Biochimica et Biophysica Acta 1798 (2010) 2258–2265

Contents lists available at ScienceDirect

Biochimica et Biophysica Acta

j ourna l homepage: www.e lsev ie r.com/ locate /bbamem

Cell-penetrating peptide exploited syndecans

Tamás Letoha a,b,⁎, Anikó Keller-Pintér c, Erzsébet Kusz a, Csongor Kolozsi a, Zsolt Bozsó d, Gábor Tóth d,Csaba Vizler a, Zoltán Oláh b,e, László Szilák f,g

a Institute of Biochemistry, Biological Research Center of the Hungarian Academy of Sciences, H-6726 Szeged, Hungaryb Pharmacoidea Ltd., H-6726 Szeged, Hungaryc Department of Biochemistry, University of Szeged, H-6720 Szeged, Hungaryd Department of Medical Chemistry, University of Szeged, H-6720 Szeged, Hungarye Acheuron Hungary Ltd., H-6726 Szeged, Hungaryf Szilak Laboratories, Bioinformatics and Molecule-Design, Szeged H-6723 Szeged, Hungaryg Institute of Biology, Savaria University Center, Western Hungarian University, H-9700 Szombathely, Hungary

⁎ Corresponding author. Institute of Biochemistry, BioHungarian Academy of Sciences, H-6726 Szeged, Temesv62 599784; fax: +36 62 409106.

E-mail address: tletoha@yahoo.com (T. Letoha).

0005-2736/$ – see front matter © 2010 Elsevier B.V. Adoi:10.1016/j.bbamem.2010.01.022

a b s t r a c t

a r t i c l e i n f o

Article history:Received 24 November 2009Received in revised form 18 January 2010Accepted 25 January 2010Available online 2 February 2010

Keywords:Intracellular deliveryCell-penetrating peptidesHeparan sulfate proteoglycansSyndecansProtein kinase C alpha

Cell-penetrating peptides (CPPs) are short peptides capable of translocating across the plasma membrane oflive cells and transporting conjugated compounds intracellularly. Fifteen years after discovering the firstmodel cationic CPPs, penetratin and TAT, CPP internalization is still challenging many questions. Particularlyit has been unknown whether CPPs enter the cells with or without mediation of a specific surface receptor.Here we report that syndecan-4, the universally expressed isoform of the syndecan family of transmembraneproteoglycans, binds and mediates transport of the three most frequently utilized cationic CPPs (penetratin,octaarginine and TAT) into the cells. Quantitative uptake studies and mutational analyses demonstrate thatattachment of the cationic CPPs is mediated by specific interactions between the heparan sulfate chains ofsyndecan-4 and the CPPs. Protein kinase C alpha is also heavily involved in the uptake mechanism. Thecollected data give the first direct evidence on the receptor-mediated uptake of cationic CPPs and mayreplace the long-thought, but already contradicted membrane penetration hypothesis. Thus our study mightgive an answer for a decade long debate and foster the development of rationalized, syndecan-4 targetednovel delivery technologies.

logical Research Center of theári krt. 62, Hungary. Tel.: +36

ll rights reserved.

© 2010 Elsevier B.V. All rights reserved.

1. Introduction

Cell-penetrating peptides (CPPs) are short peptides with a maxi-mum length of 30 amino acids capable of translocating across theplasmamembrane of live cells and transporting conjugated compoundsintracellularly. In recent biomedical research CPP-mediated import ofmembrane impermeable biomolecules (such as polypeptides andoligonucleotides) has become a popular approach to modulate cellularfunctions [1–6]. Throughout the years CPPs have proven their superbvector abilities by delivering a wide range of bioactive agents includingpeptides, proteins, oligonucleotides and drugs intracellularly both invitro and invivo [6–10]. Fifteen years after thefirst publisheddata on theearliest CPPs, the exact mechanism of CPP internalization is stillchallenging lots of questions and it is not yet known whether cellularentry of CPPs happenswith or without themediation of specific cellularreceptors [11]. Despite the hidden details of CPP translocation, several

lines of evidence suggest themajor role ofheparansulfateproteoglycans(HSPGs) in interaction with CPPs on the cell surface [12–16].Particularly, it has been shown that soluble polyanions such as heparinor heparan sulfate (HS) and removal of cell surface polyanionic HSchains of proteoglycans (PGs) using heparitinase abrogates CPPtranslocation [15,17–19]. However, the identity of the certain HSPGand its contribution to cellular internalization of CPPs has not yet beenidentified.

Cell surface HSPGs, substantially more abundant than mostreceptors, modulate encounters of extracellular protein ligands withtheir receptors by forming HS-protein complexes [20–23]. Two typesof HSPGs can be present on the cell surface: the glycosyl-phosphatidyl-inositol anchored glypicans are expressed predominantly in thecentral nervous system and the integral transmembrane proteinsyndecans (SDCs) that are more ubiquitous [24]. The four-memberfamily of SDCs belongs to the type I transmembrane proteins that bearHS chains on their extracellular domains (Fig. 1). They share a similarstructure: a conserved short, one span transmembrane domain (TM)and the following approximately 30 amino acid length cytoplasmicdomain (CD). The N-terminal, divergent extracellular domains(ectodomain) contain three glycosaminoglycan (GAG) attachmentsites for HS near the N-terminus and may bear chondroitin sulfate at

Fig. 1. Schematic representation of the SDC family.

2259T. Letoha et al. / Biochimica et Biophysica Acta 1798 (2010) 2258–2265

juxtamembrane region [25,26]. The syndecan-4 (SDC4) ectodomainsalso compromise cell-binding domains (CBDs) mediating cell–cellattachment [27]. The HS chains of SDCs consist of repeated uronic acidand N-acetylglucosamine disaccharides (40–100) forming long un-branched anionic strings, which contain sections alternately exten-sively or sparsely sulfated. Interaction between SDCs and their ligandsis mediated by specific contacts between sulfated regions of the HSchains and the basic residues on the surface of the ligand [28]. To date,there are four members of the SDC family: SDC1, expressed onepithelial and plasma cells; SDC2, on endothelial cells and fibroblasts;SDC3, predominantly found in the nervous system; and SDC4,universally expressed [25]. SDC4 is the main, Ca2+-independentactivator of the protein kinase C alpha (PKCα) [29–32]. It participatesin focal adhesions, and via its cytoplasmic domain attaches to thecytoskeleton [33]. SDC4 is targeted to lipid rafts, discrete regions of theplasma membrane enriched in cholesterol and sphingolipids. Ligandor antibody mediated clustering leads to redistribution of SDC4 tomembrane rafts and stimulation of a lipid raft-dependent, but clathrin-independent endocytosis of the core protein [34]. Endocytosis of SDC4can be induced by several agents (e.g. FGF2) and blocked by inhibitorsof macropinocytosis like amiloride. However disruption of lipid raftsand inhibition of macropinocytosis have been also shown to abrogateCPPuptake, supporting further evidence on the possible role of SDC4 inCPP translocation [35–37]. Still, apart from a previous report on SDC2enhancing the uptake of arginine-rich peptides (TAT and polyargi-nine), no othermembers of the SDC familywas directly investigated inthe cellular entry of CPPs [13]. On the other hand, SDC2 ectodomainhas been shown to inhibit gene transfer mediated by cationic polymerpolyethyleneimine [38]. Therefore it was high time to address thecontribution of the various SDC isoforms to CPP uptake.

Penetratin, a sixteen amino acid long model cationic CPP derivedfrom the third helix of the Antennapedia homeodomain, is one of thefirst and up-till-now most frequently utilized and studied CPP [39].Early studies with penetratin led to the widely accepted and longlasting cell-penetration hypothesis, rejecting the role of endocytosisand cellular receptors in CPP uptake [40,41]. However more recentresults on the endocytic uptake of CPPs shattered the formerlyproposed models of cellular entry, raising the importance to reviseearlier views, such as the neglected role of cell surface receptors inCPP translocation [11,17,42,43].

To analyze the contribution of SDCs to CPP entry, we applied thethree representative model cationic CPPs: penetratin, octaarginine(R8) and TAT on transfectants expressing the most common SDCisoforms (SDC1, SDC2 and SDC4). Using a methodology that allowsquantitative analysis of cellular internalization, we identified the SDCisoform that facilitates the cellular uptake of CPPs and carries themthrough the plasma membrane into the cytosol. Biochemical and

mutational analysis revealed the influence of various parts of the SDCextracellular domain on cellular entry of the CPPs.

The collected data give the first direct evidence on the receptor-mediated uptake of cationic CPPs and may replace the long-thought,but already contradicted membrane penetration hypothesis. Thus ourstudy might give an answer for a decade long debate and representscationic CPPs as ligands rather than self-penetrating molecules.

2. Materials and methods

2.1. Peptide synthesis and labeling

Penetratin (RQIKIWFQNRRMKWKK), R8 (RRRRRRRR) and TAT(YGRKKRRQRRR) were synthesized on a Rink Amide-AM resin (GLBiochem Ltd) using Nα-Fmoc protected L-amino acids. Peptide chainelongation was carried out using DCC/HOBt (1:1) activation andthreefold molar excess of amino acids in a DCM/DMF (1:1) mixture.The N-terminal Fmoc-group was removed by treating the peptide-resin with 20% piperidine/DMF (5+20 min). Flourescently labeledCPPs were synthesized by extending the N-terminus of the peptideswith a Gly-Gly-Gly linker followed by the coupling of fluoresceinisothiocyanate (FITC; Sigma Aldrich) with the method of Fulop et al.[44]. Peptides were cleaved from the resin by incubating it in amixture of TFA/water/thioanisole/phenol/DTT (87.5:2.5:5:2.5:2.5%)for 3 h prior to ether precipitation. The crude products were dissolvedin an acetic acid/water mixture and lyophilized. Peptides were thenpurified on a preparative C18 HPLC column, analyzed by using ananalytical C18 HPLC column and ESI mass spectrometry. The purefractions were pooled and lyophilized. The CPPs were also labeledwith Alexa Flour® 555 (Alexa555) dye. First a Cys-Gly-Gly-Gly-CPPwas synthesized and purified, then Alexa Flour® 555 C2-maleimide(Invitrogen) was reacted with five time molar excess of the peptide in0.2 M NH4OAc (pH=7.1) for 3 h. The reaction mixture was purifiedon a semipreparative C18 HPLC column.

2.1.1. Vector constructsFull-length SDC1, SDC2, SDC4 and deletion mutants were

amplified and subcloned in the mammalian expression plasmidobtained from Clontech (pcDNA3, pEGFP). The SDC-chimeras wereconstructed by inserting green fluorescent proteins (GFP) into thejuxtamembrane region of the extracellular segment. The signals werekept in all cases to orient the proteins into the membrane.

2.2. Cell culture and transfection

The human SDC DNA constructs were prepared as described aboveand transfected into K562 cells with DMRIE-C (Invitrogen) to prepare

2260 T. Letoha et al. / Biochimica et Biophysica Acta 1798 (2010) 2258–2265

stable cell-lines permanently expressing human SDC1, SDC2, SDC4 orthe truncated forms of SDC1 and SDC4. Transfections were carried outaccording to the recommendations of the manufacturer. After 24 h,cells were incubated with selection medium, containing 0.4 mg/mlG418 (Sigma). The selection medium was changed every second day.After 2 weeks, G418-resistant colonies were analyzed for the proteinexpression with flow cytometry using specific antibodies against thegiven SDC isoforms (anti-human SDC1 [Clone 359103], anti-humanSDC2 [Clone 305515], and anti-human SDC4 [all R&D Systems, Inc.]).Thus colonies exhibiting similar extent of SDC1, SDC2 or SDC4expression were chosen for further studies. The erythroleukemia cellline K562 and its clones were grown as a suspension culture in RPMI1640 medium (Gibco) supplemented with 10% fetal calf serum (FCS;Gibco) at 37 °C in a humified 5% CO2 containing air environment.

2.3. Flow cytometry

Expression of SDC1, SDC2 and SDC4 in K562 transfectants wasmeasured with a FACScan (Becton Dickinson) using specific anti-bodies (anti-human SDC1 [Clone 359103], anti-human SDC2 [Clone305515], and anti-human SDC4) against the studied SDC isoformaccording to the recommendations of the manufacturer (R&DSystems, Inc.). In the case of GFP-tagged SDC1 and SDC4 mutants,expression was also analyzed with flow cytometry measuring thefluorescence intensities of the GFP tags. Thus clones with equalamount of SDC expression were selected and used to quantify of theinternalization of the fluorescent CPPs or fluorescein-conjugatedtransferrin (Invitrogen). Briefly, 6×105 cells/ml in RPMI 1640medium (without FCS) were incubated with the fluorescently (FITCor Alexa555) labeled CPPs (penetratin, R8 and TAT, respectively) ortransferrin at a concentration of 5 μM and 25 μg/ml, respectively, for60 min at 37 °C. In the case of the FITC-labeled peptides andfluorescein-conjugated transferrin, after the 60 min of incubationthe cells werewashed twice in ice-cold PBS and resuspended in 0.5 mlof physiological saline. Equal volumes of this suspension and a stocksolution of trypan blue (Merck KGaA; 500 μg/ml dissolved in ice-cold0.1 M citrate buffer at pH 4.0) were allowed to mix for 1 min beforethe flow cytometric analyses. In this way, sample pH was lowered topH 4.0, thereby optimizing the quenching effect of trypan blue [45]. Inthe case of the SDC1 and SDC4 deletion mutants treated with theAlexa555-labeled CPPs, extracellular fluorescence of surface attachedpeptides was removed by trypsinization according to the methoddescribed by Nakase et al. [13]. Cellular uptake was then measured byflow cytometry using a FACScan (Becton Dickinson). A minimum of10,000 events per sample was analyzed. Viability of cells wasdetermined by using appropriate gating to exclude dead cells, debris,and aggregates in a forward-scatter-against-side-scatter plot.

2.4. Confocal laser scanning microscopy

Internalization of the fluorescently labeled (FITC or Alexa555) CPPsinto K562 cells and SDC transfectants was visualized by confocal laserscanningmicroscopy. Cellswere grownonpoly-D-lysine-coated glass-bottom 35-mm culture dishes (MatTek Corp.). After 24 h, the cellswere preincubated in RPMI 1640 medium (supplemented with 10%FCS) at 37 °C for 30 min before incubation with the fluorescentlylabeled (FITC or Alexa555)CPPs (penetratin, R8 and TAT, respectively).TheCPP solutionwas prepared at a concentration of 5 μMinRPMI 1640medium (without FCS) by a dilution of a 1 mM stock solution ofpeptide in phosphate-buffered saline (PBS). After 60 min at 37 °C, thecellswere rinsed three timeswith ice-cold PBS and nucleiwere stainedwith 2 μg/ml Hoechst 33258 (Sigma) for 10 min. Distribution offluorescence was then analyzed on an Olympus FV1000 confocal laserscanning microscope equipped with three lasers. A laser diode(excitation, 405 nm) and a band-pass filter (420–480 nm) were usedto capture the signal recorded as blue; an argon laser (excitation,

488 nm) and a band-pass filter (505–530 nm) were used to capturethe signal recorded as green; and finally, a helium/neon laser(excitation, 543 nm) and a band-pass filter (550–625 nm) were usedto capture the signal recorded as red. Sections presented were takenapproximately at the mid-height level of the cells. Photomultipliergain and laser power were identical within each experiments. FV10-ASW software was used for image acquisition by confocal microscopy.

2.5. Endocytosis inhibitor studies

To deplete intracellular ATP pools, cells were treated with sodiumazide (NaN3; Sigma) as described previously [46]. To study the effectof proteoglycan sulfation, cells were incubated with 60 mM sodiumchlorate (NaClO3; Sigma) for 48 h prior to CPP treatment as describedby Fadel and Eley [47]. To reveal the involvement of PKCα, some cellswere preincubated with Gö 6976 (1 μM; Sigma) 30 min before FITC-CPP treatment. Effect of macropinocytosis was investigated bypreincubating the cells with 5-(N-ethyl-N-isopropyl) amiloride(Sigma) at a concentration of 500 μM as described previously[34,48]. After incubation with these inhibitors, the cells were treatedwith FITC-labeled CPPs (penetratin, R8 and TAT, respectively) andprocessed for the flow cytometric analyses as described above.

2.6. Statistical analysis

Results are expressed as means±S.E. Differences betweenexperimental groups were evaluated by using analysis of variance(ANOVA). Values of pb0.05 were accepted as significant.

3. Results

3.1. Contribution of the most common SDC isoforms to CPP uptake

Up-till-now no specific cellular receptor that facilitates CPP uptakehas been defined. Therefore we investigated K562 cells transfectedwith the most common isoforms (1, 2 and 4) of the SDC family andtested them for internalization of the three most frequently utilizedCPPs: penetratin, R8 and TAT. First we analyzed SDC expression oftransfectantswithflow cytometry using specific antibodies against thestudied SDC isoforms. Thus transfectants with equal amount of SDCexpression were selected and along with wild-type K562 cells weretreated with the FITC-labeled CPPs (penetratin, R8 and TAT, respec-tively) at a concentration of 5 μM for 60 min at 37 °C. After incubationwith the peptides, cellular uptake into live cells was analyzed withconfocalmicroscopy. As it iswell-established, the erythroleukemia cellline K562 does not express SDCs and glypicans at the cell surface, butexpresses small amounts of betaglycan-transforming growth factorbeta receptor III (TGFßR-III) [49,50]. Accordingly, wild-type K562 cellstreated with the FITC-labeled CPPs gave relatively low level ofintracellular fluorescence, while SDC transfectants showed morepronounced fluorescence intensities, especially those expressingSDC4, demonstrating enhanced CPP entry due to SDC4 expression(Fig. 2A). Note that signals detected from the CPP-treated cells werecoming from vesicle-like intracytoplasmic structures, a characteristicfeature of endocyosis.

Contribution of SDC isoforms to CPP uptake was also quantifiedwith quantitative flow cytofluorometric assays. To enable exactassessment of CPP translocation, extracellular fluorescence of surfacebound fluorescent (FITC) peptides was quenched by adding trypanblue (dissolved at a concentration of 0.25% in ice-cold 0.1 M citratebuffer) 1 min before the analyses [45]. Consistent with previousmicroscopic studies, intracellular fluorescence signal of SDC transfec-tants was higher compared to intensities detected in wild-type K562cells. In these flow cytometric studies SDC4 transfectants presentedthe highest fluorescence signals, indicating that SDC4 enhancedinternalization of the investigated cationic CPPs the most (Fig. 2B–D).

Fig. 2. CPP uptake into K562 cells and SDC transfectants. K562 cells and transfectants of human SDC1, SDC2 and SDC4 were incubated with the FITC-labeled CPPs (penetratin, R8 andTAT, respectively) at a concentration of 5 μM for 60 min at 37 °C. After 60 min of incubation, cellular uptake was analyzed with confocal microscopy or flow cytometry. (A) Confocalmicroscopic visualization of CPP entry into SDC transfectants. Nuclei of cells were stained with Hoechst 33258 dye. Scale bar=10 μM. (B–D) Flow cytometric analyses of CPP uptake.Detected fluorescence intensities normalized to FITC-CPP-treated wild-type K562 cells as standards. The bars represent mean±S.D. of six independent experiments. Statisticalsignificance between CPP-treated cells was assessed by analysis of variance (ANOVA). *pb0.05 vs wild-type K562 cells; **pb0.01 vs wild-type K562 cells.

2261T. Letoha et al. / Biochimica et Biophysica Acta 1798 (2010) 2258–2265

Treating the cells with fluorescein-conjugated transferring, nodifference could be detected in fluorescence of transfectants andwild-type K562 cells, showing that SDC overexpression do not affectclathrin-mediated endocytosis (results not shown).

Further confocal microscopic studies on transfectants expressinggreen fluorescent protein (GFP)-tagged SDC4 showed the intracellularcolocalization of SDC4 and cationic CPPs (Fig. 3), providing additionalevidence on the significant involvement of SDC4 in CPP entry. Thesecolocalization studies demonstrated the common intracellular entry

Fig. 3. Intracellular colocalization of SDC4 and cationic CPPs. Transfectants expressingGFP-tagged SDC4 with the Alexa-labeled CPPs (penetratin, R8 and TAT, respectively) ata concentration of 5 μM for 60 min at 37 °C. After 60 min of incubation, nuclei of cellswere stained with Hoechst 33258 dye and cellular uptake was analyzed with confocalmicroscopy. Scale bar=10 μM.

of SDC4 and the CPPs, suggesting that cationic CPPs translocate to thecells bound to the SDC4 rather than just penetrating alone.

3.2. Contribution of various parts of the SDC4 ectodomain to CPP uptake

After identifying SDC4 as the main facilitator of cellular entry of thecationic CPPs, we examined the influence of various parts of the SDC4extracellular domain on CPP uptake. First we generated three deletionmutants missing different parts of SDC4 extracellular domain (Fig. 4A).Plasmid CBD.pSi4 encoded a SDC4 core protein with a mutantectodomain containing only the cell-binding domain (CBD) and signalsequence (Si), but no HS attachment (HSA) site and HS chains. PlasmidpSi4 encoded a SDC4 deletion mutant with a truncated extracellulardomain made of only the signal sequence (Si), while plasmid HSA.pSi4encoded a SDC4 mutant with and ectodomain comprising the signalsequence (Si) andHSA sitewithHS chains. Since SDC1 and SDC4possessHS chains that have almost identical structures, we also generated aplasmid (HSA.pSi1) encoding a mutant SDC1 with an ectodomainconsisting of the signal sequence (Si) and the HSAwith HS chains. All ofthe above mentioned plasmids were tagged with GFP and transfectedinto K562 cells. Clones with equal extent of SDC expression were thenselected with flow cytometry and treated with the Alexa Flour® 555(Alexa555)-labeled cationic CPPs (penetratin, R8 and TAT, respectively)at a concentration 5 μM for 60 min at 37 °C. Among the CPP-treatedcells, the highest fluorescence was measured on wild-type SDC4transfectants, demonstrating the importance of the full extracellulardomain of SDC4 in attaching and translocating cationic CPPs (Fig. 4B–D).Compared to wild-type SDC4 transfectants, pSi4 mutants (expressing atruncated ectodomain made of only a short signal sequence) displayedthe lowest level of cellular fluorescence, reflecting the low level of CPPcellular entry. CPP uptake into CBD.pSi4mutants (with a full CBD and Sicontaining ectodomain) was also quite low relative to transfectants of

Fig. 4. Contribution of various parts of the SDC4 ectodomain to CPP uptake. (A) Schematic representation of SDC1 and SDC4 deletion mutants used in the study. (B–D) Results of flowcytometric measurements. Detected fluorescence intensities are normalized to Alexa555-CPP-treated wild-type SDC4 transfectants as standards. The bars represent mean±S.D. ofsix independent experiments. Statistical significance between CPP-treated cells was assessed by analysis of variance (ANOVA). *pb0.05 vs wild-type SDC4 transfectants; **pb0.01 vswild-type SDC4 transfectants.

2262 T. Letoha et al. / Biochimica et Biophysica Acta 1798 (2010) 2258–2265

wild-type SDC4, indicating that theCBDof SDC4donot have amajor rolein interaction with the cationic CPPs. HSA.pSi4 mutants (with SDC4ectodomain containing Si and HSA site with HS chains) exhibited morepronounced cellular fluorescence that surmounted intensities detectedon SDC1 mutants with HS-only extracellular domain (HSA.pSi1),indicating that apart from the almost identical HS chains, othersubdomains of the SDC4 core protein influence the internalizationprocess.

3.3. Effect of endocytosis inhibitors on CPP uptake

Our previous assays on SDC transfectants showed that SDCs,especially SDC4 greatly enhanced uptake of the cationic CPPs.Moreover, colocalization studies showed that SDC4 and the CPPsenter the cells together. On the other hand, CPP internalization couldbe still detected on K562 cells devoid of SDCs, suggesting that SDC4-dependent uptake might be a supreme, but not the only mechanismresponsible for CPP translocation. In order to reveal the nature ofuptake mechanisms excluding SDC4 and other SDCs, we treated K562cells and transfectants with various inhibitors of endocytosis andtested their effects on CPP uptake.

Endocytic uptake of CPPs is still under a heavy dispute. Since earlystudies claimed the ATP-independent, non-endocytic uptake of CPPs[10], thus we studied cellular uptake into ATP-depleted cells. In ourassays, ATP depletion with sodium azide (NaN3) greatly diminishedCPP uptake in all of the involved cell-lines, indicating the insignifi-cance of ATP-independent, non-endocytic penetration of the cationicCPPs (Fig. 5A–C).

It has been also widely suggested that HSPGs serve as an initialreceptor for the adherence and subsequent internalization of cationic

CPPs. Interactions of PGs with cationic ligands are mediated bysulfated regions of the HS chains. This raises the possibility thatvariation in sulphation might have considerable biological ramifica-tions on CPP internalization. To examine the effect of sulfateconstituents on uptake of the studies cationic CPPs, we incubatedK562 cells and transfectants with sodium chlorate (NaClO3), an agentknown to promote undersulfation of PGs [51]. According to flowcytometric studies, chlorate treatment resulted in marked reductionof CPP uptake in K562 cells and SDC transfectants (Fig. 5A–C). As flowcytometric studies without using the quenching effect of trypan bluepointed out, sodium chlorate led to the decrease of overall (both intra-and extra-) cellular fluorescence, meaning that undersulfationabrogates even the attachment of cationic CPPs to the membranesurface (results not shown).

SDC4 can bind and activate PKCα via a tetrameric activationcomplex formed by a SDC4 dimer and two molecules of phosphati-dylinositol 4,5-bisphosphate (PIP2) [29–31,52,53]. Considering thisinterplay between SDC4 and PKCα, we tested the effect of theselective PKCα inhibitor Gö 6976 on CPP translocation. Gö 6976greatly decreased cellular internalization of all investigated CPPs. Theextent of decrease however was quite different between the variouscell-lines as SDC1 transfectants showed the less diminished CPPuptake and SDC4 transfectants the most (Fig. 5A–C).

Macropinocytosis has already been suggested as the major uptakemechanism of arginine-rich CPPs [54]. Thus we examined the effectsof amiloride, an agent reported as a potent inhibitor of macropino-cytosis [34,48], on the internalization of the cationic CPPs. In ourassays, amiloride has significantly decreased cellular uptake of allCPPs in all cell-lines (Fig. 5A–C), suggesting the macropinocytoticuptake of cationic CPPs.

Fig. 5. Effects of various endocytosis inhibitors on CPP uptake. K562 cells andtransfectants of SDC1, SDC2 and SDC4 were incubated with the FITC-labeled CPPs(penetratin, R8 and TAT, respectively) at a concentration of 5 μM for 60 min at 37 °C anduptake was measured with flow cytometry. Some cells were incubated with sodiumazide (NaN3), others with sodium chlorate (NaClO3) or Gö 6796 or amiloride prior toCPP treatment at 37 °C. Detected fluorescence intensities were normalized to CPP-treated wild-type K562 cells as standards. The bars represent mean±S.D. of sixindependent experiments.

2263T. Letoha et al. / Biochimica et Biophysica Acta 1798 (2010) 2258–2265

4. Discussion

Cell-penetration by certain biological entities (i.e. viruses) arefacilitated by relatively short cationic subdomains, so called CPPs witha length of about 30 amino acids [55–57]. Man made CPPs, even incontext with molecular conjugates, can be exploited to facilitatetranslocation of otherwise membrane impermeable biomolecules[1,2,5,58,59]. More than fifteen years after the discovery of the firstconceptualized CPPs, penetratin and TAT, cellular entry of CPPs stilladdresses many questions [11]. One of the most important questionswaiting to be answered was the role of cell surface receptors in CPPuptake [60–62]. Until now no cellular receptor that enhances theuptake of CPPs has been identified. Although the role of polyanionicmembrane proteoglycans in CPP uptake has already been implied, thespecific proteoglycan responsible for the unique internalization ofCPPs has not been revealed. The HS chains of SDCs, a family of

transmembrane proteoglycans are depicted as the main attachmentsite for many viruses, parasites and cationic gene delivery systems,but so far this is the first study providing direct evidence on SDC-mediated CPP uptake.

Our assays on SDC transfectants demonstrated that SDCs didfacilitate uptake of the most frequently utilized representatives ofcationic CPPs (penetratin, R8 and TAT), while uptake of transferrin, awell-established marker of clathrin-mediated endocytosis, was notaffected. The extent of CPP uptake however was quite differentbetween transfectants of the various SDC isoforms (1, 2 and 4).Particularly SDC4 enhanced CPP cellular entry: almost tripling cellularinternalization of all of the studied CPPs. Colocalization studiesrevealed that SDC4 and the CPPs translocate into the cytosol together,suggesting that cationic CPPs are actually transported into the cell asligands bound to SDC4. Interaction of the CPPs with SDC4 dependshighly on HS chains and less on other parts of the SDC4 ectodomainincluding the CBD that mediates cell–cell attachment. The observeddifference of cationic CPP uptake between SDC1 and SDC4 transfec-tants is indeed very interesting. Zako et al. has already reported thatHS chains of SDC1 and SDC4 are structurally almost indistinguishable[63]. Thus the difference in CPP uptake between SDC1 and SDC4mutants with truncated ectodomains made of only HSA sites and HSchains suggest that apart from the HS chains, the intracellular domainof SDC4 also contribute to the superior ability of SDC4 to mediate CPPinternalization. The interplay between the intracellular domain ofSDC4 and PKCα and is already well documented [64–66]. Therefore,diminished CPP uptake due to selective inhibition of PKCα furtherjustified the role of SDC4 in CPP internalization. And although to avarious extent, but Gö 6976 exerted general inhibition on CPP uptakeinto all of the investigated cell types, not just SDC4 transfectants,indicating the universal importance of PKCα in the internalizationmechanism. Thus PKCα activation, independent of SDC4, is necessaryfor CPP uptake. Since uptake of CPPs at a lower level could be stilldetected onwild-type K562 cells (expressing small amounts of TGFßRIII), it also became evident that internalization of the CPP was notconfined to SDC4 or the other SDCs. As studies on sodium chlorate-treated cells pointed out, uptake mechanisms excluding SDCs stillheavily rely on other PGs. According to our studies with ATP-depletedor cells, cationic CPP uptake, either mediated by SDCs or other PGs, isan ATP-dependent, endocytotic process. Macropinocytosis also playspart in the internalization of cationic CPPs.

In summary our data provides evidence that SDC4 binds andmediates transport of model cationic CPPs (penetratin, R8 and TAT)through the plasma membrane into the cells. Our study demonstratesthat cationic CPPs, unlike stated in previous studies, act as ligands,preferentially of SDC4, rather than cell-penetrating molecules duringthe internalization process. It also became evident, that apart fromSDC4, to a smaller extent other SDCs and PGs can facilitate CPPmembrane translocation that heavily relies on PKCα activity. Forsmall Trojan CPPs like penetratin, R8 and TAT, exploiting variousendocytic routes simultaneously is the key to ensure certain entry intothe cells. In this sense, SDC4 is a supreme, but not the sole mediator ofCPP uptake. The fact that cationic CPPs can utilize multiple cellularreceptors (whether it is SDC4 or other PGs) and internalizationpathways is actually favorable. And although the identification ofSDC4 as a main plasma membrane carrier of penetratin, R8 and TATpresents a great advance in the understanding of the uniquetranslocational properties of cationic CPPs, still there is a long roadahead till we could fully reveal every single route of CPP entry into thecells.

Acknowledgements

TL was supported by grants of the National Office for Research andTechnology (OMFB-01630; OMFB-01703, OMFB-01576/2006 andBAROSS_DA07-DA_TECH_07-2008-0043) and the Postdoctoral

2264 T. Letoha et al. / Biochimica et Biophysica Acta 1798 (2010) 2258–2265

Fellowship of the Zoltan Magyary Foundation. ZO was supported byMC-IRG030854-PAINKILLER; Anyos Jedlik Program NKFP-1-00019/2005; GVOP-3.3.1-05/1.-2005-05-0057/3.0, BAROSS_DA07-DA_TECH_07-2008-0043 and BAROSS_DA07-DA_TECH_07-2008-0028. CV was also supported by grants of the National Office forResearch and Technology (OM-00051/2005 and OMFB-01575/2006)and the Hungarian Ministry of Health (552/2006). LS was supportedby Anyos Jedlik Program of the National Office for Research andTechnology NKFP1a-0024-05.

References

[1] S. Deshayes, M.C. Morris, G. Divita, F. Heitz, Cell-penetrating peptides: tools forintracellular delivery of therapeutics, Cell. Mol. Life Sci. 62 (2005) 1839–1849.

[2] C. Foged, H.M. Nielsen, Cell-penetrating peptides for drug delivery acrossmembrane barriers, Expert Opin. Drug Deliv. 5 (2008) 105–117.

[3] P. Jarver, U. Langel, Cell-penetrating peptides—a brief introduction, Biochim.Biophys. Acta 1758 (2006) 260–263.

[4] M. Lindgren, M. Hallbrink, A. Prochiantz, U. Langel, Cell-penetrating peptides,Trends Pharmacol. Sci. 21 (2000) 99–103.

[5] M.C. Morris, S. Deshayes, F. Heitz, G. Divita, Cell-penetrating peptides: frommolecular mechanisms to therapeutics, Biol. Cell 100 (2008) 201–217.

[6] F. Heitz, M.C. Morris, G. Divita, Twenty years of cell-penetrating peptides:from molecular mechanisms to therapeutics, Br. J. Pharmacol. 157 (2009)195–206.

[7] S. Deshayes, M. Morris, F. Heitz, G. Divita, Delivery of proteins and nucleic acidsusing a non-covalent peptide-based strategy, Adv. Drug Deliv. Rev. 60 (2008)537–547.

[8] B.R. Meade, S.F. Dowdy, Enhancing the cellular uptake of siRNA duplexesfollowing noncovalent packaging with protein transduction domain peptides,Adv. Drug Deliv. Rev. 60 (2008) 530–536.

[9] K.M. Stewart, K.L. Horton, S.O. Kelley, Cell-penetrating peptides as deliveryvehicles for biology and medicine, Org. Biomol. Chem. 6 (2008) 2242–2255.

[10] E. Vives, J. Schmidt, A. Pelegrin, Cell-penetrating and cell-targeting peptides indrug delivery, Biochim. Biophys. Acta (2008).

[11] F. Edenhofer, Protein transduction revisited: novel insights into the mechanismunderlying intracellular delivery of proteins, Curr. Pharm. Des. 14 (2008)3628–3636.

[12] E. Ghibaudi, B. Boscolo, G. Inserra, E. Laurenti, S. Traversa, L. Barbero, R.P. Ferrari,The interaction of the cell-penetrating peptide penetratin with heparin,heparansulfates and phospholipid vesicles investigated by ESR spectroscopy, J.Pept. Sci. 11 (2005) 401–409.

[13] I. Nakase, A. Tadokoro, N. Kawabata, T. Takeuchi, H. Katoh, K. Hiramoto, M.Negishi, M. Nomizu, Y. Sugiura, S. Futaki, Interaction of arginine-rich peptideswith membrane-associated proteoglycans is crucial for induction of actinorganization and macropinocytosis, Biochemistry 46 (2007) 492–501.

[14] I. Nakase, T. Takeuchi, G. Tanaka, S. Futaki, Methodological and cellular aspectsthat govern the internalization mechanisms of arginine-rich cell-penetratingpeptides, Adv. Drug Deliv. Rev. 60 (2008) 598–607.

[15] G.M. Poon, J. Gariepy, Cell-surface proteoglycans as molecular portals forcationic peptide and polymer entry into cells, Biochem. Soc. Trans. 35 (2007)788–793.

[16] A. Ziegler, J. Seelig, Binding and clustering of glycosaminoglycans: a commonproperty of mono- and multivalent cell-penetrating compounds, Biophys. J. 94(2008) 2142–2149.

[17] S. Console, C. Marty, C. Garcia-Echeverria, R. Schwendener, K. Ballmer-Hofer,Antennapedia and HIV transactivator of transcription (TAT), “protein trans-duction domains” promote endocytosis of high molecular weight cargo uponbinding to cell surface glycosaminoglycans, J.Biol. Chem. 278 (2003)35109–35114.

[18] F. Duchardt, M. Fotin-Mleczek, H. Schwarz, R. Fischer, R. Brock, A comprehensivemodel for the cellular uptake of cationic cell-penetrating peptides, Traffic8 (2007) 848–866.

[19] J.P. Richard, K. Melikov, H. Brooks, P. Prevot, B. Lebleu, L.V. Chernomordik, Cellularuptake of unconjugated TAT peptide involves clathrin-dependent endocytosis andheparan sulfate receptors, J. Biol. Chem. 280 (2005) 15300–15306.

[20] J.R. Bishop, M. Schuksz, J.D. Esko, Heparan sulphate proteoglycans fine-tunemammalian physiology, Nature 446 (2007) 1030–1037.

[21] R.V. Iozzo, Heparan sulfate proteoglycans: intricate molecules with intriguingfunctions, J. Clin. Invest. 108 (2001) 165–167.

[22] P.W. Park, O. Reizes, M. Bernfield, Cell surface heparan sulfate proteoglycans:selective regulators of ligand-receptor encounters, J. Biol. Chem. 275 (2000)29923–29926.

[23] S. Tumova, A. Woods, J.R. Couchman, Heparan sulfate proteoglycans on the cellsurface: versatile coordinators of cellular functions, Int. J. Biochem. Cell Biol. 32(2000) 269–288.

[24] M. Bernfield, M. Gotte, P.W. Park, O. Reizes, M.L. Fitzgerald, J. Lincecum, M. Zako,Functions of cell surface heparan sulfate proteoglycans, Annu. Rev. Biochem. 68(1999) 729–777.

[25] M. Bernfield, R. Kokenyesi, M. Kato, M.T. Hinkes, J. Spring, R.L. Gallo, E.J. Lose,Biology of the syndecans: a family of transmembrane heparan sulfate proteogly-cans, Annu. Rev. Cell Biol. 8 (1992) 365–393.

[26] R. Kokenyesi, M. Bernfield, Core protein structure and sequence determine the siteand presence of heparan sulfate and chondroitin sulfate on syndecan-1, J. Biol.Chem. 269 (1994) 12304–12309.

[27] A.J. McFall, A.C. Rapraeger, Identification of an adhesion site within the syndecan-4 extracellular protein domain, J. Biol. Chem. 272 (1997) 12901–12904.

[28] A.C. Rapraeger, V.L. Ott, Molecular interactions of the syndecan core proteins,Curr. Opin. Cell Biol. 10 (1998) 620–628.

[29] E. Keum, Y. Kim, J. Kim, S. Kwon, Y. Lim, I. Han, E.S. Oh, Syndecan-4 regulateslocalization, activity and stability of protein kinase C-alpha, Biochem. J. 378(2004) 1007–1014.

[30] E.S. Oh, A. Woods, J.R. Couchman, Syndecan-4 proteoglycan regulates thedistribution and activity of protein kinase C, J. Biol. Chem. 272 (1997) 8133–8136.

[31] E.S. Oh, A. Woods, J.R. Couchman, Multimerization of the cytoplasmic domain ofsyndecan-4 is required for its ability to activate protein kinase C, J. Biol. Chem. 272(1997) 11805–11811.

[32] E. Tkachenko, J.M. Rhodes, M. Simons, Syndecans: new kids on the signaling block,Circ. Res. 96 (2005) 488–500.

[33] A. Yoneda, J.R. Couchman, Regulation of cytoskeletal organization by syndecantransmembrane proteoglycans, Matrix Biol. 22 (2003) 25–33.

[34] E. Tkachenko, E. Lutgens, R.V. Stan, M. Simons, Fibroblast growth factor 2endocytosis in endothelial cells proceed via syndecan-4-dependent activation ofRac1 and a Cdc42-dependent macropinocytic pathway, J. Cell Sci. 117 (2004)3189–3199.

[35] S.W. Jones, R. Christison, K. Bundell, C.J. Voyce, S.M. Brockbank, P. Newham, M.A.Lindsay, Characterisation of cell-penetrating peptide-mediated peptide delivery,Br. J. Pharmacol. 145 (2005) 1093–1102.

[36] I.M. Kaplan, J.S. Wadia, S.F. Dowdy, Cationic TAT peptide transduction domainenters cells by macropinocytosis, J. Control. Release 102 (2005) 247–253.

[37] J.S. Wadia, R.V. Stan, S.F. Dowdy, Transducible TAT-HA fusogenic peptideenhances escape of TAT-fusion proteins after lipid raft macropinocytosis, Nat.Med. 10 (2004) 310–315.

[38] S. Paris, A. Burlacu, Y. Durocher, Opposing roles of syndecan-1 and syndecan-2 inpolyethyleneimine-mediated gene delivery, J. Biol. Chem. 283 (2008) 7697–7704.

[39] D. Derossi, A.H. Joliot, G. Chassaing, A. Prochiantz, The third helix of theAntennapedia homeodomain translocates through biological membranes, J. Biol.Chem. 269 (1994) 10444–10450.

[40] D. Derossi, S. Calvet, A. Trembleau, A. Brunissen, G. Chassaing, A. Prochiantz, Cellinternalization of the third helix of the Antennapedia homeodomain is receptor-independent, J. Biol. Chem. 271 (1996) 18188–18193.

[41] D. Derossi, G. Chassaing, A. Prochiantz, Trojan peptides: the penetratin system forintracellular delivery, Trends Cell Biol. 8 (1998) 84–87.

[42] A.T. Jones, Gateways and tools for drug delivery: endocytic pathways and thecellular dynamics of cell penetrating peptides, Int. J. Pharm. 354 (2008) 34–38.

[43] J.P. Richard, K. Melikov, E. Vives, C. Ramos, B. Verbeure, M.J. Gait, L.V.Chernomordik, B. Lebleu, Cell-penetrating peptides. A reevaluation of themechanism of cellular uptake, J. Biol. Chem. 278 (2003) 585–590.

[44] L. Fulop, B. Penke, M. Zarandi, Synthesis and fluorescent labeling of beta-amyloidpeptides, J. Pept. Sci. 7 (2001) 397–401.

[45] S. Busetto, E. Trevisan, P. Patriarca, R. Menegazzi, A single-step, sensitive flowcytofluorometric assay for the simultaneous assessment of membrane-bound andingested Candida albicans in phagocytosing neutrophils, Cytometry A 58 (2004)201–206.

[46] T. Letoha, S. Gaal, C. Somlai, Z. Venkei, H. Glavinas, E. Kusz, E. Duda, A. Czajlik, F.Petak, B. Penke, Investigation of penetratin peptides. Part 2. In vitro uptake ofpenetratin and two of its derivatives, J. Pept. Sci. 11 (2005) 805–811.

[47] S. Fadel, A. Eley, Chlorate: a reversible inhibitor of proteoglycan sulphation inChlamydia trachomatis-infected cells, J. Med. Microbiol. 53 (2004) 93–95.

[48] S. Muro, R. Wiewrodt, A. Thomas, L. Koniaris, S.M. Albelda, V.R. Muzykantov, M.Koval, A novel endocytic pathway induced by clustering endothelial ICAM-1 orPECAM-1, J. Cell Sci. 116 (2003) 1599–1609.

[49] L. Attisano, J.L. Wrana, F. Lopez-Casillas, J. Massague, TGF-beta receptors andactions, Biochim. Biophys. Acta 1222 (1994) 71–80.

[50] S. Shafti-Keramat, A. Handisurya, E. Kriehuber, G. Meneguzzi, K. Slupetzky, R.Kirnbauer, Different heparan sulfate proteoglycans serve as cellular receptors forhuman papillomaviruses, J. Virol. 77 (2003) 13125–13135.

[51] F. Safaiyan, S.O. Kolset, K. Prydz, E. Gottfridsson, U. Lindahl, M. Salmivirta,Selective effects of sodium chlorate treatment on the sulfation of heparan sulfate,J. Biol. Chem. 274 (1999) 36267–36273.

[52] J.R. Couchman, S. Vogt, S.T. Lim, Y. Lim, E.S. Oh, G.D. Prestwich, A. Theibert, W. Lee,A. Woods, Regulation of inositol phospholipid binding and signaling throughsyndecan-4, J. Biol. Chem. 277 (2002) 49296–49303.

[53] B.K. Koo, Y.S. Jung, J. Shin, I. Han, E. Mortier, P. Zimmermann, J.R. Whiteford, J.R.Couchman, E.S. Oh, W. Lee, Structural basis of syndecan-4 phosphorylation as amolecular switch to regulate signaling, J. Mol. Biol. 355 (2006) 651–663.

[54] A.T. Jones, Macropinocytosis: searching for an endocytic identity and role in theuptake of cell penetrating peptides, J. Cell. Mol. Med. 11 (2007) 670–684.

[55] A. Joliot, Transduction peptides within naturally occurring proteins, Sci. STKE2005 (2005) e54.

[56] A. Prochiantz, Homeodomain-derived peptides. In and out of the cells, Ann. N. Y.Acad. Sci. 886 (1999) 172–179.

[57] A. Prochiantz, Messenger proteins: homeoproteins, TAT and others, Curr. Opin.Cell Biol. 12 (2000) 400–406.

[58] A. Joliot, A. Prochiantz, Transduction peptides: from technology to physiology,Nat. Cell Biol. 6 (2004) 189–196.

[59] J.S. Wadia, S.F. Dowdy, Protein transduction technology, Curr. Opin. Biotechnol. 13(2002) 52–56.

2265T. Letoha et al. / Biochimica et Biophysica Acta 1798 (2010) 2258–2265

[60] S. Futaki, Arginine-rich peptides: potential for intracellular delivery of macro-molecules and the mystery of the translocation mechanisms, Int. J. Pharm. 245(2002) 1–7.

[61] S. Futaki, I. Nakase, A. Tadokoro, T. Takeuchi, A.T. Jones, Arginine-rich peptides andtheir internalization mechanisms, Biochem. Soc. Trans. 35 (2007) 784–787.

[62] M. Kosuge, T. Takeuchi, I. Nakase, A.T. Jones, S. Futaki, Cellular internalization anddistribution of arginine-rich peptides as a function of extracellular peptideconcentration, serum, and plasma membrane associated proteoglycans, Bioconjug.Chem. 19 (2008) 656–664.

[63] M. Zako, J. Dong, O. Goldberger, M. Bernfield, J.T. Gallagher, J.A. Deakin, Syndecan-1 and -4 synthesized simultaneously by mouse mammary gland epithelial cells

bear heparan sulfate chains that are apparently structurally indistinguishable, J.Biol. Chem. 278 (2003) 13561–13569.

[64] S. Choi, E. Lee, S. Kwon, H. Park, J.Y. Yi, S. Kim, I.O. Han, Y. Yun, E.S. Oh,Transmembrane domain-induced oligomerization is crucial for the functions ofsyndecan-2 and syndecan-4, J. Biol. Chem. 280 (2005) 42573–42579.

[65] M. Simons, A. Horowitz, Syndecan-4-mediated signalling, Cell. Signal. 13 (2001)855–862.

[66] H. Son, Y. Lim, J. Kim, H. Park, S. Choi, I. Han, W.S. Kim, S. Park, Y. Bae, E.S. Oh,Protein kinase Calpha can undergo membrane localization via an alternativephosphatidylinositol 4, 5-bisphosphate-dependent pathway, Arch. Biochem.Biophys. 454 (2006) 1–6.