, J.-J. Ben, L.-L. Qian, H.-P. Huang, H. Bai, J.-H. Sha, Z.-G. He, Q. Chen, Caveolae-dependent endocytosis is required for class a macrophage scavenger receptor-mediated apoptosis in macrophages, J. Biol. Chem. 286 (2011) 8231–8239. [28] C.H.J. Choi, L. Hao, S.P. Narayan, E. Auyeung, C.A. Mirkin, Mechanism for the endocytosis of spherical nucleic acid nanoparticle conjugates, Proc. Natl. Acad. Sci. 110 (2013) 7625–7630. [29] W. Yao, K. Li, K. Liao, Macropinocytosis contributes to the macrophage foam cell formation in RAW264.7 cells, Acta Biochim. Biophys. Sin. 41 (2009) 773–780. [30] S.-H. Kang, M.-J. Cho, R. Kole, Up-regulation of luciferase gene expression with antisense oligonucleotides: implications and applications in functional assay development, Biochemistry 37 (1998) 6235–6239.

[31] N. Oskolkov, P. Arukuusk, D.-M. Copolovici, S. Lindberg, H. Margus, K. Padari, M. Pooga, Ü. Langel, NickFects, phosphorylated derivatives of transportan 10 for cellular delivery of oligonucleotides, Int. J. Pept. Res. Ther. 17 (2011) 147–157. [32] P. Säälik, K. Padari, A. Niinep, A. Lorents, M. Hansen, E. Jokitalo, Ü. Langel, M. Pooga, Protein delivery with transportans is mediated by caveolae rather than flotillin-dependent pathways, Bioconjug. Chem. 20 (2009) 877–887. [33] M. Al Soraj, L. He, K. Peynshaert, J. Cousaert, D. Vercauteren, K. Braeckmans, S.C. De Smedt, A.T. Jones, siRNA and pharmacological inhibition of endocytic pathways to characterize the differential role of macropinocytosis and the actin cytoskeleton on cellular uptake of dextran and cationic cell penetrating peptides octaarginine (R8) and HIV-Tat, J. Control. Release 161 (2012) 132–141. [34] G.M. Di Guglielmo, C. Le Roy, A.F. Goodfellow, J.L. Wrana, Distinct endocytic pathways regulate TGF-[beta] receptor signalling and turnover, Nat. Cell Biol. 5 (2003) 410–421. [35] M. Koivusalo, C. Welch, H. Hayashi, C.C. Scott, M. Kim, T. Alexander, N. Touret, K.M. Hahn, S. Grinstein, Amiloride inhibits macropinocytosis by lowering submembranous pH and preventing Rac1 and Cdc42 signaling, J. Cell Biol. 188 (2010) 547–563. [36] P.U. Le, I.R. Nabi, Distinct caveolae-mediated endocytic pathways target the Golgi apparatus and the endoplasmic reticulum, J. Cell Sci. 116 (2003) 1059–1071. [37] H. Räägel, P. Säälik, M. Hansen, Ü. Langel, M. Pooga, CPP–protein constructs induce a population of non-acidic vesicles during trafficking through endo-lysosomal pathway, J. Control. Release 139 (2009) 108–117. [38] H. Räägel, M. Hein, A. Kriiska, P. Säälik, A. Floren, Ü. Langel, M. Pooga, Cellpenetrating peptide secures an efficient endosomal escape of an intact cargo upon a brief photo-induction, Cell. Mol. Life Sci. 70 (2013) 4825–4839. [39] J. Canton, D. Neculai, S. Grinstein, Scavenger receptors in homeostasis and immunity, Nat. Rev. Immunol. 13 (2013) 621–634. [40] I. Nakase, M. Niwa, T. Takeuchi, K. Sonomura, N. Kawabata, Y. Koike, M. Takehashi, S. Tanaka, K. Ueda, J.C. Simpson, A.T. Jones, Y. Sugiura, S. Futaki, Cellular uptake of arginine-rich peptides: roles for macropinocytosis and actin rearrangement, Mol. Ther. 10 (2004) 1011–1022. [41] V. Bitsikas, I.R. Correa Jr., B.J. Nichols, Clathrin-independent pathways do not contribute significantly to endocytic flux, eLife 3 (2014), e03970. [42] T. Lehto, A. Castillo Alvarez, S. Gauck, M.J. Gait, T. Coursindel, M.J.A. Wood, B. Lebleu, P. Boisguerin, Cellular trafficking determines the exon skipping activity of Pip6a-PMO in mdx skeletal and cardiac muscle cells, Nucleic Acids Res. 42 (2014) 3207–3217. [43] A.T. Jones, Gateways and tools for drug delivery: endocytic pathways and the cellular dynamics of cell penetrating peptides, Int. J. Pharm. 354 (2008) 34–38. [44] A.H. van Asbeck, A. Beyerle, H. McNeill, P.H.M. Bovee-Geurts, S. Lindberg, W.P.R. Verdurmen, M. Hällbrink, Ü. Langel, O. Heidenreich, R. Brock, Molecular parameters of siRNA-cell penetrating peptide nanocomplexes for efficient cellular delivery, ACS Nano 7 (2013) 3797–3807. [45] L.G. Fong, D. Le, The processing of ligands by the class A scavenger receptor is dependent on signal information located in the cytoplasmic domain, J. Biol. Chem. 274 (1999) 36,808–36,816. [46] L. Pelkmans, J. Kartenbeck, A. Helenius, Caveolar endocytosis of simian virus 40 reveals a new two-step vesicular-transport pathway to the ER, Nat. Cell Biol. 3 (2001) 473–483. [47] J. Gilleron, W. Querbes, A. Zeigerer, A. Borodovsky, G. Marsico, U. Schubert, K. Manygoats, S. Seifert, C. Andree, M. Stoter, H. Epstein-Barash, L. Zhang, V. Koteliansky, K. Fitzgerald, E. Fava, M. Bickle, Y. Kalaidzidis, A. Akinc, M. Maier, M. Zerial, Image-based analysis of lipid nanoparticle-mediated siRNA delivery, intracellular trafficking and endosomal escape, Nat. Biotechnol. 31 (2013) 638–646. [48] O. Meier, K. Boucke, S.V. Hammer, S. Keller, R.P. Stidwill, S. Hemmi, U.F. Greber, Adenovirus triggers macropinocytosis and endosomal leakage together with its clathrin-mediated uptake, J. Cell Biol. 158 (2002) 1119–1131. [49] S. El Andaloussi, T. Lehto, I. Mäger, K. Rosenthal-Aizman, I.I. Oprea, O.E. Simonson, H. Sork, K. Ezzat, D.M. Copolovici, K. Kurrikoff, J.R. Viola, E.M. Zaghloul, R. Sillard, H.J. Johansson, F. Said Hassane, P. Guterstam, J. Suhorutšenko, P.M.D. Moreno, N. Oskolkov, J. Hälldin, U. Tedebark, A. Metspalu, B. Lebleu, J. Lehtiö, C.I.E. Smith, Ü. Langel, Design of a peptide-based vector, PepFect6, for efficient delivery of siRNA in cell culture and systemically in vivo, Nucleic Acids Res. 39 (2011) 3972–3987. [50] X.A. Wu, C.H.J. Choi, C. Zhang, L. Hao, C.A. Mirkin, Intracellular fate of spherical nucleic acid nanoparticle conjugates, JACS 136 (2014) 7726–7733. [51] N. Souleimanian, G.F. Deleavey, H. Soifer, S. Wang, K. Tiemann, M.J. Damha, C.A. Stein, Antisense 2′-deoxy, 2′-fluroarabino nucleic acids (2′F-ANAs) oligonucleotides: in vitro gymnotic silencers of gene expression whose potency is enhanced by fatty acids, Mol. Ther. Nucleic Acids 1 (2012), e43. [52] C.A. Stein, J.B. Hansen, J. Lai, S. Wu, A. Voskresenskiy, A. Hog, J. Worm, M. Hedtjarn, N. Souleimanian, P. Miller, H.S. Soifer, D. Castanotto, L. Benimetskaya, H. Orum, T. Koch, Efficient gene silencing by delivery of locked nucleic acid antisense oligonucleotides, unassisted by transfection reagents, Nucleic Acids Res. 38 (2010), e3. [53] G.J. Della Rocca, Y.V. Mukhin, M.N. Garnovskaya, Y. Daaka, G.J. Clark, L.M. Luttrell, R.J. Lefkowitz, J.R. Raymond, Serotonin 5-HT1A receptor-mediated Erk activation requires calcium/calmodulin-dependent receptor endocytosis, J. Biol. Chem. 274 (1999) 4749–4753. [54] A. Lorents, P.K. Kodavali, N. Oskolkov, Ü. Langel, M. Hällbrink, M. Pooga, Cellpenetrating peptides split into two groups based on modulation of intracellular calcium concentration, J. Biol. Chem. 287 (2012) 16,880–16,889. [55] E. Amiel, A. Alonso, S. Uematsu, S. Akira, M.E. Poynter, B. Berwin, Pivotal advance: toll-like receptor regulation of scavenger receptor-a-mediated phagocytosis, J. Leukoc. Biol. 85 (2009) 595–605. [56] H. Yu, T. Ha, L. Liu, X. Wang, M. Gao, J. Kelley, R. Kao, D. Williams, C. Li, Scavenger receptor A (SR-A) is required for LPS-induced TLR4 mediated NF-κB activation in macrophages, BBA Mol. Cell Res. 1823 (2012) 1192–1198.